Evanston Northwestern Healthcare Corporation

Text (OCR of the scan at left; may contain errors)

IN THE MATTER OF EVANSTON NORTHWESTERN HEALTHCARE CORPORATION AND ENH MEDICAL GROUP, INC.

OPINION OF THE COMMISSION AND ORDER IN REGARD TO ALLEGED VIOLATIONS OF SEC. 7 OF THE CLAYTON ACT Docket 9315; File No. 011 0234 Opinion, August 2, 2007 – Order, August 2, 2007 Complaint, February 10, 2004 – Initial Decision, October 20, 2005 The complaint concerns the merger of Evanston Northwestern Healthcare Corporation (“ENH”) and Highland Park Hospital in January 2000. The merger also folded the Highland Park Independent Physician Association into ENH Medical Group. The complaint alleged that the merger violated Section 7 of the Clayton Act by significantly reducing competition in the market for general acute acute care inpatient hospital services sold to private payers in northeast Cook County and southeast Lake County, Illinois. The complaint also alleged that ENH ENH Medical Group violated Section 5 of the FTC Act requiring private payers to to accept increased prices for hospital and physician services or face termination of termination of both. Following an administrative hearing, the Administrative Law Law Judge issued an Initial Decision, 144 F.T.C. 19, ruling that the effect of the the acquisition of Highland Park by ENH substantially lessened competition in violation of Section 7 of the Clayton Act and ordered the divestiture of the Highland Park Hospital Assets. On appeal, the Commission upheld the Initial Decision, finding that the merger enabled ENH to exercise market power, and that that ENH used this market power to increase its average net prices for acute inpatient hospital services by a substantial amount, but rejected divestiture as a remedy. The Commission issued an Order to cease and desist. The Order also requires the Respondents to file with the Commission a detailed proposal for implementing the injunctive relief ordered. VOLUME 144 Complaint Participants For the Commission: Thomas H. Brock, Jeff Dahnke, Philip M. Eisenstat, Jamel R. Franklin, Renee S. Henning, Albert Y. Kim, Jeanne H. Liu, John Martin, Christopher L. Marvine, Paul J. Nolan, Chul Pak, Nancy Park, Stephanie A. Parks, Anthony R. Saunders, Stephen W. Sockwell, Jr., Kristina E. Van Horn, Steve Vieux, Oscar Voss, and Goldie V. Walker.

For the Respondents: Tracy J. Allen, Keitha Clopper, David E. Dahlquist, Michael A. Del Negro, Christopher B. Essig, Duane M. Kelley, J. Ethan McComb, and Scott C. Walton, Winston & Strawn LLP, Chicago, Illinois; Charles B. Klein, Jay L. Levine, Rebecca C. Morrison, Adam S. Nadelhaft, Jennifer H. Scott, Michael L. Sibarium, and Patrick A. Tillou, Winston & Strawn LLP, Washington, D.C.

COMPLAINT Pursuant to the provisions of the Federal Trade Commission Act, and by virtue of the authority vested in it by said Act, the Federal Trade Commission, having reason to believe that respondent Evanston Northwestern Healthcare Corporation (“ENH”) has violated and is violating Section 7 of the Clayton Act, and that respondent ENH Medical Group, Inc. (“ENH Medical Group”), has violated and is violating Section 5 of the Federal Trade Commission Act, and it appearing to the Commission that a proceeding by it in respect thereof would be in the public interest, hereby issues its complaint stating its charges in that respect as follows: NATURE OF THE CASE 1. This complaint concerns the merger of ENH and Highland Park Hospital (“Highland Park”) in January 2000. The merger combined ENH’s Evanston and Glenbrook hospitals located in Cook EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 3 Complaint Cook County, Illinois with Highland Park Hospital, the nearest hospital to the north. Shortly after the merger, ENH negotiated uniform prices for the three hospitals as a single system and raised prices at all three locations, the largest of which was at ENH. The price increases that resulted from the merger are large and far beyond those achieved by comparable hospitals during this time period.

2. The merger also folded the Highland Park Independent Physician Association (“IPA”) into ENH Medical Group, creating a larger group that included both ENH salaried physicians as well as other independent physicians. Following the merger, ENH Medical Group engaged in price fixing of physician services by negotiating with third party payers for uniform prices for both the salaried physicians and non-salaried, independent physicians. This conduct deprived commercial payers, employers, and individuals the benefits of competition in physician services.

3. After merging the hospitals and the physician groups, ENH conducted negotiations with private payers by offering hospital services and physician services as a package. In many instances, ENH required private payers to accept its terms for both hospital and physician services or face termination of both hospital and physician contracts.

BACKGROUND ON THE ENH HOSPITALS AND MEDICAL GROUP 4. ENH is a non-profit corporation organized, existing, and doing business under, and by virtue of, the laws of Illinois, with its office and principal place of business located at 1301 Central Street, Evanston, Illinois 60201. For the fiscal year ending September 30, 2000, ENH had revenues of about $735 million. 5. ENH owns and operates Evanston Hospital (“Evanston”), a a 466-bed acute care hospital located in Evanston, Illinois, Glenbrook Hospital (“Glenbrook”), a 136-bed acute care hospital VOLUME 144 Complaint located near Evanston, and Highland Park, a 234-bed acute care hospital also located near Evanston.

6. Prior to the merger Highland Park was offering a broad range of medical and surgical services. In addition, Highland Park was pursuing the offering of open heart surgery through regulatory filings with the state of Illinois and through formation of a joint venture with Evanston.

7. ENH is the sole member or owner of ENH Faculty Practice Associates (“Faculty Practice Associates”), an Illinois non-profit corporation located at 1301 Central Street, Evanston, Illinois 60201. Faculty Practice Associates was organized in 1990 under its former name Evanston Medical Specialists Foundation. It currently employs about 500 physicians who primarily serve the patients of ENH. 8. ENH Medical Group is a for-profit corporation organized, existing, and doing business under, and by virtue of, the laws of Illinois, with its office and principal place of business located at 1301 Central Street, Evanston, Illinois 60201. Faculty Practice Associates, which ENH controls, is the sole shareholder of ENH Medical Group.

JURISDICTION 9. ENH is, and at all relevant times has been, engaged in commerce within the meaning of the Clayton Act. Before their merger with ENH, Highland Park, a non-profit Illinois corporation, and its parent Lakeland Health Services, Inc., a non-profit Illinois corporation, were engaged in commerce within the meaning of the Clayton Act. ENH’s merger with Highland Park constitutes an acquisition under the Clayton Act.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 5 Complaint 10. ENH Medical Group is, and at all relevant times has been, engaged in commerce within the meaning of the Federal Trade Commission Act.

11. ENH Medical Group is a corporation within the meaning of Section 4 of the Federal Trade Commission Act. THE MERGER 12. On or about January 1, 2000, ENH and Lakeland Health Services, Inc., completed a merger by which Lakeland Health Services, Inc., and its subsidiary, Highland Park, merged with and into ENH. There was no merger or acquisition price in connection with this transaction. In August 1999, ENH estimated the fair market value of Highland Park at $233,528,000.

13. The merger placed Evanston, Glenbrook, and Highland Park under the control of ENH. The merger established one board of directors, one management staff, and one medical staff. Since the merger, ENH has collectively negotiated prices for all three hospitals.

14. Prior to the merger, ENH and Highland Park, along with several other hospitals, were members of a joint venture known as the Northwestern Healthcare Network. Under that joint venture, ENH and Highland Park and the other members maintained separate management and negotiated prices independently. At the time of the merger negotiations, members of the Northwestern Healthcare Network planned to exit from or dissolve the joint venture. They dissolved the joint venture on January 3, 2000, two days after ENH and Highland Park consummated the merger. COUNT I: MERGER OF HOSPITALS IN VIOLATION OF CLAYTON ACT §7 15. The allegations of paragraphs 1 through 14 are incorporated by reference as though fully set forth herein. VOLUME 144 Complaint PRODUCT MARKET 16. The relevant product market is general acute care inpatient hospital services sold to private payers, including commercial payers, managed care plans, and self-insurance plans (collectively, “private payers”). General acute care inpatient hospital services are a broad cluster of basic medical and surgical diagnostic and treatment services that include an overnight stay in the hospital by the patient. General acute care inpatient hospital services exclude (i) services at hospitals that serve solely military and veterans; (ii) services at outpatient facilities that provide same-day service only; (iii) sophisticated services known in the industry as “tertiary services” that include such services as open heart surgery and transplants; and (iv) psychiatric, substance abuse, and rehabilitation services. GEOGRAPHIC MARKET 17. The relevant geographic market in which to analyze the merger is the geographic area directly proximate to the three ENH hospitals and contiguous geographic areas in northeast Cook County and southeast Lake County, Illinois. This geographic area, in which a significant number of individuals who seek hospital care at the three ENH hospitals reside, spans (and may be narrower than) the densely populated suburban corridor that runs for about 15 miles north-south along the shore of Lake Michigan, and extends roughly ten miles west of the Lake. The existence of this relevant geographic market is evidenced, among other things, by the ability of ENH, once it controlled Highland Park as well as the Evanston and Glenbrook hospitals, profitably to impose significant and nontransitory price increases upon private payers in their purchase of acute care hospital services at those hospitals. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 7 Complaint CONCENTRATION 18. As a result of the merger, ENH has been able to exercise market power in the relevant market. The merger of ENH and Highland Park created the largest hospital system in the relevant market. This market is highly concentrated and the combination significantly increased market concentration. The merger resulted in a post-merger HHI increase in excess of 500 points to a level exceeding 3000 points.

ENTRY CONDITIONS 19. It is unlikely that entry into the market would remedy, in a timely manner, the anticompetitive effects from the merger. Entry is difficult and likely to take more than two years because of the time required to plan for and to complete construction of an acute care hospital.

20. Government regulations also make entry difficult. The Illinois Health Facilities Planning Act, 20 Illinois Code § 3960, restricts entry in this market. The Act prevents firms from entering the market by building a hospital without first obtaining a permit from the Illinois Health Facilities Planning Board (“Planning Board”), which administers the Act. The Planning Board has issued detailed regulations, 77 Illinois Administrative Code § 1100, governing the administration of the Act. 21. For a prospective entrant, the prospects for receiving from the Planning Board a permit to build a new hospital are highly uncertain. The Illinois Health Facilities Planning Act, along with the regulations issued by the Planning Board, authorize the Planning Board to deny applications for permits based on various factors. These include, among others, the potential for duplication of health care services; the desire for orderly development of health care facilities; and the background, character, and financial fitness of the applicant.

VOLUME 144 Complaint 22. Obtaining a permit to build a new hospital may take several years. The Illinois Health Facilities Planning Act authorizes adversely affected companies to seek judicial review under Illinois Administrative Review Law of any final decision of the Planning Board. The regulations of the Planning Board define adversely affected persons to include the incumbent hospitals in the area. These hospitals have a right to intervene in the Planning Board proceedings and to seek judicial review. The time period from application at the Planning Board to completion of judicial review can take several years.

23. The Illinois Health Facilities Planning Act also restricts expansion by current market participants. It requires a permit to expand capacity by more than 10 beds or more than 10 percent of current capacity, whichever is less.

LACK OF MERGER EFFICIENCIES 24. The merger was not necessary to permit the parties to achieve overriding efficiencies to vindicate the merger. Should the matter of efficiencies be placed properly in issue, the evidence establishes that the merger has not led to lower costs at ENH that led to lower prices for consumers. Rather, the merger has led to large cost increases at ENH that coincided with large price increases for consumers. The ability of ENH and Glenbrook hospitals to increase these operating costs and their charges for general acute care inpatient hospital services, without a corresponding improvement in quality of care, further reflects the market power exercised by the hospitals after the merger.

25. Prior to the merger, ENH’s Evanston and Glenbrook hospitals had operating costs comparable to area hospitals and other comparable hospitals. Following the merger, the operating costs at the Evanston and Glenbrook hospitals increased substantially, and much more than experienced by area hospitals and other comparable hospitals.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 9 Complaint 26. Salaries account for the largest portion of operating costs. Following the merger, salary expenses at ENH’s Evanston and Glenbrook hospitals increased substantially, and much more than experienced by area hospitals and other comparable hospitals. VIOLATION 27. The merger of ENH and Highland Park has substantially lessened competition in the relevant market, in violation of Section 7 of the Clayton Act, as amended, 15 U.S.C. § 18. COUNT II: MERGER OF HOSPITALS IN VIOLATION OF CLAYTON ACT §7 28. The allegations of paragraphs 1 through 14 and 19 through 26 are incorporated by reference as though fully set forth herein. COMPETITIVE EFFECTS OF MERGER 29. Following the merger, ENH established a strategy of negotiating with private payers on behalf of the three hospitals as a single system. In many instances, this policy, with the addition of Highland Park to ENH, effectively forced private payers to accept price increases that were significantly higher than the price increases of other comparable hospitals, or face the loss of all three hospitals from their networks. Such a loss would have a significant adverse impact on their ability to market their managed care products. 30. Following the merger, ENH raised prices more than the price price increases implemented by other comparable hospitals. Private Private payers regarded the ENH price increases as unwarranted. ENH also required many private payers to agree to pay prices set at at a discount off of ENH’s list prices in lieu of predetermined per diem prices for each day of inpatient care, a feature of many of the the hospitals’ pre-merger contracts with their major payers. Any VOLUME 144 Complaint pricing system based on list prices makes hospital payments less predictable for private payers and facilitates the hospitals’ ability to to impose unilateral price increases (by raising list prices). ENH raised its list prices several times following the merger. 31. Following the merger, ENH proposed large price increases to its major private payers. All but one of these large customers accepted ENH’s significant postmerger increases rather than try to sell a health plan without any of the three ENH hospitals. In each of the following cases in which it sought to raise prices, ENH also negotiated with the payer hospital and physician services as a package, requiring each payer to accept ENH’s terms for the package or otherwise lose both contracts. (a) United Healthcare of Illinois, Inc. (“United”) is a commercial payer that conducts business in the state of Illinois. As a result of the merger, United faced significantly higher prices for inpatient care. In 2000, ENH raised United’s (i) health maintenance organization (“HMO”) rates by about 52% at the Evanston and Glenbrook hospitals and 38% at Highland Park and (ii) preferred-provider-organization (“PPO”) rates by about 190% for the Evanston and Glenbrook hospitals and 20% for Highland Park as measured by United. As is typical for commercial payers, the vast majority of United’s payments to ENH and other local hospitals are made at HMO or PPO rates. ENH also forced United to pay on the basis of discounts from list prices, which makes payments for hospital services less predictable and potentially even more costly. (b) Private Healthcare Systems (“Private Healthcare”) is a a commercial payer that conducts business in the state of Illinois. As a result of the merger, Private Healthcare faced significantly higher prices for inpatient care. In 2000, ENH raised Private Healthcare’s rates at the Evanston and Glenbrook Glenbrook hospitals by about 40% as measured by Private EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 11 Complaint Healthcare. Evanston also forced Private Healthcare to pay for for some services on the basis of discounts from list prices, which makes payments for hospital services less predictable and and potentially even more costly.

(c) CIGNA Corporation (“CIGNA”) is a commercial payer that conducts business in the state of Illinois. As a result of the merger, CIGNA faced significantly higher prices for inpatient care. In 2000, ENH raised CIGNA’s (i) HMO rates by about 15- 20% and (ii) PPO rates by about 30% as measured by CIGNA. Evanston also forced CIGNA to pay on the basis of discounts from list prices, which makes payments for hospital services less predictable and potentially even more costly. (d) Aetna Inc. (“Aetna”) is a commercial payer that conducts business in the state of Illinois. As a result of the merger, Aetna faced significantly higher prices for inpatient care. In 2000, ENH raised Aetna’s rates by about 45-50% over three years or about 15% per year as measured by Aetna. (e) Humana Inc. (“Humana”) is a commercial payer that conducts business in the state of Illinois. As a result of the merger, Humana faced significantly higher prices for inpatient care. In 2000, ENH raised Humana’s PPO rates by about 50- 60% as measured by Humana.

(f) Preferred Plan, Inc. (“Preferred Plan”) is a commercial payer that conducts business in the state of Illinois. As a result of the merger, Preferred Plan faced significantly higher prices for inpatient care. In 2000, ENH raised Preferred Plan’s rates by about 24% as measured by Preferred Plan. ENH also forced Preferred Plan to pay on the basis of discounts from list prices, which makes payments for hospital services less predictable and potentially even more costly.

VOLUME 144 Complaint (g) HFN, Inc. (“HFN”) is a commercial payer that conducts business in the state of Illinois. As a result of the merger, HFN faced significantly higher prices for inpatient care. In 2000, ENH raised HFN’s exclusive provider organization (“EPO”) rates by about 21% for Highland Park and 25% at Evanston and Glenbrook hospitals and raised HFN’s PPO rates by higher amounts as measured by HFN.

(h) Blue Cross is a commercial payer that conducts business in the state of Illinois, and the largest commercial payer in the Chicago area. Following the merger, ENH proposed a large price increase in both inpatient care and physician services to Blue Cross. Blue Cross challenged ENH’s physician pricing practices as illegal, after which ENH withdrew the proposed price increases to Blue Cross.

VIOLATION 32. The merger of ENH and Highland Park enabled ENH to raise its prices to private payers above the prices that the hospitals would have charged absent the merger. Consequently, the merger has substantially lessened competition in a line of commerce in a section of the country, in violation of Section 7 of the Clayton Act, as amended, 15 U.S.C. § 18.

COUNT III: PRICE FIXING OF PHYSICIAN SERVICES IN VIOLATION OF FTC ACT § 5 33. The allegations of paragraphs 1 through 14, 26 and 31 are incorporated by reference as though fully set forth herein. 34. In many instances, ENH also followed a strategy of negotiating hospital services and physician services (through ENH Medical Group) as a package deal, requiring private payers to accept the terms offered for both hospital and physician services, or face termination of both.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 13 Complaint 35. Faculty Practice Associates, which ENH controls, employs about 460 physicians. These salaried physicians have medical offices in several locations in Cook and Lake counties. For these salaried physicians, Faculty Practice Associates or ENH owns or rents office space for them, employs nurses and other staff that work at the offices, purchases computer technology and other office equipment, and purchases malpractice insurance. ENH Medical Group negotiates prices for the services performed by these salaried physicians. These salaried physicians provide services for a fee charged to commercial payers that ENH Medical Group collects. 36. ENH Medical Group also negotiates prices on behalf of about 450 non-salaried or independent physicians. ENH refers publicly to these physicians as affiliated physicians in contrast to the salaried physicians. These independent or affiliated physicians work at several dozen medical offices in Cook and Lake counties. The independent physicians rent their own office space, hire nurses and other staff, pay for their own computer technology and other office equipment, and purchase their malpractice insurance. The independent physicians provide services for a fee charged to commercial payers that they collect through their own office personnel or administrators.

37. Both the salaried physicians and independent physicians include specialists and primary care physicians that provide comparable services in the same geographic area. In the absence of the price fixing described herein, the salaried physicians and the independent physicians compete in the sale of physician services. This competition reduces the cost of physician services charged to commercial payers that offer health plans to employers and individuals. This competition also improves the quality of services. 38. The ENH Medical Group has negotiated and entered into commercial contracts that contain uniform price terms that cover the VOLUME 144 Complaint the services of both the salaried physicians and the independent physicians. Nearly all of the commercial contracts provide for reimbursement on the basis of fee-for-service, as opposed to capitation or other alternative reimbursement methods. For these commercial contracts, the salaried physicians and the independent physicians do not share expenses, revenues, or profits, or otherwise otherwise share any financial risk.

39. The salaried physicians and the independent physicians have not engaged in any meaningful efficiency-enhancing integration. They do not share information technology systems to enhance services. Nor do they comply or seek to comply with common performance standards or clinical protocols to enhance services. 40. About 300 of the 450 independent or affiliated physicians formerly contracted through the Highland Park IPA. Following the merger, the ENH Medical Group established prices for about 910 physicians – about 460 salaried physicians and 450 independent physicians, including about 300 formerly affiliated with the Highland Park IPA. Following the merger, the ENH Medical Group raised prices.

41. The prices charged for physician services are often set by reference to Medicare’s Resource Based Relative Value System (“RBRVS”), a system used by the U.S. Centers for Medicare and Medicaid Services to determine the amount to pay for physician services to Medicare patients. The RBRVS approach provides a method to determine fees for specific services. Commercial payers often contract with individual physicians or physician groups at a price level specified as some percentage of the RBRVS fee for a particular year, such as 110% of RBRVS.

42. An alternative reimbursement method is for physicians to charge on the basis of capitation. Under capitation, the physician or EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 15 Complaint or physician group charges a set per-member-per-month fee rather than separate fees for specific services. 43. In 2000, ENH Medical Group negotiated price increases for the salaried physicians and independent physicians. In some instances, ENH Medical Group converted capitated contracts to feefor-service contracts with higher effective rates. In other instances, ENH Medical Group raised the amount of the fee-for-service reimbursement. The price increases negotiated and implemented in 2000 after the merger include the following: (a) ENH Medical Group negotiated an increase in the price for Private HealthCare’s PPO from 125% of Medicare RBRVS to 140%.

(b) ENH Medical Group negotiated an increase in the price for United’s PPO from 125% of Medicare RBRVS to 140%, and for United’s HMO from a capitated rate that was comparable to 110% of Medicare RBRVS to 125%.

(c) ENH Medical Group negotiated an increase in the price for Aetna’s PPO from 110% of Medicare RBRVS to 140%. (d) ENH Medical Group negotiated an increase in the price of CIGNA’s PPO from 135% of Medicare RBRVS to 150%, and for CIGNA’s HMO from 115% of Medicare RBRVS to 135%. (e) ENH Medical Group negotiated an increase in the price for One Health’s HMO from 125% of Medicare RBRVS to 140%, and for One Health’s PPO from 130% of Medicare RBRVS to 152.5%.

44. By establishing these and other price increases on behalf of of the salaried physicians and the independent physicians, ENH Medical Group engaged in illegal price fixing in restraint of trade. VOLUME 144 Complaint This conduct deprived commercial payers, employers, and individuals of the benefits of competition among physicians. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 17 Complaint VIOLATION 45. The contracting for physician services engaged in by ENH Medical Group on behalf of its independent physicians constitutes unfair methods of competition in violation of Section 5 of the Federal Trade Commission Act, as amended, 15 U.S.C. § 45. NOTICE Notice is hereby given to the respondents that the tenth day of May, 2004, at 10:00 a.m., or such later date as determined by an Administrative Law Judge of the Federal Trade Commission, is hereby fixed as the time and Federal Trade Commission offices, 600 Pennsylvania Avenue, NW, Room 532, Washington, D.C. 20580, as the place when and where a hearing will be had before an Administrative Law Judge of the Federal Trade Commission, on the charges set forth in this complaint, at which time and place you will have the right under the Federal Trade Commission and Clayton Acts to appear and show cause why an order should not be entered requiring you to cease and desist from the violations of law charged in the complaint.

You are notified that the opportunity is afforded you to file with the Commission an answer to this complaint on or before the twentieth (20th) day after service of it upon you. An answer in which the allegations of the complaint are contested shall contain a concise statement of the facts constituting each ground of defense; and specific admission, denial, or explanation of each fact alleged in the complaint or, if you are without knowledge thereof, a statement to that effect. Allegations of the complaint not thus answered shall be deemed to have been admitted.

If you elect not to contest the allegations of fact set forth in the the complaint, the answer shall consist of a statement that you admit admit all of the material facts to be true. Such an answer shall constitute a waiver of hearings as to the facts alleged in the complaint and, together with the complaint, will provide a record VOLUME 144 Complaint basis on which the Administrative Law Judge shall file an initial decision containing appropriate findings and conclusions and an appropriate order disposing of the proceeding. In such answer, you you may, however, reserve the right to submit proposed findings and and conclusions under § 3.46 of the Commission’s Rules of Practice Practice for Adjudicative Proceedings and the right to appeal the initial decision to the Commission under § 3.52 of said Rules. Failure to answer within the time above provided shall be deemed to constitute a waiver of your right to appear and contest the allegations of the complaint and shall authorize the Administrative Law Judge, without further notice to you, to find the facts to be as alleged in the complaint and to enter an initial decision containing such findings, appropriate conclusions, and order. The Administrative Law Judge will schedule an initial prehearing scheduling conference to be held not later than 14 days after the last answer is filed by any party named as a respondent in the complaint. Unless otherwise directed by the Administrative Law Judge, the scheduling conference and further proceedings will take place at the Federal Trade Commission, 600 Pennsylvania Avenue, NW, Room 532, Washington, D.C. 20580. Rule 3.21(a) requires a meeting of the parties’ counsel as early as practicable before the prehearing scheduling conference, and Rule 3.31(b) obligates counsel for each party, within 5 days of receiving a Respondent’s answer, to make certain initial disclosures without awaiting a formal discovery request.

NOTICE OF CONTEMPLATED RELIEF Should the Commission conclude from the record developed in in any adjudicative proceedings in this matter that the merger of ENH and Highland Park, or any joint venture that combines them, challenged in this proceeding violates Section 7 of the Clayton Act, Act, as amended, the Commission may order such relief against EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 19 Complaint respondents as is supported by the record and is necessary and appropriate, including, but not limited to: 1. Divestiture of Highland Park, and associated assets, in a manner that restores the hospital as a viable, independent competitor in the relevant market, with the ability to offer such services as Highland Park was offering and planning to offer prior to its acquisition by ENH.

2. A ban, for a period of time, on any transaction between ENH and the restored Highland Park that combines their hospitals or other health facilities in the relevant section of the country, except as may be approved by the Commission. 3. A requirement that, for a period of time, ENH provide prior notice to the Commission of acquisitions, mergers, consolidations, or any other combinations of its hospital or other health facilities in the relevant markets with other hospitals or health facilities in the relevant market. 4. A requirement to file periodic compliance reports with the Commission.

5. Any other relief appropriate to correct or remedy the anticompetitive effects of the transaction or to restore Highland Park as a viable, independent competitor in the relevant market.

Should the Commission conclude from the record developed in any adjudicative proceedings in this matter that ENH Medical Group is in violation of Section 5 of the Federal Trade Commission Act, as alleged in the complaint, the Commission may order such relief as is supported by the record and is necessary and appropriate, including, but not limited to, an order that ENH Medical Group shall: VOLUME 144 Initial Decision 1. Cease and desist from fixing, raising, stabilizing, establishing, maintaining, adjusting, or tampering with any fee or aspect of the fee charged for any physician’s service, where the physician is not employed by ENH Faculty Practice Associates, and such conduct is not ancillary to an integrated joint venture.

2. File periodic compliance reports with the Commission. 3. Take other appropriate measures of steps to correct or remedy, or prevent the recurrence of, the anti-competitive practices engaged in by ENH Medical Group. IN WITNESS WHEREOF, the Federal Trade Commission has caused this complaint to be signed by its Secretary and its official seal to be hereto affixed, at Washington, D.C. this tenth day of February, 2004.

By the Commission, Commissioner Harbour dissenting. INITIAL DECISION By Stephen J. McGuire, Chief Administrative Law Judge I. INTRODUCTION A. Overview and Summary of Decision In January 2000, Evanston Hospital (“Evanston”) and Glenbrook Glenbrook Hospital (“Glenbrook”) merged with Highland Park Hospital (“Highland Park”) to form the Evanston Northwestern EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 21 Initial Decision Healthcare Corporation (“ENH” or “Respondent”). Over four years years later, on February 10, 2004, Complaint Counsel for the Federal Federal Trade Commission (“FTC”) filed a Complaint challenging challenging the merger under Section 7 of the Clayton Act, 15 U.S.C. § 18, asserting that the merger has substantially lessened competition.

This case presents a rare opportunity to examine “the actual effect of concentration on price in the hospital industry.” See United States v. Rockford Memorial Corp., 898 F.2d 1278, 1280 (7th Cir. 1990). Since the enactment of the Hart-Scott-Rodino Antitrust Improvements Act of 1976 (15 U.S.C. § 18a), most enforcement actions are initiated prior to the proposed merger. In those cases, courts must rely on predictions based on market concentration data. In this consummated merger case, however, there is significant postacquisition evidence to evaluate in assessing whether the probable effect of the merger will be to “substantially lessen competition.” This opinion follows the traditional Clayton 7 approach in assessing whether there is a reasonable probability that the merger is likely to result in anticompetitive effects in a relevant market. First, the relevant product market and geographic market are determined. Then, the Court analyzes whether anticompetitive effects are probable, using both market concentration statistics and postacquisition evidence. Finally, Respondent’s procompetitive justifications and affirmative defense are assessed. The relevant product market in this case is found to be general acute care inpatient services sold to managed care organizations, including primary, secondary, and tertiary inpatient services. The relevant geographic market is found to be the area encompassing the the following seven hospitals: Evanston, Glenbrook, Highland Park, Park, Lake Forest, Advocate Lutheran General, Rush North Shore, Shore, and St. Francis. See Attachment 1 (DX 8173, map). The postpost-merger market concentration level, as measured by the Herfindahl-Hirschman Index (“HHI”), is found to be 2739, with an VOLUME 144 Initial Decision an increase of 384. This corresponds to a “highly concentrated” market and the presumption that the merger is likely to “create or enhance market power.” Horizontal Merger Guidelines, § 1.51 (1992, as amended 1997), reprinted in 4 Trade Reg. Rep. (CCH) P 13,104.

Contemporaneous and post-acquisition evidence establishes that ENH exercised its enhanced post-merger market power to obtain price increases significantly above its premerger prices and substantially larger than price increases obtained by other comparison hospitals. As a result of the elimination of Highland Park as a competitor, Respondent was able to convert existing price methodologies to managed care organizations to much more favorable post-merger terms than either Evanston or Highland Park could have achieved alone. The evidence further shows that Respondent, in 2002 and 2003, continued to unilaterally raise rates in its chargemaster, which significantly increased the prices paid by managed care organizations for ENH services. The empirical evidence presented by Complaint Counsel’s expert ruled out explanations for the price increases other than market power. Complaint Counsel’s expert compared price increases implemented by ENH post-merger to price increases implemented by other hospitals in her control groups and found that, across all managed care plans, ENH’s price increases exceeded the control groups by 11 to 18%, i.e., if other hospitals raised their prices by 10%, ENH raised its prices by 21 to 28%. Even under Respondent’s expert’s calculations, ENH’s post-merger price increases were 9 to 10% higher than price increases by hospitals in his control groups. This evidence confirms the predictive assessments made by the structural market analysis of market concentration. The evidence presented by Respondent fails to rebut the government’s prima facie case. Upon review, the Court has EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 23 Initial Decision determined that Respondent’s learning about demand theory is flawed, is inconsistent with Respondent’s contemporaneous actions, actions, and is based upon unreliable empirical analysis. In addition, addition, Respondent’s few merger specific improvements to Highland Park do not constitute a sufficiently procompetitive justification to outweigh the competitive harm resulting from the merger. Thus, neither of Respondent’s main defenses, the learning learning about demand theory, nor the quality of care improvements improvements argument, justify the substantial post-merger price increases to managed care organizations and, ultimately, consumers. consumers. Respondent’s other defenses -- its nonprofit status, ease ease of entry, and that Highland Park was a failing firm -- and Respondent’s affirmative defense – that Evanston and Highland Park were already a single entity at the time of the merger -- are similarly unpersuasive. The only viable explanation for Respondent’s anticompetitive prices is that the merger, through elimination of a competitor, enhanced ENH’s market power. Complaint Counsel proved that the challenged merger has substantially lessened competition in the product market of general acute inpatient services and in the geographic market of the seven hospitals described above. Therefore, Complaint Counsel has established a violation of Section 7 of the Clayton Act under Count I of the Complaint. Count II of the Complaint, an alternate pleading, is not dispositive and therefore dismissed as moot. The appropriate remedy for the violation is full divestiture of Highland Park from ENH, which, with ancillary relief, is specified more fully in the attached Order. This is the most effective remedy to restore competition to that which would have existed without the merger and which is necessary and in the public interest to eliminate the ill effects of the acquisition offensive to the statute. B. Summary of Complaint and Answer VOLUME 144 Initial Decision The Complaint in this case charges three counts. Count I alleges that the merger of ENH and Highland Park has substantially lessened competition in the alleged relevant product and geographic market, in violation of Section 7 of the Clayton Act. Complaint PP 16-17, 27. Count II also charges that the merger of ENH and Highland Park has substantially lessened competition, in violation of Section 7 of the Clayton Act, but does not allege a relevant product or geographic market. See Complaint PP 28-32 (the paragraphs alleging the relevant product and geographic markets in Count I, paragraphs 16-18, are not incorporated by reference into Count II). Complaint Counsel argues that Counts I and II are alternative approaches to establishing a violation of Section 7 of the Clayton Act. CCB at 51; Closing argument, Tr. 6546-47. Count III of the Complaint, which includes all claims against ENH Medical Group, Inc., was resolved by a consent agreement with the Commission. The consent agreement was approved and ordered by the Commission on May 17, 2005. Respondent filed an Answer to the Complaint on March 17, 2004; a First Amended Answer on July 12, 2004; and a Second Amended Answer on January 11, 2005 “Answer”). In its Second Amended Answer, Respondent denied the material allegations of Counts I and II of the Complaint and asserted the following defenses: the Complaint fails to state a claim upon which relief can be granted; prior to the merger, Evanston and Highland Park were not separate persons as required for the application of Section 7 of the Clayton Act; the Complaint and the relief sought are not in the public interest; the merger yielded significant procompetitive efficiencies that outweigh any alleged anticompetitive effects; and the merger facilitated significant improvements in the quality of patient care throughout the ENH system that outweigh any alleged anticompetitive effects. Answer, p. 1-15, 20-21. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 25 Initial Decision C. Procedural Background The final prehearing conference was held on February 8, 2005. Trial commenced on February 10, 2005 and continued for eight weeks. Over 1600 exhibits were admitted and forty-two witnesses testified in person. On May 20, 2005, the parties filed post hearing briefs, proposed findings of fact, and conclusions of law. On June 24, 2005, the parties filed responses in reply to the briefs and proposed findings of fact. Closing arguments were heard on July 7, 2005. The hearing record was closed pursuant to Commission Rule 3.44(c) by Order dated July 18, 2005.

By Orders dated February 9, 2005, April 6, 2005, June 8, 2005, August 8, 2005, and October 7, 2005, the Rule 3.51(a) deadline for filing the Initial Decision within one year of the Complaint was extended to December 12, 2005. This Initial Decision is filed within ninety days of the close of the record, pursuant to Commission Rule 3.51(a).

D. Evidence This Initial Decision is based on the exhibits properly admitted admitted in evidence, the transcript of trial testimony, and the briefs, briefs, proposed findings of fact and conclusions of law, and replies replies thereto submitted by the parties. Citations to specific numbered findings of fact in this Initial Decision are designated by by “F.”1 1 References to the record are abbreviated as follows: CX – Complaint Counsel Exhibit RX – Respondent’s Exhibit JX – Joint Exhibit Tr. – Transcript of Testimony before the Administrative Law Judge Dep. – Transcript of Deposition CCFF – Complaint Counsel’s Proposed Findings of Fact CCRFF – Complaint Counsel’s Response to Respondents’ Proposed Findings of Fact CCB – Complaint Counsel’s Post Hearing Brief VOLUME 144 Initial Decision Under the Commission’s Rules of Practice, a party or a nonparty may file a motion seeking in camera treatment for material, or portions thereof, offered into evidence. 16 C.F.R. § 3.45(b). The Administrative Law Judge may order that such material be placed in camera only after finding that its public disclosure will likely result in a clearly defined, serious injury to the entity requesting in camera treatment. 16 C.F.R. § 3.45(b). Pursuant to Commission Rule 3.45(b), several orders were issued granting in camera treatment to material that met the Commission’s strict standard. In addition, when the parties sought to elicit testimony at trial that revealed information that had been granted in camera treatment, the hearing went into an in camera session.

In instances where a document or trial testimony had been given in camera treatment, but the portion of the material cited to in this Initial Decision does not require in camera treatment, such material is disclosed in the public version of this Initial Decision, pursuant to Commission Rule 3.45(a) (the ALJ “may disclose such in camera material to the extent necessary for the proper disposition of the proceeding”). in camera material that is used in this Initial Decision is indicated in bold font and braces (“{ }”) in the in camera version; it is redacted from the public version of the Initial Decision, in accordance with 16 C.F.R. § 3.45(f).

This Initial Decision is based on a consideration of the whole record relevant to the issues and addresses the material issues of fact fact and law. All findings of fact in this Initial Decision are CCRB – Complaint Counsel’s Post Hearing Reply Brief RFF – Respondent’s Proposed Findings of Fact RRFF – Respondent’s Response to Complaint Counsel’s Proposed Findings of Fact RB – Respondent’s Post Hearing Brief RRB – Respondent’s Post Hearing Reply Brief EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 27 Initial Decision supported by reliable, probative, and substantial evidence, as required by 16 C.F.R. § 3.51(c)(1) and In re Chicago Bridge & Iron Iron Co., 2005 WL 120878, Dkt. No. 9300, at 2 n.4 (Op. of FTC Commu January 6, 2005) (also available at http://www.ftc.gov/ os/adjpro/d9300/index.htm). Administrative Law Judges are not required to discuss the testimony of each witness or all exhibits that that are presented during the administrative adjudication. In re Amrep Corp., 102 F.T.C. 1362, 1670 (1983). Further, administrative administrative adjudicators are “not required to make subordinate findings on every collateral contention advanced, but only upon those issues of fact, law, or discretion which are 'material.’” Minneapolis & St. Louis Ry. Co. v. United States, 361 U.S. 173, 193-94 (1959). Proposed findings of fact not included in this Initial Initial Decision were rejected, either because they were not supported by the evidence or because they were not dispositive or material to the determination of the allegations of the Complaint or or the defenses thereto.

II. FINDINGS OF FACT A. The Merger 1. The Merging Parties a. Evanston Northwestern Healthcare 1. Evanston Northwestern Healthcare (“ENH”) is a nonprofit corporation with its office and principal place of business located at 1301 Central Street, Evanston, Illinois 60201. Complaint P 4; Answer P 4.

2. Prior to merging with Lakeland Health Services in 2000, Evanston was comprised of Evanston Hospital, Glenbrook Hospital, ENH Medical Group, ENH Research Institute, and ENH Homecare Services. CX 84 at 6.

VOLUME 144 Initial Decision 3. Evanston Hospital has been affiliated with the Northwestern Feinberg School of Medicine (“Northwestern Medical School”) since at least 1930. Neaman, Tr. 1282. Evanston strengthened its academic relationship with Northwestern Medical School between 1992 and 1996. RX 584 at ENH JH 2951-52; RX 132 at ENH JH 275-77.

4. Mark Neaman, who joined Evanston in 1973, has served as its Chief Executive Officer (“CEO”) since 1992. Neaman, Tr. 1278. Jeffrey Hillebrand, who joined Evanston on a full time basis in 1979, has served as its Chief Operating Officer (“COO”) since 1998. Hillebrand, Tr. 1826-27, 2009.

(1) Evanston Hospital 5. Evanston Hospital has more than 400 beds and is located in Evanston, Illinois. Neaman, Tr. 1291.

6. Evanston had .34 residents per bed in 1999. RX 1912 at 60. 7. Evanston offered obstetrical services, pediatric services, a skilled nursing facility, psychiatric care, neurosurgery, radiation therapy, cardiology services, orthopedics, trauma centers, and the Kellogg Cancer Care Center. CX 84 at 8, 15; CX 681 at 2; Newton, Tr. 299; Spaeth, Tr. 2083-84; Neaman, Tr. 1292. 8. Evanston provides a wide array of inpatient and outpatient services, from basic hospital services (such as obstetrics) to more intensive services (such as cardio-angiogenesis). Rosengart, Tr. 4496; Neaman, Tr. 1291.

(2) Glenbrook Hospital 9. Glenbrook Hospital “Glenbrook”), located in Glenview, Illinois, is a community hospital that was developed, built, and EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 29 Initial Decision opened by Evanston in 1977. CX 84 at 7; Neaman, Tr. 1286, 1292; 1292; Hillebrand, Tr. 1827.

10. Glenbrook is located 12.6 miles and 26 minutes west of Evanston. RX 1912 at 20-21,in camera.

11. Glenbrook has approximately 125 to 150 beds. Neaman, Tr. 1292; CX 681 at 1-2.

12. Glenbrook provides inpatient and outpatient services, but it does not provide obstetrics services. Neaman, Tr. 1292. 13. Glenbrook has a Kellogg Cancer Care Center, center of excellence in orthopedics, and does a significant amount of work in neurology, particularly movement disorders. Neaman, Tr. 1292; CX 681 at 2.

(3) ENH Research Institute 14. The ENH Research Institute, founded in 1996, performs translational clinical research, meaning research that is taken to the bedside. Neaman, Tr. 1289-90. The ENH Research Institute’s translational research is directly related to ENH’s nucleus of clinical activities, such as oncology, cardiology, imaging, and patient outcomes. Hillebrand, Tr. 2007.

15. The ENH Research Institute receives funding from the federal government, including the National Institutes for Health (“NIH”), the National Cancer Institute, and the Department of Defense. Hillebrand, Tr. 2007-08; Neaman, Tr. 1290. 16. In 2004, NIH restructured its clinical research initiatives, including the creation of the Patient Reported Outcome Measurement Information System (“PROMIS”), which is a top NIH NIH priority for measuring the quality of healthcare. Hillebrand, Tr. Tr. 2008. In 2004, and as part of the PROMIS initiative, the ENH VOLUME 144 Initial Decision Research Institute was named the National Coordinating Center for for NIH’s patient outcome studies. Hillebrand, Tr. 2009. 17. ENH has over $100 million in NIH grants. Neaman, Tr. 1290. In terms of NIH funding, ENH ranks twelfth nationally and first in Illinois. Neaman, Tr. 1290.

b. Lakeland Health Services, Inc.

18. Lakeland Health Services, Inc. (“Lakeland Health”), the parent company of Highland Park Hospital (“Highland Park”) prior to the merger, was a nonprofit Illinois corporation with its principal place of business located at 718 Glenview Avenue, Highland Park, Illinois 60035. CX 541 at 1; Newton, Tr. 472; RX 563 at ENH TH 1572.

19. Before merging with Evanston, Lakeland Health Services was comprised of Highland Park Hospital, Highland Park Hospital Foundation, and the for profit Lakeland Health Ventures, Inc. CX 84 at 11. Lakeland Health Services was incorporated in 1982 as a holding company. CX 84 at 12; RX 563 at ENH TH 1572; RX 218 at ENHL TH 229-30.

(1) Highland Park Hospital 20. Highland Park Hospital (“Highland Park”) is located at 718 Glenview Avenue, Highland Park, Illinois 60035, and first opened in 1918. CX 1874 at 1; CX 84 at 12; RX 123. 21. Highland Park is located 13.7 miles and 27 minutes north of Evanston, along Lake Michigan. RX 1912 at 20-21, in camera; Belsky, Tr. 4889.

22. Highland Park has approximately 150 to 200 beds. Neaman, Tr. 1292; CX 84 at 11, 16, In 1999, Highland Park had no residents. RX 1912 at 60.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 31 Initial Decision 23. Highland Park had a medical staff of 562 physicians in 1999. CX 84 at 1, 12.

24. Prior to the merger, Highland Park offered obstetrical services, including: a level II perinatal center; pediatric services; diagnostic services; a skilled nursing facility; a fertility center; psychiatric care; neurosurgery; radiation therapy; cardiology services, including an adult cardiac catheterization lab; an oncology program; and a level II trauma center. CX 84 at 13, 15; CX 699 at 24; Newton, Tr. 299; Spaeth, Tr. 2083-88. 25. Ronald Spaeth was Highland Park’s president and CEO from 1983 up until the merger. Spaeth, Tr. 2235. (2) Highland Park Hospital Foundation 26. The Highland Park Hospital Foundation was Highland Park’s fund-raising arm before the merger. Styer, Tr. 4954. The Highland Park Foundation was tasked with soliciting funds to support Highland Park from individuals and corporations in the general Highland Park community. Styer, Tr. 4954-55, 5001. The Highland Park Foundation was dissolved immediately before, and in anticipation of, the merger. Styer, Tr. 4953. (3) Other Ventures 27. Lakeland Health Ventures, Inc. were for-profit entities owned by Lakeland Health Services. These entities included: Lakeland Primary Care Associates, physician practice management services, real estate ventures, and joint ventures, including a fitness center and a mail order pharmacy. CX 681 at 3; RX 563 at ENH TH 1572. 28. Highland Park also owned 50% of Highland Park Healthcare, Healthcare, Inc., a physician-hospital organization “PHO”). RX 563 563 at ENH TH 1572. The remaining 50% was owned by the VOLUME 144 Initial Decision Highland Park Independent Physicians Association. Chan, Tr. 789, 789, in camera.

2. Premerger Background a. NH North 29. As early as 1994, Neaman and Spaeth, the CEOs of Evanston and Highland Park, respectively, shared the view that hospitals should “stand united” in order to get “better pricing” and “leverage.” CX 1802 at 2-3.

30. Evanston, Highland Park, and Northwest Community Hospital discussed a collaboration as far back as 1996. CX 6305 at 7 (Stearns, Dep.); Neaman, Tr. 1017-18. The entity that would have been created as the result of the proposed merger of Highland Park, Evanston, and Northwest Community Would have been called NH North. Neaman, Tr. 1017-18.

31. One principle of NH North was to be “an entity that differentiates its product, its brand and is indispensable to the marketplace.” CX 395 at 2. The idea behind this branding strategy was to use name-brand to differentiate NH North in such a way that it would be very distinctive and very desirable in the minds of customers. Neaman, Tr. 1363-64.

32. An August 1996 planning document for NH North prepared prepared by Evanston’s CEO, Neaman and Evanston’s COO, Hillebrand explained that for NH North to achieve “market influence” and “indispensability,” it had to achieve “differentiation” “differentiation” and “cost leadership.” CX 394 at 13; Neaman, Tr. Tr. 1018-19; Hillebrand, Tr. 1790-91. According to the planning document, “differentiation” was to be achieved through “superior outcomes,” “brand equity,” and “best physicians.” CX 394 at 13; Hillebrand, Tr. 2020-21. “Cost leadership” was to be achieved through reducing “cost per unit of care,” “develop[ing] pathways,” EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 33 Initial Decision pathways,” and “hospital & physicians common incentives.” CX 394 at 13; Hillebrand, Tr. 2020-21.

33. Bain & Company (“Bain”), a consulting firm to Evanston, was involved in strategizing for NH North. Neaman, Tr. 1024. Bain listed two “key tactics” that should be used by NH North to “gain incremental market share.” RX 477 at ENH JH 349. The two “key tactics” were: (1) “improved/coordinated physician recruitment and development”; and (2) “developing and leveraging brand name.” RX 477 at ENH JH 349.

34. The three-way discussions between Highland Park, Evanston, and Northwest Community with regard to the creation of NH North broke down in 1997 as the result of differences over the proposed merged entity’s organization (such as the composition of the board), personality conflicts, and a lack of interest on the part of Northwest Community. CX 6305 at 7-9 (Stearns, Dep.); Neaman, Tr. 1035; Hillebrand, Tr. 1791-92.

b. Northwestern Healthcare Network 35. The Northwestern Healthcare Network (“Network”) was a system of Chicago area hospitals formed pursuant to an affiliation agreement dated October 23, 1989. CX 6306 at 2 (Mecklenburg, Dep.); RX 22 at NHN 322.

36. The earliest formal discussions concerning the formation of the Network were among a group of hospitals already related to one another through a common affiliation with Northwestern University Medical School. These hospitals included Evanston, the Rehabilitation Institute of Chicago, and Children’s Memorial Medical Center (“Children’s Memorial”). CX 6306 at 2 (Mecklenburg, Dep.).

37. The founding members of the Network were Evanston, Lakeland Health (Highland Park’s parent), Northwestern Memorial, VOLUME 144 Initial Decision Memorial, and Children’s Memorial. Neaman, Tr. 963; CX 1780 at at 1.

38. Pursuant to the affiliation agreement, the Network became the “sole member” of the member hospitals, in accordance with the Illinois General Not For Profit Corporation Act of 1986, as amended. RX 22 at NHN 339, 372.

39. Under the Network, the member hospitals continued to operate as independent entities, operating for their own self-interest. Newton, Tr. 307, 311.

(1) Purposes of the Network 40. The Network hospitals came together to respond to anticipated marketplace behavior in terms of managed care contracting and in terms of exclusive contracting with certain managed care organizations. RX 70 at NHN 873; CX 6306 at 4 (Mecklenburg, Dep.).

41. In particular, the Network was formed, in part, with an eye toward handling the anticipated trend towards capitated contracts, pursuant to which a managed care organization paid a group of providers a fixed amount of dollars per member per month, thus placing all financial risk on that group of providers. Neaman, Tr. 1360.

42. The Network negotiated contracts for the provision of hospital services by its member hospitals with the International Brotherhood of Teamsters, Health Network, Great West, and MultiPlan. CX 6307 at 18 (Schelling, Dep.). The Network negotiated a capitated home health, services agreement with Humana and entered into an agreement with North American Medical Management. CX 6307 at 5-6 (Schelling, Dep.). EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 35 Initial Decision 43. While capitated contracts did come to Chicago in the mid- 1990's, they never became the major factor many had predicted. Neaman, Tr. 1360-61. Thus, one of the driving forces behind the formation of the Network did not materialize in the Chicago area marketplace. RX 584 at ENH JH 2951.

44. Evanston participated in the Network based on its belief that the then-existing Rush, Humana (at that time, Humana owned several hospitals in the Chicago area, including the former Michael Reese Hospital), and Evangelical (a precursor to the Advocate system) systems of ownership of several hospitals in the Chicago area would be the operating model for the future. RX 357 at ENH JH 10385.

45. Highland Park participated in the Network to enhance its quality of care and its perception in the marketplace. Spaeth, Tr. 2194.

(2) Structure of the Network 46. The Network Affiliation Agreement among the four hospital members created a council of governors, consisting of seven representatives named by each of the member hospitals. RX 22 at NHN 340; CX 1780 at 12. The Network Affiliation Agreement gave the council of governors control over the Network, including, inter alia, the authority to appoint and to remove members of the board of directors of the Network. CX 1780 at 14. 47. In addition, the Network had its own executive and its own board of directors. CX 1780 at 12; CX 6306 at 5-6 (Mecklenburg, Dep.); Newton, Tr. 457.

(a) Separate Administrations 48. Under the Network Affiliation Agreement, the governing boards of each hospital retained “local autonomy and control,” of their own hospitals. CX 1777 at 50, 52, 68. VOLUME 144 Initial Decision 49. Each hospital member developed its own budget and operated independently. CX 6307 at 12-13 (Schelling, Dep.). Under the Network Affiliation Agreement, the Network hospitals were autonomous in their financial operations. CX 1777 at 50; CX 6307 at 12-13 (Schelling, Dep.).

50. Under the Network Affiliation Agreement, each institution continued to select, appoint, and employ its own chief executive officer (“CEO”). The duties, functions, and obligations continued to be determined by each institution. CX 1780 at 25. 51. Hospital members did not share day-to-day operational functions. Newton, Tr. 312.

52. The Network could not exercise its discretion to terminate the employment of the administrators of the individual member hospitals, except for limited, specifically defined reasons. CX 1831 at 13.

53. Under the Network Affiliation Agreement, a member of the Network could petition to withdraw from the Network if the Network attempted to implement network-wide managed care agreements that substantially favored one member hospital to the detriment of the withdrawing hospital. CX 1831 at 9-11. 54. Under the Network Affiliation Agreement, a member of the Network could petition to withdraw from the Network if the Network failed to exercise reasonable efforts to support the academic affiliation of that hospital. CX 1831 at 10. (b) Separate Staffs 55. Each hospital in the Network maintained its own medical staff. Hillebrand, Tr. 1786; Newton, Tr. 312. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 37 Initial Decision 56. Each hospital in the Network was responsible for the quality of care at its hospital. Newton, Tr. 312. 57. Under the Network Affiliation Agreement, each member of the Network retained the exclusive authority over granting medical staff privileges at its hospital. CX 1777 at 72. 58. Under the Network Affiliation Agreement, a member of the Network could petition to withdraw from the Network if the Network attempted to require members of that hospital’s medical staff to become members or employees of a network-wide organization. CX 1831 at 10.

59. Under the Network Affiliation Agreement, the medical staff of each hospital remained autonomous. CX 1777 at 49-50, 52. (c) Separate Services and Operations 60. Under the Network Affiliation Agreement, each institution retained autonomy and control over the local-based decisions related to the delivery of health care services. CX 1777 at 52. 61. Each member hospital of the Network developed its own hospital program expansion plans. CX 6307 at 12-13 (Schelling, Dep.).

62. Under the Network Affiliation Agreement, a member of the Network could petition to withdraw from the Network if the Network attempted to implement program expansions or consolidations that substantially favored one member hospital to the detriment of the withdrawing hospital. CX 1831 at 9-10. 63. Each hospital member of the Network maintained its own self-funded health insurance programs for its employees. CX 6307 at 22 (Schelling, Dep.).

VOLUME 144 Initial Decision 64. Each hospital member of the Network maintained its own structure and staff for managed care contracting. Newton, Tr. 312. 65. Each hospital member of the Network retained the authority to enter into a contract or to refuse to enter into a contract with each individual managed care organization. The Network did not have the authority to enter into a contract binding on the individual member hospitals. CX 6307 at 18, 20-21 (Schelling, Dep.). 66. The hospitals that were members of the Network continued to compete with each other, unilaterally negotiating contracts with managed care companies, “'slicing’ each other up in the market,” and “undercutting each other.” CX 1768 at 3. (d) Financial Independence 67. Under the Network Affiliation Agreement, the network hospitals were autonomous in their financial operations. CX 1777 at 50; see CX 6307 at 12-13 (Schelling, Dep.). 68. Members of the Network only shared the cost of running the Network. There was no combined profit and loss or profit-sharing. Members’ balance sheets were separate. Newton, Tr. 311; Neaman, Tr. 973.

69. Member hospitals were not responsible for any debts incurred by other members of the Network. CX 6304 at 4 (Livingston, Dep.) (Evanston); CX 6306 at 5 (Mecklenburg, Dep.) (Northwestern Memorial Healthcare).

70. The Network Affiliation Agreement restricted the authority of the Network to transfer assets of any individual member hospital. CX 1777 at 62.

71. Under the Network Affiliation Agreement, a member of the the Network could petition to withdraw from the Network if the EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 39 Initial Decision Network attempted to impose certain obligations to transfer assets to to another member of the Network. CX 1831 at 9. (3) Dissolution of the Network 72. By 1998, the Network had evolved into a “trade association.” Neaman, Tr. 1008. As a trade association, the Network consisted of a general grouping of hospitals designed to support the general wellbeing of the association. Neaman, Tr. 1008-09. 73. The Network had limited success negotiating contracts with managed care organizations, in part, because it could not bring together the members for contract negotiations. Neaman, Tr. 965-66. Some members were not convinced the Network could get better terms from managed care organizations and, instead, negotiated independently. Neaman, Tr. 966.

74. The cost of running the Network outweighed the value received from the Network, and some questioned whether the Network could generate sufficient value. CX 6306 at 12 (Mecklenburg, Dep.).

75. All members of the Network, including Evanston and Lakeland Health, authorized the dissolution of the network on October 26, 1999. CX 1833 at 2; Neaman, Tr. 1017; CX 872 at 7; RX 592A at ENH RS 880.

76. The member hospitals voted to dissolve the Network rather than submit themselves to the “full control” of the Network. CX 2231 at 4; CX 872 at 7; CX 1833 at 2; Neaman, Tr. 1016-17; RX 592A at ENH RS 880; CX 6306 at 2 (Mecklenburg, Dep.); CX 6305 at 6-7 (Stearns Dep.).

77. The articles of dissolution were adopted by the Network on on December 22, 1999. CX 1833 at 2. The dissolution agreement went into effect on January 2, 2000. Neaman, Tr. 1016; CX 5 at 4. VOLUME 144 Initial Decision The articles of dissolution were filed on January 3, 2000. CX 1833 1833 at 1-2.

3. Merger Agreement 78. The merger discussions that resulted in the merger between Evanston and Highland Park started in late 1998 or early 1999. CX 1 at 2; CX 2 at 7; CX 1879.

79. Neaman, Evanston’s CEO, led the merger discussions from Evanston’s side, while Spaeth, Highland Park’s CEO, led Highland Park’s efforts. Neaman, Tr. 1320; Spaeth, Tr. 2283. Neaman had overall responsibility for the merger and the subsequent merger integration. Neaman, Tr. 955.

80. In April 1999, Evanston and Highland Park signed an agreement to develop a cardiac surgery program at Highland Park. Rosengart, Tr. 4527-30; CX 2094 at 1, 6. In November 1999, the state approved a certificate of need (“CON”) for an open heart surgery program at Evanston and Highland Park. Newton, Tr. 423. 81. The merging parties, including Evanston Northwestern Healthcare, Lakeland Health, and Highland Park, signed a letter of intent to merge effective July 1, 1999. Neaman, Tr. 1328; RX 567 at ENH MN 1365, 1390.

82. Simultaneous with the execution of the letter of intent, Evanston and Highland Park sent a press release to area employers, elected officials, managed care companies, and the press describing the merger. RX 563 at ENH TH 1568-76; Hillebrand, Tr. 1857-58; RX 564.

83. On October 29, 1999, the parties entered into an Agreement and Plan of Merger. RX 651. The effective date of the merger was January 1, 2000. RX 651 at ENH MN 1517.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 41 Initial Decision 84. In the merger agreement, the parties agreed that Lakeland Health and Highland Park would be merged into Evanston Northwestern Healthcare and that Lakeland Health and Highland Park would no longer exist as separate corporations. CX 501 at 17. 85. The merger was consummated on January 1, 2000. CX 501 at 17.

86. ENH subsequently shut down most of the premerger joint ventures operated by Lakeland Health Ventures. Newton, Tr. 449. 4. Post-Merger ENH a. ENH Hospitals 87. Since the merger, the nonprofit ENH healthcare delivery system consists of, among other things, the three hospitals (Evanston, Highland Park, and Glenbrook), a physician multispecialty faculty group practice, a research enterprise, and a charitable foundation. Neaman, Tr. 1281-83. 88. All three ENH hospitals operate as though they are a single hospital entity. Hillebrand, Tr. 1839-42. ENH has one Medicare identification number for all three hospitals. Hillebrand, Tr. 1840-41. 89. ENH consolidated all corporate activities at the Evanston campus and eliminated all corporate functions at Highland Park -including human resources, purchasing, payor contracting, the business office, and information systems. Hillebrand, Tr. 1839-40; Neaman, Tr. 1345-46.

90. ENH instituted one billing system and one business office. Hillebrand, Tr. 1839-40. For example, ENH implemented a coordinated registration, scheduling, and charging system throughout its three hospitals. Hillebrand, Tr. 1840. VOLUME 144 Initial Decision 91. After the merger, Highland Park physicians became part of the medical staff of Evanston and Glenbrook. If a physician had clinical privileges with ENH after the merger, the clinical privileges were good at any of the three hospital sites. RX 518 at ENH GW 2082; Hillebrand, Tr. 1840-41.

92. There are no other hospitals located between Highland Park, Glenbrook, and Evanston. Ballengee, Tr. 167-68; see Attachment 1 (DX 8173 (map)).

93. The three ENH hospitals form a triangle along Chicago’s north shore. Newton, Tr. 351-52; see Attachment 1 (DX 8173 (map)).

94. The driving time from Evanston to Highland Park, or vice versa, is 27 minutes, and the distance is approximately 14 miles. RX 1912 at 20-21, in camera; Spaeth, Tr. 2157. b. Healthcare Foundation of Highland Park 95. As a result of the merger, Evanston and Highland Park also created the Healthcare Foundation of Highland Park on January 1, 2000. Styer, Tr. 4951, 4971; Belsky, Tr. 4894; Spaeth, Tr. 2281. Evanston and the Highland Park Foundation signed the agreement creating the Healthcare Foundation of Highland Park in December 1999. RX 2037; Styer, Tr. 4977-78.

96. The establishment of a separate, post-merger foundation to serve Highland Park was designed to compensate the Highland Park community for the loss of control when Highland Park merged with Evanston. Kaufman, Tr. 5855-56.

97. The Foundation Agreement establishing the Healthcare Foundation of Highland Park describes the Foundation’s mission to to support Highland Park and healthcare in the general Highland EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 43 Initial Decision Park community. RX 2037 at HFHP 1356; Styer, Tr. 4951, 4979; Neaman, Tr. 1373.

98. Spaeth (President and CEO of Highland Park before the merger) has been the president of the ENH Foundation since February 2005. Spaeth, Tr. 2236; Neaman, Tr. 1326. 99. As the head of the ENH Foundation, Spaeth is responsible for growing “friends and funds” from ENH’s communities and to ensure that ENH has the support from these communities for the various healthcare programs that ENH provides. Spaeth, Tr. 2237; Neaman, Tr. 1327.

c. ENH Faculty Practice Associates 100. ENH Faculty Practice Associates is comprised of about 500 employed primary and specialty care physicians. Neaman, Tr. 1287- 88.

101. The ENH Faculty Practice Associates does not include the approximately 1200 non-employed, private practice physicians who have admitting privileges at the three ENH hospitals. Neaman, Tr. 1282.

B. The Health Care Industry 1. Managed Care 102. The competitive dynamics of healthcare markets are distinguishable from other markets in the United States economy. Haas-Wilson, Tr. 2453. This is in part because hospital services are differentiated products. Haas-Wilson, Tr. 2492; Baker, Tr. 4763; Noether, Tr. 5901.

103. “In the context of a differentiated product, it’s difficult to draw a bright line that hospitals inside the bright line are all competitors to each other, and then as soon as you cross that line, there’s no competitive pressure that’s exerted.” Noether, Tr. 5931. VOLUME 144 Initial Decision 104. In addition, in the healthcare market, direct price competition for patients is often attenuated: patients generally pay only a portion of their bill and thus do not react to the entire amount of any change in price made by a hospital. Haas-Wilson, Tr. 2464. 105. There are four different institutional relationships relevant to understanding the competitive dynamics of hospital services. These institutional relationships are between: (1) hospitals and managed care organizations; (2) managed care organizations and employers; (3) employers and employees; and (4) patients and hospitals. Haas-Wilson, Tr. 2456, 2460-64 (discussing DX 7026). a. Hospital – Managed Care Organization 106. The first institutional relationship related to competition for hospital services is the institutional relationship between hospitals and managed care organizations. This relationship is referred to as first stage competition in the economics literature. Haas-Wilson, Tr. 2456.

107. The first institutional relationship between hospitals and managed care organizations is particularly important because it is through this relationship that hospital prices are determined. Haas- Wilson, Tr. 2456. Hospitals sell their services to managed care organizations, and the managed care organizations are the consumer in this first stage competition. Haas-Wilson, Tr. 2456-57; Noether, Tr. 5906.

108. The managed care organization puts together its network of of health care providers by choosing which hospitals will be included in its different plans’ networks, as well as which physician physician organizations and which other ancillary healthcare EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 45 Initial Decision providers will be included in the hospital networks that are offered offered as part of the health plan. Haas-Wilson, Tr. 2456-57. 109. Hospitals compete to be on the hospital network of the health plans offered by managed care organizations. Haas-Wilson, Tr. 2456-57. Managed care organizations build hospital networks to compete effectively with other managed care organizations for employer health plan contracts. Haas-Wilson, Tr. 2456-57. 110. The “customer” in the sale of inpatient hospital services is the managed care organization (as opposed to the individual patient). Noether, Tr. 5924-25; Haas-Wilson, Tr. 2456-57. b. Managed Care Organization – Employer 111. The second institutional relationship related to competition for hospital services is the institutional relationship between the managed care organizations and employers. Health plans sell their products, such as HMO and PPO products, to prospective buyers or employers. Haas-Wilson, Tr. 2460-61.

112. In the employment-based healthcare insurance system in the United States, the employer selects which products of managed care organizations to offer as a fringe benefit to employees. Haas-Wilson, Tr. 2460-61.

113. Employers want to limit the amount of money that they spend on employee health benefits, and, as a result, price competition among managed care organizations is important. Haas- Wilson, Tr. 2461. Therefore, managed care organizations are interested in obtaining the lowest rates possible from the providers that they include in their networks, and this fosters price competition among hospitals and with other providers. Haas-Wilson, Tr. 2457- 58.

114. Viewed from the standpoint of this second institutional relationship, managed care organizations compete with each other to VOLUME 144 Initial Decision to offer hospital networks that are both more attractive to employees employees and that have a low “premium” or price. Haas-Wilson, Tr. Tr. 2461. To be attractive to employers, managed care organizations organizations must provide adequate networks that span the range of of basic and specialty services that employers demand, have good quality reputations, and are geographically convenient to employees employees and their families. Noether, Tr. 5936-37, 5944-45. 115. Consumers prefer a broad choice of hospitals in a hospital network. Haas-Wilson, Tr. 2461.

116. All health plan products have financial incentives to use providers who participate in the plan, although they vary in how “harsh” those incentives are. Haas-Wilson, Tr. 2461-62. 117. Managed care organizations compete on many factors, but the two most important factors are the attractiveness of the network and the price. Haas-Wilson, Tr. 2461.

118. Managed care organizations “are in the business of competing in part based on the provider networks that they put together.” Noether, Tr. 5936. The “managed care organization, to be able to compete, has to have a network that is attractive to enrollees who are the ultimate patients.” Noether, Tr. 5948. c. Employer – Employee 119. The third institutional relationship related to competition for hospital services is the institutional relationship between employers and their employees. Employers who choose to offer health insurance to their employees are offering this health insurance coverage as a form of compensation to their employees. Nevertheless, the employee still bears costs of the health insurance because economic theory shows that the cost of that insurance is “shifted back” to the employee in the form of lower wages. Haas- Wilson, Tr. 2463.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 47 Initial Decision 120. Managed care organizations construct hospital networks to create plans that are attractive to employers. Elzinga, Tr. 2407. The employers, in turn, are driven to provide a plan that is attractive to their employees, because employees may consider health care benefits in deciding where to accept employment. Elzinga, Tr. 2407. Therefore, managed care organizations must take patient preferences into consideration in constructing their hospital networks. Elzinga, Tr. 2407-08; Haas-Wilson, Tr. 2803, in camera; F. 252-55. 121. From the managed care organization’s perspective, the criteria for placing and retaining a hospital in a network include price, reputation, services offered, and location. Mendonsa, Tr. 485 (discussing importance of location); Neary, Tr. 587 (discussing importance of competitive prices); Holt-Darcy, Tr. 1421 (discussing importance of licensing and accreditation); Dorsey, Tr. 1451 (discussing importance of offering appropriate level of care and services).

d. Patient – Hospital 122. The fourth institutional relationship related to competition for hospital services is the institutional relationship between patients and hospitals. When an employee or family member covered under an employer-based health insurance plan needs hospitalization, that patient will, together with his or her physician, select the hospital from which to obtain care. Frequently, the employee, because of the financial incentive offered by the health plan, will choose a hospital in the network. Haas-Wilson, Tr. 2463-64 (discussing DX 7026). 123. Hospitals compete, although not on price, to attract patients patients who are covered by the managed care organizations with which the hospital has contracts. Haas-Wilson, Tr. 2464. This competition for patients after the hospital has entered into contracts VOLUME 144 Initial Decision contracts with managed care organizations is called “second stage competition.” Haas-Wilson, Tr. 2465.

124. To attract patients, hospitals compete, in part, on the quality of care delivered. Noether, Tr. 6011 (“Patients are made better off when quality is improved, and they certainly use quality to the extent that they can evaluate it as one of the dimensions by which they choose hospitals.”).

125. The four institutional relationships related to competition for hospital services have changed over time as a result of the increasing prevalence of managed care. Prior to managed care, most people were covered by “indemnity-based” insurance. Under indemnity-based insurance, discussed more below, these four different institutional relationships would not have existed as is the case today under managed care competition. Haas-Wilson, Tr. 2463- 65.

2. Government Insured and Uninsured 126. In the United States, the majority of people with private health insurance have their health insurance purchased through their employer. However, not everyone is covered by employer-based healthcare insurance. As discussed below, some people have government insurance, while other people are uninsured. Haas- Wilson, Tr. 2454.

a. Government Insured 127. Close to half of ENH’s hospital services are paid by the federal government. Neaman, Tr. 1312. The rates and schedules at which hospitals are reimbursed by the government for providing goods and services to individuals covered by Medicare and Medicaid are publicly available and non-negotiable. Neaman, Tr. 1312, 1317-18; Hillebrand, Tr. 1721.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 49 Initial Decision 128. The prices in public health insurance programs are not determined by competitive market forces or negotiation, but rather are set unilaterally by the government. Haas-Wilson, Tr. 2455; Neaman, Tr. 1317-18.

129. The Medicare program “is a federal health insurance program that provides health insurance for the elderly and those individuals suffering from ... kidney failure and needing renal dialysis.” Haas-Wilson, Tr. 2454.

130. The Medicaid program is “a joint federal/state program” under which “individuals of low income receive health insurance coverage.” Haas-Wilson, Tr. 2454.

131. The federal insurance programs pay a case rate on the basis of the Diagnosis Related Group (“DRG”), which is a “grouping of inpatients into hundreds of separate categories based on their diagnoses and the procedures they undergo while hospitalized.” JX 8 at 5.

132. The DRG reimbursement is “a method of payment in which the reimbursement for inpatient hospital services is set based on the DRG into which a patient is classified. As a general rule, the amount of payment will not vary if the hospital renders significantly greater or less services in treating the patient than is the estimated average, or if the hospital incurs costs that are greater or less than the typical cost incurred by hospitals.” JX 8 at 5.

133. According to a 1999 document, 45% of Highland Park’s revenue came from managed care, 41% from Medicare, 2% from Medicaid, and 12% from other sources. CX 84 at 13. “[E]ssentially, the major payor mix was commercial and Medicare.” Newton, Tr. 301.

134. According to a 1999 document, 51% of Evanston’s revenue came from managed care, 34% from Medicare, 3% from Medicaid, and 12% from other sources. CX 84 at 8.

VOLUME 144 Initial Decision 135. At the start of trial in February 2005, nearly 50% of ENH’s revenue came from government sources such as Medicare and Medicaid. Neaman, Tr. 1312.

b. Uninsured 136. People who do not have health insurance, either through the public sector or commercial plans, are referred to as “uninsured.” Haas-Wilson, Tr. 2454.

137. After Medicare, Medicaid, and the top health plans, there remains for ENH approximately ten percent of gross revenues that fall into a separate category. Neaman, Tr. 1312. Most of this ten percent increment is charity care, although there are a small number of self pay patients in that mix as well. Neaman, Tr. 1312; Newton, Tr. 301.

138. Self pay patients are charged for services based on the hospital’s chargemaster, which are essentially list prices. Porn, Tr. 5685; see F. 174-75.

3. Types Of Managed Care Plans 139. The purpose of a network is to provide employers and their employees with access to the facilities they want and a discount for using those hospitals. Mendonsa, Tr. 485. 140. Managed care plans generally fall within the broad HMO, POS, and PPO categories. “Nevertheless, the different types of managed care plans are difficult to distinguish because, over time, the managed care organizations have modified each type of plan to incorporate different elements of the other plans that consumers demand.” JX 8 at 7.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 51 Initial Decision a. HMO 141. A Health Maintenance Organization (“HMO”) product provides prepaid health insurance coverage to members through a network of physicians, hospitals, and other health care providers that contract with the HMO to furnish such services. RX 1743 at 6. Under an HMO, the insurance company takes the risk. Neary, Tr. 585.

142. Traditionally, an HMO requires that a member’s primary physician approve access to hospitals, specialty physicians, and other health care providers. As a result, the HMO product is the most restricted form of managed care. RX 1743 at 6. The primary physician is called a gatekeeper, who manages the relationship with the patient and will refer the patient to a selected panel of specialists. Hillebrand, Tr. 1834. Pediatricians, family-medicine physicians, internists, and occasionally obstetricians act as gatekeepers. Hillebrand, Tr. 1834.

143. In an HMO network, there are significant economic incentives for the patient to only go to in-network providers. Hillebrand, Tr. 1759-60. HMO networks work on a fixed reimbursement methodology, and only provide benefits to patients if they go to in-network hospitals. Hillebrand, Tr. 1759-60. HMO members receive no benefits for out-of-network usage. Mendonsa, Tr. 477.

144. The “gatekeeper” HMO model has not sold well in Chicago. Hillebrand, Tr. 1834; Mendonsa, Tr. 479; Holt-Darcy, Tr. 1544, in camera. Consumers have rejected closed-panel HMOs and increasingly have demanded “choice.” RX 987 at FTC-LFH 229; Hillebrand, Tr. 1834; Mendonsa, Tr. 479. At most, Chicago had 25% HMO penetration, as compared to 50 to 60% in Los Angeles, New York, and the District of Columbia. Mendonsa, Tr. 479. VOLUME 144 Initial Decision 145. In recent years, consumers have demanded broad hospital networks with few restrictions from their managed care plans. Hillebrand, Tr. 1761-62; RX 1189 at ENHL JL 14126; RX 1346 at BCBSI-ENH 5539. More tightly controlled, traditional HMOs have given way largely to more loosely structured Preferred Provider Organizations (“PPOs”) with large hospital networks and few financial incentives. RX 987 at FTC-LFH 229; Hillebrand, Tr. 1834. 146. At the same time, the distinctions between HMOs and PPOs have blurred. Noether, Tr. 5982. Many HMO plans offer substantial networks, and gatekeeper referrals are no longer always necessary. Noether, Tr. 5982; Hillebrand, Tr. 1834; Foucre, Tr. 881. b. PPO 147. A PPO includes some elements of managed health care, but typically includes more cost-sharing with the member, through copayments and annual deductibles. RX 1743 at 6. With a self-insured PPO product, the employer that contracts with the insurance company is responsible ultimately for the payment of expenses beyond the co-payment and deductible. Neary, Tr. 586. 148. PPOs provide members more freedom to choose a hospital or physician. RX 1743 at 6. In a PPO, the member is encouraged, through financial incentives, to use participating health care providers that have contracted with the PPO to provide services at more favorable rates. RX 1743 at 6. If a member chooses not to use a participating health care provider, the member maybe required to pay a greater portion of the provider’s fees. RX 1743 at 6. 149. A PPO plan offers employers the ability to have different co-payments, deductibles, and other means to make employees EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 53 Initial Decision partially accountable and responsible for paying for their own care. care. Hillebrand, Tr. 1833-34.

c. POS 150. A point of service (“POS”) product tends to have a different configuration and generally involves a network smaller than a PPO network. Ballengee, Tr. 142. POS plans are traditionally between HMOs and PPOs in terms of flexibility and price. Ballengee, Tr. 142-43; Mendonsa, Tr. 479.

151. “A point of service product is one where the in-network benefit or the higher benefit is accessed if [a patient] utilize[s] a primary care physician as opposed to just in and out of network, but there is an out-of-network benefit in that product.” Mendonsa, Tr. 479.

152. With POS products, like with PPO products, the companies “that contracted with the insurance company are responsible ultimately for the payment of [healthcare services].” Neary, Tr. 586. d. Indemnity Insurance 153. In the 1980's, the predominant form of managed care insurance in Chicago was indemnity insurance. Hillebrand, Tr. 1831- 32. Managed care plans grew in importance, crowding out traditional indemnity insurance. Managed care became “the predominant form of commercial health insurance.” Hillebrand, Tr. 1832.

154. Indemnity insurance was insurance “where benefits were given to subscribers. Prices weren’t negotiated with the insurer.” Instead, the insurance company would pay the benefit on behalf of the patient. Hillebrand, Tr. 1832.

VOLUME 144 Initial Decision 155. Under indemnity insurance, the individual covered by insurance could select any hospital, and the insurance company would reimburse the individual for the cost of care according to the plan benefits. Under indemnity insurance, the customer of the hospital would be the individual patient, in contrast to under managed care, where the managed care organization acts as the consumer in first-stage competition. Haas-Wilson, Tr. 2465-67. 156. Under indemnity insurance, hospitals did not have to compete to be part of a network, so there was not the same kind of competition as there is under managed care. Because there was no competition for a place in the provider network under indemnity insurance, hospitals were not competing on price to obtain contracts with managed care organizations. Haas-Wilson, Tr. 2466. e. Self Insurance 157. Administrative services only, or ASO, is the name given when the managed care insurer provides the administrative services, like claims processing, network development, and upkeep, for an employer who chooses to self insure. The employer bears the insurance risk and hires the insurance company to do just the administrative work, such as the bill-paying and the claims processing. Haas-Wilson, Tr. 2571, in camera. 4. Managed Care Contracting a. Selective Contracting 158. Managed care organizations typically do not contract with with all the hospitals in a given geographic area. Instead, they engage in selective contracting – the process by which managed care care organizations negotiate with hospitals. A managed care organization seeks to put together an attractive network for potential EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 55 Initial Decision potential buyers, while at the same time keeping premiums (the prices at which it sells its products) low. Haas-Wilson, Tr. 2457. 159. Through the process of selective contracting, the managed care organization seeks to negotiate a lower price with the hospital while the hospital seeks to negotiate for a higher price. A bargain is struck between the two price objectives. Haas-Wilson, Tr. 2457-58. 160. The managed care organization will only include those hospitals in its hospital network with which there is this sort of bargain over price. Haas-Wilson, Tr. 2457-58. 161. The ability of the managed care organization to exclude a hospital from its network is a powerful tool that defines each side’s bargaining position. Haas-Wilson, Tr. 2470; Noether, Tr. 6189. 162. “Selective contracting” has been one of the fundamental tools of managed care. Noether, Tr. 5980-81. 163. Different managed care plans include different numbers of hospitals depending on the extent to which selective contracting is used. Haas-Wilson, Tr. 2459-60. For example, in the Chicago area, the Great West Healthcare PPO includes 70 hospitals in its hospital network while the Blue Cross Blue Shield PPO includes 93. Haas- Wilson, Tr. 2459-60.

164. Private Healthcare Systems (“PHCS”) contracts with 75 of around 80 to 85 general acute care hospitals in the Chicago area. Ballengee, Tr. 154. PHCS has excluded hospitals because their rates were too high relative to comparable hospitals, including the exclusion of the University of Chicago. Ballengee, Tr. 155-56, 189- 90.

165. Aetna contracts with about 88 out of a total of 100 hospitals hospitals in the Chicago area. Mendonsa, Tr. 484. Aetna terminated VOLUME 144 Initial Decision terminated the Rush hospital system because Rush demanded higher higher prices than Aetna wanted to pay and because Aetna could maintain a viable network without the inclusion of Rush. Mendonsa, Mendonsa, Tr. 568-69, in camera.

166. In general, PPO plans tend to include more hospitals then HMO plans, which tend to have more restrictive networks. Haas- Wilson, Tr. 2460.

167. Highland Park’s CEO testified that he understood that every major insurer in the market had threatened to or actually had left hospitals out of their contracts. Spaeth, Tr. 2193. b. Steering 168. Typically, managed care organizations are able to obtain discounts from providers’ list prices if the managed care organizations can credibly promise to steer patient volume toward the providers. Dorsey, Tr. 1474-75. Such steerage can only occur if certain providers are “preferred” members of the plan’s network. Hillebrand, Tr. 1760-61. Patients are given financial incentives, through lower out-of-pocket expenditures, to use the preferred providers. Hillebrand, Tr. 1759-60. Use of other providers is discouraged by forcing patients to pay larger amounts themselves. RX 1393 at ENHL BW 3691, in camera.

169. With the exception of capitation contracts, managed care organizations in Chicago have not successfully engaged in steering their enrollees from one hospital to another in exchange for better rates. Hillebrand, Tr. 1760-63.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 57 Initial Decision c. Reimbursement Methodologies 170. There are several price arrangements by which a managed care organization and a hospital can contract. The managed care organization can pay charges, per diem, per case, or discount off charges. see, e.g., Holt-Darcy, Tr. 1521, in camera; Ballengee, Tr. 227, 229, in camera.

171. Hospitals use a variety of contract reimbursement methodologies. Hillebrand, Tr. 1833. The different reimbursement methodologies described below can be used for different types of services in the same managed care organization contract. RX 387 at H 2637; RX 1503 at 3651, 3656-67, in camera. 172. These contracts are the result of individualized negotiations between the hospital system and managed care organizations. see, e.g., Ballengee, Tr. 174-76; Mendonsa, Tr. 535-36, in camera; Dorsey, Tr. 1434-38; Foucre, Tr. 886-87; Holt-Darcy, Tr. 1503-04, in camera.

(1) Discount Off Charges 173. A discount off charges rate is a negotiated discount from a hospital’s list price or chargemaster. Chan, Tr. 667. A discount off charges contract is an arrangement by which managed care organizations pay a percentage discount off of the hospital’s chargemaster list price for each component of a service rendered. Chan, Tr. 667; JX 8 at 5.

174. A charge description master, also known as a chargemaster, is a line-by-line listing of all of the clinical activities performed at a hospital. Neaman, Tr. 1349; Porn, Tr. 5638. The chargemaster contains all services provided at a hospital, including inpatient and outpatient services. Porn, Tr. 5646.

VOLUME 144 Initial Decision 175. A hospital’s chargemaster reflects tens of thousands of predetermined itemized amounts (list prices) to be billed for each good or service the hospital provides. Each hospital maintains its own chargemaster. JX 8 at 4; Neaman, Tr. 1349; Hillebrand, Tr. 1710; Chan, Tr. 674; H. Jones, Tr. 4143. 176. ENH’s chargemaster has 15,000-20,000 line items. Neaman, Tr. 1349; RX 641 at ENH KG 627.

177. Escalator clauses may protect a managed care organization from a hospital’s chargemaster increases. Newton, Tr. 459. Such clauses are put into a discount off charges contract to protect the managed care organization in case charges go up. Mendonsa, Tr. 566-67, 558, in camera. For example, where a contract is for 50% of charges and the escalator clause is 5%, if the hospital were to raise its prices by 10%, then the discount would increase to 55% percent to offset the charge increase. Mendonsa, Tr. 567, in camera. (2) Per Diem 178. Under the per diem reimbursement, the fixed rate per day is an all-inclusive amount for each day that the patient is in the hospital, regardless of the amount of services or the costs or charges for the services that actually must be rendered to that patient. JX 8 at 8-9.

179. A per diem is a predictable expense. Mendonsa, Tr. 524-25, in camera. A per diem means that managed care organizations pay a fixed amount to the hospital per day of inpatient stay regardless of what services are provided. Ballengee, Tr. 228, in camera. There can be different per diems for different categories of service, e.g., medical/surgical versus intensive care unit. Ballengee, Tr. 228, in camera.

(3) Case Rates EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 59 Initial Decision 180. A per case rate is an all-inclusive charge for an entire case (such as the delivery of a baby based on the length of stay). Ballengee, Tr. 229, in camera. Managed care organizations prefer case rates because, like per diem rates, they allow the managed care organizations to fix their costs and price their products accordingly for the coming year. Sirabian, Tr. 5740. (4) Capitation 181. In capitated contracts, the parties typically negotiate a fixed amount that the provider receives for agreeing to care for each patient, regardless of how much care the patient seeks during the period in question. Mendonsa, Tr. 525, in camera; Holt-Darcy, Tr. 1537-38, in camera. Capitated contracts shift financial risk to providers, to align the incentives of those who provide care (the hospitals and physicians) with those who must pay for it. Mendonsa, Tr. 525, in camera; Holt-Darcy, Tr. 1537-39, in camera. 182. When health plans pay a fixed per diem or per case rate, it is not capitation. Hospital capitation has not been common in the Chicago market. Spaeth, Tr. 2129-30; Holt-Darcy, Tr. 1537-39, in camera; Mendonsa, Tr. 525, in camera.

5. Hospital Costs 183. Congress passed the Balanced Budget Act of 1997 (“Balanced Budget Act”) as part of a larger deficit reduction package. Pub. L. 105-33, 1997 H.R. 2015; Neaman, Tr. 1314; H. Jones, Tr. 4106. Overall, the Balanced Budget Act was intended to reduce the annual rate of Medicare spending growth. Neaman, Tr. 1314. The Balanced Budget Act did, in fact, reduce expenditures in a number of areas, including: general hospital payments, teaching, research, home care, and payments to physicians. Neaman, Tr. 1314- 15.

VOLUME 144 Initial Decision 184. The reduction in general hospital payments placed significant strain on hospitals’ abilities to cover many of their high fixed (or shared) costs. H. Jones, Tr. 4106, 4145-47; Noether, Tr. 5973. Additionally, these reductions limited hospitals’ abilities to care for their uninsured patients. According to federal regulations, hospitals must provide emergency care to all who require it, regardless of their ability to pay. 42 U.S.C. 1395dd; 42 C.F.R. § 489.24.

185. Passage of the Balanced Budget Act coincided with a continuing decline in the growth of payments from managed care organizations. RX 1346 at BCBSI-ENH 5540. Traditionally, payments from private payors helped hospitals meet the costs of providing unprofitable services -- such as caring for the uninsured and training residents. RX 1393 at ENHL BW 3681, in camera. Meeting costs via cross-subsidization was practiced by some hospital administrators. Haas-Wilson, Tr. 2684-85. 186. Along with Medicare payment reductions and a declining ability to shift costs, hospitals have encountered other payment challenges since the Balanced Budget Act’s passage: rising liability insurance costs’ stock market declines; new expensive technological developments; and increased labor costs. RX 1393 at ENHL BW 3681, in camera; H. Jones, Tr. 4108.

187. Managed care organizations could absorb provider price increases without passing them on to consumers. For instance, Health Care Service Corporation, the parent of Blue Cross, posted net gains of over $624 million in 2003, $387 million in 2001 and $173 million in 2000. RX 1587 at 7; RX 1198 at 6-7. Humane is one one of the nation’s largest publicly traded health benefits companies, companies, based on 2003 revenues of $12.2 billion. RX 1743 at 4, 4, 27. In 2003, PHCS reported that its net revenue climbed to $153 $153 million, an increase of 6% over 2002. RX 1615 at 3. Further, PHCS’s earnings increased by “an astounding 50%” in 2003. RX 1615 at 3. Cigna posted net income of $668 million in its 2003 EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 61 Initial Decision financial statements. RX 1742 at 54. As of February 2005, United Health Group was worth over $30 billion. Foucre, Tr. 939; RX 1662 1662 at 225, 227. First Health, which acquired CCN in August 2001, 2001, had net income of $152,734,000 in 2003, up from $132,938,000 in 2002, $102,920,000 in 2001, and $82,619,000 in 2000. RX 1661 at 50; RX 1469 at 104.

188. Managed care representatives testified that employees ultimately bear the cost of higher health care prices. When hospitals raise their rates to managed care organizations, those higher rates are passed on to the managed care organizations’ employer groups and further to the employer groups’ employees. Ballengee, Tr. 239, in camera (PHCS); Mendonsa, Tr. 483 (Aetna); Dorsey, Tr. 1450 (One Health).

189. A self-insured customer or large employer group, in the event of unforeseen increases in expenses, may pass on the costs to its employees. Mendonsa, Tr. 483-84; Ballengee, Tr. 239, in camera. Large employers can “raise the deductible, raise the co-payments and also charge more out of [the employee’s] paycheck for the coverage.” Mendonsa, Tr. 549, in camera. 190. “The big impact” of managed care organizations passing on large increases to their smaller business customers is “small insureds dropping coverage altogether and people not having insurance.” Mendonsa, Tr. 483-84.

C. Relevant Market 1. Product Market 191. The relevant product market is the market for “general acute care inpatient services sold to managed care organizations.” Haas-Wilson, Tr. 2451-52; see F. 192-211. VOLUME 144 Initial Decision 192. Primary, secondary, and tertiary services are included in the relevant product market. Haas-Wilson, Tr. 2661; see F. 197-200. 193. ENH’s economic expert, Dr. Monica G. Noether, Vice President, Charles River Associates, agrees that specialty hospitals that do not provide the full range of hospital services, that may be specialized either in a particular service or for a particular category of patients, are excluded from the relevant product market. Noether, Tr. 5924.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 63 Initial Decision a. Definitions 194. Acute care hospital services are “[s]ervices furnished to patients with acute needs for health care services, as distinguished from services furnished for chronic physical conditions through the provision of long-term inpatient care.” Noether, Tr. 5905; JX 8. 195. Inpatient hospital services are furnished to a patient who, to obtain the services, must stay overnight at the hospital. Ballengee, Tr. 144; Neary, Tr. 590; JX 8.

196. Outpatient hospital services are furnished to patients who do not require an overnight stay at the facility. CX 6321 at 82; Newton, Tr. 302; JX 8.

197. Primary services refers to the basic care that is typically provided by physicians or nurse practitioners who work with general and family medicine, internal medicine, pregnant women, and children. Noether, Tr. 6159. Primary services could include things such as basic hospital outpatient services and basic minor surgery. Neaman, Tr. 1293.

198. Secondary services refers to care given by a specialist or a facility upon referral by a primary care provider, and generally requires more skill, expertise, or equipment than primary care services. Noether, Tr. 6159.

199. Tertiary services refers to more complicated services than primary or secondary, but less complicated services than quaternary services. Haas-Wilson, Tr. 2491. Tertiary care generally means major surgical or medical procedures that are done within a hospital setting. Neaman, Tr. 1294.

200. Quaternary services refers to high-end services that are performed at some hospitals and not others. Neaman, Tr. 1294; Haas-Wilson, Tr. 2701, in camera. Quaternary services, which VOLUME 144 Initial Decision include solid organ transplants and treatment for severe burns, require very specific human capital, including trained nurses and doctors, and very specialized physical capital, including specialized specialized equipment. Haas-Wilson, Tr. 2701, in camera. b. Services Provided by the Merging Parties 201. Before the merger, both Highland Park and Evanston had, among other things, operating rooms, pediatric services, obstetrical services, radiation therapy, cancer services, and psychiatric services. Spaeth, Tr. 2083-88.

202. Before the merger, both Highland Park and Evanston provided primary and secondary services. Holt-Darcy, Tr. 1507, in camera; Haas-Wilson, Tr. 2491, 2316. Evanston provided tertiary services before the merger, while Highland Park generally did not. Haas-Wilson, Tr. 2491.

203. None of the hospitals comprising ENH offer advanced, quaternary services, such as organ transplants and severe burn care. Haas-Wilson, Tr. 2665; Ballengee, Tr. 188-89. c. Outpatient Services Not a Substitute for Inpatient Services 204. None of the outpatient centers in the Evanston area have 24-hour nursing or lodging of patients. Spaeth, Tr. 2076. 205. The physician determines whether a patient should be admitted to the hospital. Hillebrand, Tr. 1756; Spaeth, Tr. 2076; Newton, Tr. 302.

206. If a patient requires more than a day of medical or surgical services as an inpatient, managed care organizations cannot substitute outpatient services. Holt-Darcy, Tr. 1422-23; Newton, Tr. 302.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 65 Initial Decision 207. Changes in inpatient pricing have no impact on patients switching from inpatient services to outpatient services. Neaman, Tr. 1210; Hillebrand, Tr. 1755-56.

208. When faced with a price increase for inpatient care from a hospital, managed care organizations could not add to the network outpatient only providers and exclude the higher priced hospitals. Haas-Wilson, Tr. 2663.

209. ENH set its inpatient rates independent of its outpatient rates and without concern that patients would switch to outpatient services. Neaman, Tr. 1210-11; Newton, Tr. 330-31. 210. When ENH developed its plan to negotiate higher prices, ENH management did not prepare or ask for any documents analyzing whether more patients would switch from inpatient to outpatient services as a result of changes in inpatient prices. Neaman, Tr. 1210-11; see Hillebrand, Tr. 1756. 211. ENH’s expert agrees that inpatient and outpatient services are not functionally interchangeable. Noether, Tr. 6194. 2. Geographic Market a. Elzinga-Hogarty Test and Patient Flow Data Are Not Relevant to the Geographic Market Analysis 212. The Elzinga-Hogarty test, which was developed for the beer and coal industries prior to the development of the Merger Guidelines, has been utilized in a number of hospital merger cases. Elzinga, Tr. 2374-76.

213. The Elzinga-Hogarty test is premised on the assumption that patient flow data affects market prices. Elzinga, Tr. 2356. VOLUME 144 Initial Decision 214. Patient flow data is data regarding where patients go to obtain hospital services. Elzinga, Tr. 2356, 2375; Noether, Tr. 6203- 04.

215. Under the Elzinga-Hogarty test, the geographic market is based on the area from which the hospital attracts its patients (its service area) and where patients within that service area go to receive healthcare. Elzinga, Tr. 2380-81. 216. Patient-flow data and the Elzinga-Hogarty test are inapplicable to geographic market definition for a differentiated product such as hospital services. Elzinga, Tr. 2384-85. 217. The first problem with use of patient flow data and the Elzinga-Hogarty test is the “payor problem,” which recognizes that in the hospital industry, managed care organizations pay for hospital services but patients are the ones who use the services. Elzinga, Tr. 2395.

218. Because patients do not set the price of hospital services, their willingness to travel tells us nothing about their sensitivity to price changes by the merging hospitals. Elzinga, Tr. 2395-97. 219. The second problem with patient flow analysis is that it incorrectly assumes that if some patients are willing to travel to distant hospitals, then others will travel as well in response to a change in hospital prices, thereby incorrectly suggesting a broader geographic market. Elzinga, Tr. 2385-90. 220. A “silent majority” of people will not travel in response to a change in hospital prices, and those people can be subject to an anticompetitive price increase. Elzinga, Tr. 2385-90. 221. Hospitals frequently consider patient flow data in evaluating competition and service areas. RX 518 at ENH GW EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 67 Initial Decision 2055-57, 2059; RX 2021 at ENH DL 3443, in camera; RX 135 at 4; 4; RX 1361 at 1; RX 1564, in camera.

222. However, basing geographic market definition on patient migration and patient flow data inherently will overstate the size of the geographic market for hospital services. Elzinga, Tr. 2393. 223. Patient flow data should not be used to determine the geographic market for hospital services, even apart from the Elzinga-Hogarty test, because the same payor and silent majority problems exist. Elzinga, Tr. 2417-18.

224. While Respondent’s expert, Noether, did not use the Elzinga-Hogarty test for the purpose of defining the geographic markets, she did use patient flow analysis as one factor in defining the proposed geographic market. see, e.g., Noether, Tr. 5947-48. 225. Noether conceded that patient flow data is focused on which hospitals patients ultimately choose for care and that one would not want to rely on patient flow data by itself to determine the geographic market. Noether, Tr. 6203-04. b. Market Participant Views (1) Managed Care Organizations 226. In the Chicago area, provider networks must include local hospitals. For example, PHCS’s representative stated that people “do not like to drive by a local hospital and have to go to another hospital.” Ballengee, Tr. 184.

227. Local hospitals in this particular geographic area are important to include in hospital networks because this was an area populated by “senior executives and decision-makers” who are “educated” and “outspoken” and it would be “real tough” for any managed care organization and employer “whose CEOs either use VOLUME 144 Initial Decision this place or that place to walk from [ENH] and 1700 of their doctors.” CX 4 at 2; Foucre, Tr. 901-02, 926; Spaeth, Tr. 2242; Newton, Tr. 360-61 (Within the triangle formed by the ENH hospitals live many executives who “make decisions about health benefits for their employers, employees,” and have “immense influence and power with the health plans.”). 228. This managed care testimony is consistent with economic literature findings that affluent consumers may be less willing to travel because they “impute a higher value to their time and consequently travel becomes more costly to them in the opportunity cost sense ... affluent people have to stay close to home ... so they can move on earning their – the high income that makes them affluent.” Elzinga, Tr. 2408.

229. Managed care representatives described Evanston and Highland Park as each other’s “main” competitors or “primary” alternative, thereby permitting managed care organizations to “trade off one for the other” or “work them against each other” in contract negotiations. Neary, Tr. 600-02; Ballengee, Tr. 166-70. 230. Aetna could constrain Evanston’s prices by utilizing Highland Park (and others) in its network as an alternative (and vice-versa). Mendonsa, Tr. 520, 530, in camera. 231. PHCS knew that if rate negotiations were not “going well” at Evanston or Highland Park, PHCS could turn to the other as the alternative and use this fact to work the negotiations favorably its way. Ballengee, Tr. 166-67.

232. One Health viewed Evanston and Highland Park as “main competitors” because their services were “comparable,” and the two hospitals drew patients from the same general population. Neary, Tr. 600-01.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 69 Initial Decision 233. Managed care representatives testified that they needed ENH in their hospital networks. Ballengee, Tr. 179-80 (PHCS customers made it “very clear” that a network without ENH was not “marketable.”); Foucre, Tr. 901-02, 925-26, 931-34 (United concluded it “could not have a viable network that would support our sales and growth objectives” without ENH). For example, there was testimony that “people would choose either to go north to [Highland Park] or south to [Evanston]. They could go either way and receive the same services at the same level. So, it was pretty well assumed that we could have one or the other hospital in the network.” Ballengee, Tr. 166, 168 (migration tends to be northsouth.).

234. The Unicare representative testified that she could have a viable network comprised of Highland Park, Advocate Lutheran General, Rush North Shore, and St. Francis or Evanston and Lake Forest. Holt-Darcy, Tr. 1518-20, in camera. Either of these alternative networks could “provide medical services adequately” and meet the “geographic access standards” of local Unicare customers. Holt-Darcy, Tr. 1519-20, in camera. 235. The Aetna representative testified that Evanston competed locally with Rush North Shore and St. Francis and that Highland Park competed locally primarily with Lake Forest. Mendonsa, Tr. 562, in camera.

236. The PHCS representative testified that premerger, Advocate Lutheran General, Rush North Shore, and St. Francis, were significant competitors to Evanston, and that Lake Forest was a significant competitor to Highland Park. Ballengee, Tr. 211-12. 237. The PHCS representative testified that for purposes of developing its network and deciding which hospitals to include in its its network, she viewed the service and quality of Advocate Lutheran General, possibly Rush North Shore, and possibly VOLUME 144 Initial Decision Advocate Northside to be comparable to Evanston. Ballengee, Tr. 191-93.

238. When PHCS notified its customers about the merger, it identified “other contract providers within the same geographical area as that of Highland Park Hospital and Evanston,” including: Lake Forest, Advocate Lutheran General, Rush North Shore, St. Francis, and Holy Family Medical Center. RX 712 at PHCS 891; Ballengee, Tr. 213-14.

239. The Great West representative testified that the main alternatives to ENH were: Advocate Lutheran General, St. Francis, Condell, and Northwestern Memorial. Neary, Tr. 630-31. 240. Great West provided its subscribers with a list of hospitals that were in its network that could be alternatives to ENH, including: Lake Forest, Advocate Lutheran General, St. Francis, and to the north, St. Therese and Victory Memorial (now the Vista hospitals). Dorsey, Tr. 1479-80.

241. The Unicare representative testified that ENH competes with Lake Forest, Rush North Shore, St. Francis, and Advocate Lutheran General to some degree. Holt-Darcy, Tr. 1596-98, in camera. According to the Unicare representative, Evanston also competes with the other tertiary hospitals in the Chicago area and may compete with Louis A. Weiss to some degree. Holt-Darcy, Tr. 1596-97, in camera. When asked whether Highland Park competes with Condell, Holt-Darcy replied “[l]ess so, because it is a little further west.” Holt-Darcy, Tr. 1596, in camera. 242. The United representative testified that Evanston competes competes with Advocate Lutheran General, Rush North Shore, and and St. Francis and that Highland Park primarily competes with Lake Forest and Condell. Foucre, Tr. 941-44. The United representative also testified that Evanston competes with EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 71 Initial Decision Northwestern Memorial in respect to certain services. Foucre, Tr. 946.

VOLUME 144 Initial Decision (2) ENH 243. Evanston and Highland Park viewed each other as competitors premerger. CX 1868 at 3; Neaman, Tr. 1046; Spaeth, Tr. 2088.

244. Highland Park, prior to the merger, considered its closest or primary competitor to be Lake Forest, although it also was “reasonably close” to Advocate Lutheran General, Rush North Shore, Evanston, and Condell. Spaeth, Tr. 2239-40; Chan, Tr. 730; CX 6305 at 5 (Stearns, Dep.); Krasner, Tr. 3700. 245. Spaeth, Highland Park’s President, indicated that he believed that managed care organizations could exclude Highland Park from a network and substitute Evanston, Lake Forest, Advocate Lutheran General, Rush North Shore, St. Francis, and Condell. Spaeth, Tr. 2299.

246. Neaman, Evanston’s CEO, testified that Condell and Lake Forest were competitors of Evanston, but that Highland Park was not a substantial competitor of Evanston. Neaman, Tr. 1381-82. 247. ENH described its combined core service area as including: Evanston, Highland Park, Lake Forest, Advocate Lutheran General, Rush North Shore, St. Francis, downtown teaching hospitals, and “other” hospitals. CX 359 at 16.

248. According to ENH representatives, ENH’s major competitors for “more sophisticated” or “tertiary” services include: Lake Forest, Advocate Lutheran General, Rush North Shore, St. Francis, Condell, Northwestern Memorial, Rush-Presbyterian-St. Luke’s, and University of Chicago, because all of these hospitals offer a comparable breadth and type of services. Hillebrand, Tr. 1748-51; Neaman, Tr. 1301.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 73 Initial Decision 249. The merging parties recognized that hospital competition is local. “What Evanston does is provide total concentration” and that “[i]f one of your key objectives is to get geographic leverage on the employers in this area getting Northwestern [Memorial] doesn’t do much for you.” CX 4 at 9; Spaeth, Tr. 2213-14. see also CX 4 at 9 (board member noted that a merger with Northwestern Memorial would not provide “critical mass in the same area.”). 250. At an April 5, 1999, meeting of the medical staff executive committee at Highland Park, Neaman commented on the “geographic advantages” of a merger between Evanston and Highland Park. Spaeth, Tr. 2213-14; CX 2 at 7. 251. In a joint 1999 submission to an Illinois healthcare agency for approval to extend Evanston’s heart surgery program to Highland Park, the hospitals stated:

Last, a concept that is often misunderstood by persons not living in suburban communities is that many suburban residents rarely travel from their general area of residence for shopping, business and health care services. For this reason, many of the anxiety and convenience-related issues related to a resistance to travel for care, that are typically associated with smaller communities, also exist in the suburbs.

CX 413 at 83.

c. Other Factors Relevant to the Geographic Market Determination 252. Managed care organizations consider a variety of factors in in building their hospital networks, including: patient preferences, geographic needs, marketing needs, credentialing, physician preferences, quality of services, breadth of services, ease of VOLUME 144 Initial Decision accessibility, and residence of the individuals who negotiate contracts with managed care plans. Elzinga, Tr. 2407; Haas-Wilson, Haas-Wilson, Tr. 2803-05, in camera; Noether, Tr. 5937, 5949; Foucre, Tr. 885; Mendonsa, Tr. 485; Holt-Darcy, Tr. 1420-21; Ballengee, Tr. 151-53.

253. Employers are concerned about where their employees want to seek hospital care. Noether, Tr. 5936-37, 5948. Consequently, to the extent that patients value convenience, there is a derived demand by the managed care organizations for hospitals that are convenient to their enrollees. Noether, Tr. 5937; Elzinga, Tr. 2407. 254. The Unicare representative testified that a managed care organization wanted “to make sure that members have access to a hospital within 30 miles of where they live or work” in order “to meet the standards that the plans put together.” Holt-Darcy, Tr. 1420. Thus, the Unicare representative testified that “[y]ou look at geographic need, you look at marketing needs, you look at access” and that “[y]ou want to see what population you have or potentially have, what marketing thinks that they need in a particular service area.” Holt-Darcy, Tr. 1420.

255. Driving times may be a better measure of geographic proximity than driving distances because distances do not account for variations in road and/or traffic patterns that can affect patient preferences. Noether, Tr. 5933.

256. Noether computed the driving times from Evanston and Highland Park to other area hospitals. RX 1912 at 20-21, in camera. The actual driving time will vary for each patient, depending on where he or she lives or works. see Noether, Tr. 5929. 257. According to a Lake Forest customer survey report, dated November 8, 2001, consumers are willing to travel, on average, up EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 75 Initial Decision up to 16 minutes for emergency care, 28 minutes to a primary care physician for routine care, 31 minutes for outpatient services, and 35 35 minutes to a hospital for an overnight stay. RX 1179 at LFH 845. 845.

258. The average driving distance from Lake Forest, Advocate Lutheran General, Rush North Shore, and St. Francis to the closer of Evanston or Highland Park is 5.75 miles, while the average driving time is 12.75 minutes. see RX 1912 at 20, in camera. 259. The average driving distance from Condell and Resurrection to the closer of Evanston or Highland Park is 12.4 miles, while the average driving time is 24.5 minutes. see RX 1912 at 20, in camera.

260. By either mileage or minutes, the travel time from the closer of Highland Park or Evanston to the hospitals excluded from the geographic market is almost double the mileage or minutes from the closer of Highland Park or Evanston to the hospitals included in the geographic market. Compare F. 258 to F. 259. 261. Physician admitting practices are significant “because the physician is the one who is often the most responsible for choosing where a particular patient is going to be admitted to a hospital.” Noether, Tr. 5949.

d. Hospitals Included in the Geographic Market 262. The hospitals below, which are part of the geographic market, were all included in Respondent’s proposed geographic market. see Noether, Tr. 5928, 5960. In addition, Respondent’s proposed geographic market included two additional hospitals which are discussed infra in section II.C.2.e. VOLUME 144 Initial Decision (1) Evanston 263. see F. 5-8, 201-03.

(2) Glenbrook 264. see F. 9-13, 203.

(3) Highland Park 265. see F. 20-24, 201-03.

(4) Lake Forest 266. Lake Forest is 6.1 miles and 13 minutes (driving time) northwest of Highland Park. Neaman, Tr. 1304; Spaeth, Tr. 2239-40; Mendonsa, Tr. 555, in camera; RX 1310 at FTC-LFH 669; RX 1912 at 20-21, in camera.

267. Lake Forest is a 142 bed hospital with a very active obstetrics program, roughly the same size as Highland Park’s obstetrics program. Hillebrand, Tr. 2005; RX 1912 at 60. Lake Forest Hospital does not provide any tertiary care. Neaman, Tr. 1304.

268. Lake Forest had no residents per bed in 1999. RX 1912 at 60.

269. There was a substantial overlap of admitting physicians who had privileges and admitted patients at both Highland Park and Lake Forest prior to the merger. Noether, Tr. 5950; RX 653 at ENH DL 4497. Once the merger was announced, a number of these physicians shifted their admissions to Lake Forest. Noether, Tr. 5950; RX 653 at ENH DL 4498.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 77 Initial Decision 270. Lake Forest was identified in contemporaneous PHCS and Great West correspondence to patients as an alternative to ENH. RX 712 at PHCS 891; Ballengee, Tr. 213-14; Dorsey, Tr. 1478-80. 271. Managed care representatives testified that Lake Forest is a significant competitor to ENH. Ballengee, Tr. 212 (PHCS); Foucre, Tr. 944 (United); Holt-Darcy, Tr. 1596, in camera (Unicare); Mendonsa, Tr. 562, in camera (Aetna); Spaeth, Tr. 2239, 2299 (Highland Park).

(5) Advocate Lutheran General 272. Advocate Lutheran General is 10.2 miles or 21 minutes (driving time) west and slightly south of Evanston. Neaman, Tr. 1297; RX 1912 at 20-21, in camera; see also Mendonsa, Tr. 556, in camera.

273. Advocate Lutheran General is a 521 bed tertiary care hospital that is the largest hospital in the Advocate system, which itself consists of eight hospitals. Neaman, Tr. 1296-97; see also Ballengee, Tr. 225, in camera; RX 1503 at PHCS 3667, in camera; RX 1912 at 60; Mendonsa, Tr. 558, in camera. Through the end of 2000, Advocate Health Care was the overall market share leader in the Chicago metropolitan area and the largest hospital system in the Chicago area. RX 1053 at AHHC 363, in camera; Mendonsa, Tr. 558, in camera.

274. Advocate Lutheran General provides all basic services, cardiac surgery, and most everything in between. Neaman, Tr. 1297. Advocate Lutheran General also has a teaching relationship with the University of Illinois at Chicago Health Services Center. Neaman, Tr. 1297.

275. Advocate Lutheran General had .36 residents per bed in 1999. RX 1912 at 60.

VOLUME 144 Initial Decision 276. In terms of range of services, Advocate Lutheran General is the most similar to Evanston Hospital. Haas-Wilson, Tr. 2706, in camera. The United representative stated: “Lutheran General is the most comparable facility from type of services, quality of services, size of facility; however, it is the furthest away. It’s got a bit of geographic disadvantage, but it’s not terribly far away.” Foucre, Tr. 944.

277. Before the merger, patients who went to the emergency room at Highland Park or Lake Forest with a heart attack were referred to Advocate Lutheran General for more advanced care. Spaeth, Tr. 2241-42.

278. ENH, during contract negotiations with PHCS, considered giving a better rate to PHCS if PHCS excluded Advocate Lutheran General from its hospital network. Ballengee, Tr. 181-82. 279. Advocate Lutheran General was identified in contemporaneous PHCS and Great West correspondence to patients as an alternative to ENH. RX 712 at PHCS 891; Ballengee, Tr. 213- 24; Dorsey, Tr. 1479-80.

280. Managed care representatives testified that Advocate Lutheran General is a significant competitor to ENH. Ballengee, Tr. 211 (PHCS); Foucre, Tr. 941-42 (United); Neary, Tr. 630-31 (Great West); Holt-Darcy, Tr. 1597, in camera (“to some degree”) (Unicare); Spaeth, Tr. 2239-40, 2299 (Highland Park). (6) Rush North Shore 281. Rush North Shore is 3.7 miles or 9 minutes (driving time) southwest of Evanston Hospital. Spaeth, Tr. 2239-40; RX 1912 at 20-21, in camera.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 79 Initial Decision 282. Rush North Shore has 150 to 200 beds and as of February 2005 it was affiliated with Rush-Presbyterian-St. Luke’s, a major tertiary and academic hospital. The Rush-Presbyterian affiliation improved the breadth, quality, and the perception of services offered at Rush North Shore. Neaman, Tr. 1302.

283. Rush North Shore is geographically close to Evanston but does not have the same tertiary facilities as Advocate Lutheran General. Foucre, Tr. 945.

284. Rush North Shore had .12 residents per bed in 1999. RX 1912 at 60.

285. Rush North Shore was identified in contemporaneous PHCS correspondence to patients as an alternative to ENH. RX 712 at PHCS 891; Ballengee, Tr. 213-14.

286. Managed care representatives testified that Rush North Shore is a significant competitor to ENH. Ballengee, Tr. 211-12 (PHCS); Foucre, Tr. 941 (United); Spaeth, Tr. 2239-40, 2299 (Highland Park); Holt-Darcy, Tr. 1597, in camera (Unicare). (7) St. Francis 287. St. Francis is located in Evanston and is 3 miles south of Evanston Hospital on the same street -- Ridge Avenue, only an 8 minute drive past Evanston. Neaman, Tr. 1303; Foucre, Tr. 941; RX 1912 at 20-21, in camera.

288. St. Francis has 300 to 400 beds. As of February 2005, St. Francis was part of the Resurrection System. Neaman, Tr. 1303. St. Francis’s services range from cardiology and obstetrics to general surgery. RX 1854 at ENHE F16 426.

289. St. Francis is geographically close to Evanston but does not not have the same tertiary facilities that Advocate Lutheran General VOLUME 144 Initial Decision General has and has less of a perception as an equivalent facility. Foucre, Tr. 945.

290. St. Francis had .36 residents per bed in 1999. RX 1912 at 60.

291. St. Francis was identified in contemporaneous PHCS and Great West correspondence to patients as an alternative to ENH. RX 712 at PHCS 891; Ballengee, Tr. 213- 14. 292. Managed care representatives testified that St. Francis is a significant competitor to ENH. Ballengee, Tr. 212 (PHCS); Foucre, Tr. 942, 944-45 (United); Neary, Tr. 631 (Great West); Holt-Darcy, Tr. 1596, in camera (Unicare).

e. Hospitals Excluded from the Geographic Market (1) Condell 293. Condell is 12.7 miles and 24 minutes (driving time) northwest of Highland Park. Neaman, Tr. 1304-05; Hillebrand, Tr. 2006; Spaeth, Tr. 2239-40; Mendonsa, Tr. 555, in camera; RX 1912 at 20-21, in camera.

294. Condell is a 163 bed hospital located in Libertyville, Lake County, which is one of the fastest growing areas in metropolitan Chicago. Neaman, Tr. 1326; Hillebrand, Tr. 2006; Mendonsa, Tr. 562, in camera; RX 1912 at 60.

295. As of February 2005, Condell provided a full array of services, including everything from general obstetrics to cardiac surgery. Neaman, Tr. 1305.

296. Condell had no residents per bed in 1999. RX 1912 at 60. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 81 Initial Decision 297. Condell is not a significant competitor to ENH. Lake Forest, which is between Highland Park and Condell, is a more significant competitor to Highland Park. Holt-Darcy, Tr. 1596, in camera (Unicare) (Highland Park competes with Condell, “[l]ess so, because it is a little further west.”); Mendonsa, Tr. 562, in camera (Aetna) (Highland Park competes “[m]uch more with Lake Forest than Condell.”). But see Foucre, Tr. 944 (agreeing that Highland Park competes with Condell and Lake Forest) and Neary, Tr. 631 (Condell competes with ENH); Spaeth, Tr. 2239-40, 2299 (Highland Park).

(2) Resurrection 298. Resurrection Medical Center is 12.1 miles or 25 minutes (driving time) southwest of Evanston. Neaman, Tr. 1303-04; Ballengee, Tr. 263, in camera; RX 1912 at 20-21, in camera. 299. Resurrection has 350 staffed beds. RX 1912 at 60. 300. Resurrection had 0.17 residents per bed in 1999. RX 1912 at 60.

301. The Resurrection system is a large system, described by one managed care representative as a “system which we really need to keep.” Ballengee, Tr. 263, in camera. The Resurrection system includes St. Francis. Ballengee, Tr. 263, in camera. 302. There is conflicting testimony regarding whether the Resurrection system negotiated all of its hospitals as one contract or separately. Compare Ballengee, Tr. 263, in camera, with Foucre, Tr. 890-91.

303. Resurrection is not a significant competitor to ENH and was not identified by managed care organizations as an alternative to ENH. see F. 234-42.

(3) Other Hospitals VOLUME 144 Initial Decision 304. Noether testified that “certainly from a geographic perspective, some of the other hospitals that are quite near the sort of minimum geographic area that I’ve described certainly probably place at least competitive pressure and maybe potentially could even be in the market” including: Holy Family, Swedish Covenant, and the Vista hospitals. Noether, Tr. 5930-31. Noether also testified that Northwestern Memorial places “substantial competitive constraint” on ENH and the other hospitals in the proposed geographic market even though it is located in downtown Chicago. Noether, Tr. 5931 305. Holy Family is 11.3 miles or 23 minutes (driving time) west of Evanston. RX 1912 at 20-21, in camera. Holy Family has 260 staffed beds and .02 residents per bed in 1999. RX 1912 at 60. PHCS was the only managed care organization which mentioned Holy Family. RX 712 at PHCS 891; Ballengee, Tr. 213-14. 306. Swedish Covenant is an urban hospital located 6.8 miles or 19 minutes (driving time) south of Evanston. Neaman, Tr. 1305; Newton, Tr. 296; RX 1912 at 20-21, in camera. As of February 2005, Swedish Covenant had 325 beds. Newton, Tr. 296. In 1999, Swedish Covenant had .13 residents per bed. RX 1912 at 60. Swedish Covenant operates an open heart surgery program with Evanston. Newton, Tr. 423-24; Hillebrand, Tr. 2045-46. The managed care representatives did not mention Swedish Covenant as a significant competitor to ENH.

307. The Vista hospitals include Vista Health St. Therese and Vista Health Victory Memorial and are located in Waukegan in northern Illinois and Victory Memorial is “almost up to Wisconsin.” Wisconsin.” Dorsey, Tr. 1480; Noether, Tr. 5956. The Vista hospitals hospitals are an average of 15.9 miles or 30 minutes (driving time) time) north of Highland Park. RX 1912 at 20-21, in camera; Ballengee, Tr. 163. Great West was the only managed care EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 83 Initial Decision organization which mentioned the Vista hospitals as an alternative to to ENH. see Dorsey, Tr. 1479-80.

308. Northwestern Memorial is located in downtown Chicago, roughly 13 miles or 26 minutes (driving time) south of Evanston. Neaman, Tr. 1298; RX 1912 at 20-21, in camera. Northwestern Memorial is a tertiary hospital with more than 700 beds. Neaman, Tr. 1298. Northwestern Memorial is affiliated with the Northwestern Medical School and had .56 residents per bed in 1999. Neaman, Tr. 1299; RX 1912 at 60. Northwestern Memorial is the number one provider of obstetrical services in Illinois. Neaman, Tr. 1298. It has the premier obstetrics brand in Chicago because of its Prentice Women’s Hospital and possesses the largest volume of delivering mothers in the Chicago area. Hillebrand, Tr. 2003-04. Great West was the only managed care organization which mentioned Northwestern Memorial as an alternative to ENH. see Dorsey, Tr. 1479-80.

D. Effects on Competition 1. Anticompetitive Effects a. Market Concentration 309. Given the available data, Respondent’s expert, Noether, was not able to calculate exact market shares. Noether, Tr. 5961. Noether did, however, calculate proxy shares using the best available information, contained in the Medicare Cost Reports. Noether, Tr. 5961. The Medicare Cost Reports provide information on total net revenues, both inpatient and outpatient, across all managed care organizations. Noether, Tr. 5961. Noether calculated revenues of both inpatient and outpatient services and for inpatient services alone. Noether, Tr. 5961-62, 5964. 310. Noether also calculated Herfindahl-Hirschman Index (“HHI”) statistics. Noether, Tr. 5962. HHI is a measure suggested by by the Merger Guidelines as a way of capturing market VOLUME 144 Initial Decision concentration to take into account all of the players in the market, and it takes the shares of each of those firms, squares them, and then then sums the squared shares. Thus, HHI is a statistic that can range range from zero, in the case of a infinite number of very small players, up to 10,000, which is 100 squared, if there were a single monopolist in the market. Noether, Tr. 5962-63. 311. Noether properly treated St. Francis and Resurrection Medical Center as separate hospitals, although the hospitals had merged in the late 1990's. Noether, Tr. 5963; Noether, 6248-49, in camera; RX 531 at 13916.

312. Noether prepared a chart of net inpatient revenue and market shares (annualized) from 1997 to 2002 including the hospitals in her proposed geographic market. The net inpatient revenue from each hospital for each year was added to establish the market total. Each hospital’s revenue was divided by the market total to establish that hospital’s market share. Noether, Tr. 5962; RX 1912 at 58, in camera.

313. Noether calculated the HHI using 1999 market shares. RX 1912 at 58, in camera; Noether, Tr. 5965. 314. Noether calculated the post-merger HHI by summing the squares of the market shares of the hospitals in her proposed geographic market as follows: { } = 1919. see Noether, Tr. 5962- 65; RX 1912 at 58, in camera.

315. Noether calculated the change in HHI for her proposed market as 222. Noether, Tr. 5963; RX 1912 at 58, in camera. 316. Using the market shares from Noether’s proposed geographic market, but recalculated to reflect the Court’s defined geographic market allows a determination of the premerger HHI as follows: { } = 2355. see F. 323; RX 1912 at 58, in camera. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 85 Initial Decision 317. Using the market shares from Noether’s proposed geographic market, but recalculated to reflect the Court’s defined geographic market allows a determination that the combined market shares of Evanston and Highland Park in 1999 was { } see F. 322; RX 1912 at 58, in camera.

318. Using the market shares from Noether’s proposed geographic market, but recalculated to reflect the Court’s defined geographic market allows a determination of the post merger HHI as follows: { } = 2739. see F. 322; RX 1912 at 58, in camera. 319. Using the concentration figures in F. 316 and F. 318, the increase in the HHI is 384 (2739 minus 2355). 320. The post-merger HHI of over 2700 in the Court’s defined geographic market is well above the Merger Guidelines’ threshold of 1800 indicating a concentrated market (Noether, Tr. 5963) and the increase of over 350 far exceeds the Merger Guidelines’ threshold of 50 as signifying a significant increase in concentration. 321. To reflect the geographic market in this case, excluding Condell and Resurrection from Noether’s chart of net inpatient revenue and market shares (annualized) from 1997 to 2002 yields the following net inpatient revenues:

NET INPATIENT REVENUE 1997 1998 1999 2000 2001 2002 Evanston Northwest Healthcare { Highland Park Hospital Lake Forest Hospital Advocate VOLUME 144 Initial Decision Lutheran General Rush North Shore Medical Center Saint Francis Hospital MARKET TOTAL } See RX 1912 at 58, in camera.

322. To reflect the geographic market in this case, excluding Condell and Resurrection from Noether's chart of net inpatient revenue and market shares (annualized) from 1997 to 2002, provides the following market shares:

MARKET SHARES 1997 1998 1999 2000 2001 2002 Evanston Northwest Healthcare { Highland Park Hospital Lake Forest Hospital Advocate Lutheran General Rush North Shore Medical Center Saint Francis Hospital } See RX 1912 at 58, in camera.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 87 Initial Decision 323. Even the HHI of 1919 calculated by Noether using Respondent’s proposed geographic market exceeds the Merger Guidelines’ threshold of 1800 indicating a concentrated market and the increase of 222 exceeds the Merger Guidelines’ threshold of 50 as signifying a significant increase in concentration. see Noether, Tr. 5963; Merger Guidelines § 1.51.

324. Using the market shares from Noether’s proposed geographic market, but recalculated to reflect the Court’s defined geographic market, ENH increased its combined market share from approximately 35 to 40% from 1999 to 2002 while the market shares of the four competing hospitals in the geographic market fell from 1999 to 2002. F. 322.

325. In 1999, ENH identified the market share in its combined core service area as: Evanston, 44%; Highland Park, 11%; Lake Forest, 3%; Advocate Lutheran General, 7%; Rush North Shore, 14%; St. Francis, 7%; downtown teaching hospitals, 7%; and other, 7%. CX 84 at 21.

b. Contemporaneous and Post-Acquisition Evidence 326. The direct effects evidence of the ENH merger demonstrates that: (1) ENH achieved substantial price increases as a result of the merger; (2) empirical analysis establishes that ENH prices rose relative to other hospitals; and (3) alternative explanations of price increases are ruled out. F. 327-755. (1) ENH Achieved Substantial Price Increases as a Result of the Merger 327. The evidence demonstrates that: (a) Evanston and Highland Highland Park sought market power from the merger; (b) ENH sought to increase prices through contract negotiations and chargemaster increases; (c) managed care testimony confirms price VOLUME 144 Initial Decision price increases; and (d) ENH highlighted the managed care price increases as a merger accomplishment. F. 328-468. (a) Evanston and Highland Park Sought Market Power from the Merger 328. Present and former ENH executives testified that the contemporaneous assessment of the consequences of the merger found in ENH documents is an accurate reflection of contemporaneous discussions in the premerger and post-merger period. Neaman, Tr. 1192-95, 1196-97, 1200, 1203-05, 1207, 1209; Hillebrand, Tr. 1811-12; Spaeth, Tr. 2210-11; Newton, Tr. 369-70, 372-73.

329. ENH’s board meeting minutes were reviewed by key personnel, including Neaman, Evanston’s CEO, and accurately represented what occurred at the meetings. Attendees were free to speak candidly and honestly. Neaman, Tr. 1192-95. (i) Evanston 330. Evanston’s CEO, Neaman, acknowledged that one of Evanston’s goals of the merger with Highland Park was to obtain better prices and better terms on contracts from managed care organizations for ENH. Neaman, Tr. 1036. In the late 1990's, health plans were decreasing rates for hospital services. Neaman, Tr. 1037- 38. ENH and Highland Park hoped that the merged entity could strengthen the negotiating position of the hospitals with managed care organizations. Neaman, Tr. 1039; CX 19. 331. In 1998, Evanston’s CEO and Highland Park’s CEO wrote wrote about the business environment confronting Evanston and Highland Park, stating that: “[p]ricing pressures will escalate on healthcare providers from both government and managed care.” CX CX 19 at 1; Neaman, Tr. 1037-38. The recommendations included: included: “[s]trengthen negotiating positions with managed care EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 89 Initial Decision through merged entities and one voice” and “[m]maintain and enhance enhance local community ties for long-term success -- make indispensable to marketplace.” CX 19 at 1; see also CX 442 at 4-5; 4-5; CX 2 at 7 “geographic advantages” of merger). 332. At a January 4, 1999 meeting between Evanston and Highland Park board members and medical staff leaders, Evanston representatives identified the opportunity to “strengthen negotiation capability with managed care companies through merged entities” as well as to bring advanced oncology and cardiac surgery to Highland Park. CX 1 at 3 (physician groups should “‘not compete with self’”). 333. The minutes of an April 5, 1999 meeting record an Evanston representative as saying: “[g]rowth was seen as a real benefit to a possible merger. This would be an opportunity to join forces and grow together rather than compete with each other.” CX 2 at 7.

334. In a June 25, 1999 presentation about the proposed merger to Evanston’s board of directors, management reminded the board of the risk of “not undertaking [the] merger.” CX 84 at 58. Skokie Valley Community Hospital, located three miles to Evanston’s south, had been a “sleeping dog” competitor until it affiliated with the Rush system of hospitals, at which point Rush renamed it “Rush North Shore,” invested heavily in the hospital, and the former “sleeping dog” awoke to become a new, strong hospital. Hillebrand, Tr. 1794- 97. The point of the story was clear: if Evanston did not act first, the same problem could occur to Evanston’s north, and another hospital system would come in to further strengthen Highland Park. Hillebrand, Tr. 1797.

335. In a September 29, 1999 meeting, Neaman reported to ENH department heads that the addition of Highland Park helps Evanston to “[i]ncrease our leverage, limited as it might be, with the managed care players and help our negotiating posture.” CX 1566 at 9; Neaman, Tr. 1138, in camera.

VOLUME 144 Initial Decision 336. Neaman’s November 18, 1999 speech to the board of directors emphasized the same potential to increase leverage and enhance the negotiating posture with managed care players through the merged entity. RX 2015 at ENHL MO 3485. (ii) Highland Par 337. As early as the fall of 1998, Highland Park leadership “had been approached and approached again by [Evanston]” to discuss the possibility of a relationship between the two institutions. CX 3 at 1.

338. Transcript remarks from a fall of 1998 meeting of Highland Park leadership state that: “[n]obody is able to apply or assemble enough power to deal with managed care areas. An affiliation [with Evanston] would enable [Highland Park] to exploit an area of the market in a meaningful way -- Evanston has a large effect.” CX 3 at 1-2.

339. The three merging hospitals would form a triangle and “together would have a significant market penetration in these very affluent, attractive communities.” Newton, Tr. 352. 340. Highland Park management foresaw that a merger with Evanston would build “negotiating strength with payers.” CX 1869 at 7. Highland Park saw Evanston, Lake Forest, Northwest Community, and Condell as merger candidates, the attractiveness of each turning on “how concentrated could this market be for us.” CX 1869 at 6; Newton, Tr. 353-54. Merging with Evanston would build the greatest pricing strength with managed care organizations. Newton, Tr. 349-50.

341. In November 1998, Highland Park responded to Evanston’s Evanston’s merger proposal. CX 1879 at 1-2. With respect to “competition and signals,” Neele Steams, Highland Park’s board chairman, recognized that a merger would allow the two health care EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 91 Initial Decision care providers to “[s]top competing with each other.” CX 1879 at 3- 3-4.

342. In 1999, Highland Park board members and doctors met to frankly discuss the merger. During this meeting, Spaeth, the president of Highland Park, stated:

[T]he reality in my view is that we are not looking at a rosie future economically on this site. Neither are they [Evanston]. We are not looking at the opportunity to control this market individually. The largest ... payors in this arena have consolidated and are big enough, strong enough, and probably bent on assuring that the physicians who practice here and at Evanston and the institutions don’t make a hell of a lot of money. That is the reality and I am not even laying that on the insurers I am laying that on the employers. The same speech I have made over and over.

CX 4 at 1-2; Spaeth, Tr. 2210-11.

343. Spaeth continued by stating:

I think the ultimate benefit to these communities is pretty positive. There are cost economies, there are quality issues, there are ways to at least I think to push back on the managed care phenomenon and get the rates back where they ought to be if you are a big enough concerted enough entity which is important enough to the employers in this community. I think it would be real tough for any of the Fortune 40 companies in this area whose CEOs either use this place or that place to walk from Evanston, Highland Park, Glenbrook, and 1700 of their doctors. VOLUME 144 Initial Decision CX 4 at 2; Spaeth, Tr. 2210-11.

344. At that same meeting, another Highland Park representative expressed concerns regarding “the relative negotiating power of the payors,” which-had become an “economic issue” for the hospital. CX 4 at 9; Spaeth, Tr. 2211-12.

345. At that same meeting, there was a comment on “the economic benefit of not being out there doing battle with one another in what will be a common battle ground if you want to call it that.” CX 4 at 1.

346. Another Board member stated: “I’ll tell you can put in the bank now Dr. and that is that the Fortune 40 are gonna win they have the economic power and as long as we maintain the divided front on the provider side you’re gonna get hammered its just economics always work.” CX 4 at 11.

(b) ENH Sought to Increase Prices Through Contract Negotiation and Chargemaster Increases 347. The record shows that ENH exercised its market power, attained through the merger, to raise prices. At least six mechanisms were employed to raise prices: (1) utilizing the higher Evanston or Highland Park rate until new contracts were negotiated; (2) moving managed care organizations to one contract for all three hospitals; (3) in renegotiating contracts, demanding the higher of Evanston or Highland Park rates plus a premium and discount off rates; (4) increasing discount off charges arrangements; (5) adopting the higher of the Evanston or Highland Park chargemaster prices; and (6) increasing ENH’s chargemaster prices four times in 2002 and 2003. F. 348-391; see, e.g., CX 30 at 1, 3; CX 23 at 2; CX 26 at 1; CX 25 at 9; CX 31 at 1.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 93 Initial Decision (i) Higher of Evanston or Highland Park Rates Utilized Until New Contracts Negotiated 348. In a September 24, 1999 memorandum, Terry Chan, who was responsible for managed care contracting for Highland Park, compared Evanston and Highland Park inpatient rates, and stated that: “if the merged hospital and physician entities were successful in renegotiating hospital and physician contracts by January 1, 2000, with rates that are more favorable than the current Highland Park or ENH rates, (whichever is higher), there could be great potentials in improving payment rates for both hospitals and physicians.” CX 30 at 3.

349. In December 1999, ENH negotiators sent consent to assignment agreements to managed care organizations authorizing assignment of the higher of the Evanston or Highland Park rates. CX 5900 at 2-7; CX 5901 at 2; CX 5902 at 2, in camera. 350. In January 2000, while the status of many contracts was still in limbo, Chan instructed ENH’s billing department to “continue to use the current Highland Park Hospital rates” -- in the instances in which Highland Park had higher rates -- until all of the hospital contracts had been renegotiated. CX 5900 at 1; CX 1373 at 14, in camera.

351. Many managed care organizations that did not immediately consent to assign the higher of the two rates across all three hospitals later agreed during the negotiations with ENH. Ballengee, Tr. 174- 75; Neary, Tr. 763-64, in camera; CX 5900 at 1. 352. “Conver[ting] all payer contracts to the most favorable rates” of the two hospitals was an “Opportunity Ite[m]” for the merged entity that Ernst & Young projected could provide anywhere from $500,000 to $1,000,000 in possible revenue enhancements. CX 2386 at 2.

VOLUME 144 Initial Decision 353. In fact, as of March 2000, converting the payor contracts to the more favorable rates had exceeded ENH’s opportunity targets seven-fold. CX 2386 at 2. Ernst & Young’s March 2000 update showed that ENH had enhanced its revenue by $7 million dollars, a figure that was “ongoing.” CX 2386 at 2; see CX 2234 at 2. 354. One month later, in May 2000, Ernst & Young reported that converting the payor contracts to the more favorable of the Highland Park or Evanston contract had increased ENH’s revenue another $3 million dollars, for a total of $10 million in revenue enhancements that was “ongoing.” CX 23 at 2.

(ii) Managed Care Organizations Moved to One Contract for All Three Hospitals 355. ENH began managed care contract renegotiations on behalf of both Evanston and Highland Park in the fall of 1999 and continued to the fall of 2000. Chan, Tr. 833-34, in camera; Hillebrand, Tr. 1868-69, 1707.

356. Evanston engaged Bain for consulting advice at the time of the merger. Neaman, Tr. 1159. The focus of Bain’s 1999 merger consulting work for ENH was “growing net income by leveraging contracting and service line opportunities created by the Highland Park merger.” CX 74 at 3. Bain assisted ENH in creating a “unified contracting strategy reflecting the combined entities” of Highland Park and Evanston. CX 66 at 2.

357. Bain representatives themselves helped negotiate certain ENH managed care contracts in the renegotiations relating to the Highland Park merger. Neaman, Tr. 1217-18. Bain issued its final report on the merger project on February 1, 2000. CX 67 at 1. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 95 Initial Decision 358. During the winter of 1999, ENH senior management decided that the merged entity would put the three ENH facilities on the same contract and charge the same rate for all three facilities. Hillebrand, Tr. 1703-04; Newton, Tr. 363-65. 359. ENH demanded and received the same rate for all three facilities regardless of the level or complexity of services provided at each hospital. Foucre, Tr. 890; Ballengee, Tr. 176-77; Neary, Tr. 602; Neary, Tr. 756-60, in camera; Dorsey, Tr. 1447-50; CX 262 at 2, in camera.

360. Some managed care organizations opposed moving all three of the ENH facilities to the higher rates of the Evanston or the Highland Park contract because they did not value the three facilities equally. Neary, Tr. 603, 606; Holt-Darcy, Tr. 1560-61, in camera. 361. ENH presented an “all-or-nothing deal” to managed care organizations, regardless of complexity of services provided at each hospital. Holt-Darcy, Tr. 1528-29, in camera; Ballengee, Tr. 176-77; Neary, Tr. 602, Neary, Tr. 756, in camera; Dorsey, Tr. 1447-50; CX 262 at 2, in camera.

362. Under ENH’s billing system, managed care organizations “can’t distinguish between services at the three hospitals” to determine which services were rendered at a particular hospital in the system. Foucre, Tr. 890-92.

363. Evanston, Glenbrook, and Highland Park do not merit equal reimbursement rates. Dorsey, Tr. 1446-47. 364. ENH successfully moved all three ENH hospitals to the same contract and equalized the charges for all three sites postmerger. see, e.g., { }.

365. ENH’s request to move all three hospitals in its system to one set of rates was unusual for a hospital system in the Chicago VOLUME 144 Initial Decision area. Foucre, Tr. 890-92; see Ballengee, Tr. 163-65; Dorsey, Tr. 1445-46; RX 1503 at PHCS 3648, in camera; Holt-Darcy, Tr. 1528, 1528, in camera.

366. Other hospital systems in the Chicago area differentiate rates based upon the level and complexity of service offerings of each hospital in the system. Foucre, Tr. 890-92; Ballengee, Tr. 163- 65; Dorsey, Tr. 1446-47; RX 1503, in camera; see Holt-Darcy, Tr. 1528-30, in camera.

(iii)Higher of Evanston or Highland Park Rates Plus a Premium and Discount Off Rates Demanded 367. Recognizing ENH’s “additional negotiating power and leverage with the payors”– one of the “benefits of the merger” – during the winter of 1999, ENH senior management decided that “the combined entity would use the better of the Highland Park or Evanston [contract rate] and then add a premium to that.” Newton, Tr. 364-65; Hillebrand, Tr. 1705; Chan, Tr. 709-10. 368. Bain advised ENH to “sell” ENH’s benefits to managed care by: emphasizing “the value ENH brings to a payor’s network” such as brand, patient access, cost management, and quality, and to “[j]ustify premium pricing (i.e., above the competitive average).” CX 67 at 49.

369. The merged entity successfully negotiated prices above the premerger rates of either Evanston or Highland Park for numerous payors. Hillebrand, Tr. 1705.

370. Among ENH’s “accomplishments” were the renegotiations renegotiations of the United, PHCS, Aetna, Blue Cross Blue Shield, Shield, and Cigna contracts, which collectively resulted in an annualized economic value of $15 million for ENH ($3 million per per managed care organization). CX 17 at 5-8. ENH realized an EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 97 Initial Decision additional $3 million annually from the renegotiation of the Humana Humana contract and from the renegotiation of other smaller PPO contracts combined ($2 million for Humana and $1 million for some some “smaller” PPO contracts combined). CX 17 at 5, 8. 371. Evanston “had never achieved” a price increase as high as $18 million before the merger. Hillebrand, Tr. 1722. 372. Except for losing One Health for a short period of time, ENH lost no managed care organization customers over the course of the 2000 renegotiations. Hillebrand, Tr. 1707-08. (iv) Increased Discount Off Charges Arrangements 373. Post-merger, ENH succeeded with numerous managed care organizations in negotiating discount off charges arrangements, which were “more favorable” for ENH. CX 1373 at 14, in camera; RX 663 at ENHL TC 16939, in camera. Fixed rates tend to result in greater discounts – “up to 50%” – than discount off charges. Chan, Tr. 675.

374. As the Unicare representative explained, in discount off charges arrangements, the “hospital sets their own prices,” and managed care organizations “have no control over ... what the services are going to cost in any given admission or service.” Holt- Darcy, Tr. 1522-23, in camera.

375. Managed care organizations have no control over a hospital’s chargemaster increases. Neary, Tr. 609; Newton, Tr. 366; Holt-Darcy, Tr. 1522, in camera; Foucre, Tr. 898-900, 889; Mendonsa, Tr. 524-28, in camera. Under a discount off charges contract, the price that the managed care organization must pay to the hospital increases as the chargemaster list price increases, to the extent that the managed care organization does not negotiate a “ceiling,” such as a maximum or escalator clause. Porn, Tr. 5670. VOLUME 144 Initial Decision 376. The merged entity was successful in moving a number of managed care organizations to discount off charges arrangements. Hillebrand, Tr. 1706, 1893; Hillebrand, Tr. 1947, in camera; { } 377. A discount off charges arrangement would be even more favorable to the merged entity if “Highland Park Hospital is adopting ENH’s charge master which is expected to generate higher gross charges than gross charges generated by Highland Park Hospital’s current charge master.” RX 663 at ENHL TC 16939, in camera.

(v) Chargemaster Consolidation in 2000 to Higher of Evanston or Highland Park Charge 378. As part of the merger integration process, ENH consolidated the Highland Park and Evanston chargemasters in 2000. Hillebrand, Tr. 1710; Porn, Tr. 5643. 379. ENH created a combined chargemaster with the same rates for all three hospitals. Hillebrand, Tr. 1704; Porn, Tr. 5643. 380. In a “fairly simplistic analysis,” ENH examined the chargemasters at the two hospitals and adopted the higher of the Highland Park or Evanston chargemaster rates for each line item. Hillebrand, Tr. 1711, 1714-15; Noether, Tr. 6193; see CX 2240 at 11. 381. In January 2000, ENH’s transition team projected the overall increase in gross revenue from combining and increasing the charges at the three hospitals to be at least $100,000,000. CX 2237 at 1; CX 42 at 2; CX 2462 at 1. Later ENH documents estimated the overall increase in gross revenues at $100,000,000. CX 2238 at 1; CX 2239 at 1; CX 2384 at 2.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 99 Initial Decision 382. For example, upon completion of merging the chargemaster items related to renal dialysis, that transition team’s report reflected ENH’s objective: “[h]ighest charge comparing those of EH and HPH utilized on new Charge Master.” CX 2383 at 2. For renal dialysis alone, ENH’s finance department estimated a $1,324,497 “revenue enhancement” from selecting the higher of the Highland Park and Evanston rates. CX 2383 at 2.

383. As of September 30, 2000, only nine months after the merger, Neaman reported to ENH’s board of directors that ENH’s “Unified Pricing Structure” for the chargemaster had already resulted in $5 million of annualized economic value. CX 2382 at 6. (vi) Four Chargemaster Price Increases Instituted in 2002 and 2003 384. ENH increased its chargemaster rates four times between 2002 and 2003. RX 1687 at ENHL BW 27653, in camera. 385. On April 15, 2002, ENH implemented increases to its chargemaster. These changes were projected to { } CX 45 at 8. This increase had a { } impact on ENH’s fee schedule, depending on which estimate is used. CX 44 at 3; CX 45 at 8; RX 1687 at ENHL BW 27653, in camera.

386. After ENH raised its chargemaster prices in April 2002, Tom Hodges, ENH’s executive vice-president for finance, wrote to ENH managers that “[f]or a number of reasons we want to be as quiet as possible and there are relatively few people who have seen the scope of the changes.” CX 44 at 1.

387. According to Hillebrand, for chargemaster increases, “the only notification we make is to Blue Cross.” Hillebrand added, “[w]e should not notify anyone beyond those where we have a contractual obligation to do so.” CX 54 at 1. VOLUME 144 Initial Decision 388. On October 1, 2002, ENH raised prices for its three hospitals by { } RX 1687 at ENHL BW 27653, in camera. 389. On June 1, 2003, ENH raised prices for its three hospitals by { } RX 1687 at ENHL BW 27653, in camera. 390. On October 1, 2003, ENH raised prices for its three hospitals by { } RX 1687 at ENHL BW 27653, in camera. 391. From 2002 to 2003, ENH’s four chargemaster increases, taken together, represent a { } increase in the fee schedule. CX 44 at 3; CX 45 at 8; RX 1687 at ENHL BW 27653, in camera. (c) Managed Care Testimony Confirms Price Increases 392. Managed care representatives from United, PHCS, One Health (Great West), Aetna, and Unicare testified about their experiences negotiating contracts with the combined ENH entity. see F. 393-456.

(i) United 393. United, which was the second largest managed care organization in the Chicago area, had various contracts throughout the 1990's with both Evanston and Highland Park under the names of United affiliates including Share, Metlife, Metropolitan Life, Chicago HMO, Travelers, and MetraHealth. CX 5910 at 36-42; Hillebrand, Tr. 1868.

394. Before the merger in 2000, Highland Park and Evanston representatives formulated a strategy for the renegotiation of a contract with United. Hillebrand, Tr. 1873-74; Chan, Tr. 834, in camera.

395. Bain identified the United contract as a “1st Priority” contract with “upside revenue potential” for which the merged entity EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 101 Initial Decision entity had “enough leverage to improve terms.” CX 75 at 9-10; CX CX 74 at 10. Bain advised ENH that United reimbursed Evanston 45 45 to 50% less than it paid Highland Park. Hillebrand, Tr. 1869; RX RX 684 at BAIN 44, in camera. Moreover, Bain informed Evanston Evanston that its outdated contract with United had cost the hospital hospital $30 million over the preceding five years. Hillebrand, Tr. 1870; Neaman, Tr. 1340-41; RX 684 at BAIN 73, in camera; Haas- Haas-Wilson, Tr. 2851-52, in camera.

396. The negotiations resulted in { } Foucre, Tr. 890; CX 5174 at 11-12, in camera.

397. The United contract expired at the end of 2002. CX 5174 at 7. If neither party provided written notice of termination, then the contract renewed automatically for successive one-year terms. CX 5174 at 7. A separate provision of the contract allowed United to terminate the agreement at any time upon 90 days written notice if ENH’s standard charges increased by more than 6%. CX 5174 at 7. 398. In 2002, United stated that the merger had enabled ENH to “dominat[e] Chicago’s north shore, providing the only hospital locations ... ranging between Evanston and Highland Park, as well as a significant stretch of territory moving inland” and noting “the strategic importance of ENH’s geographic exclusivity.” CX 21 at 5. 399. In August 2002, United requested a renegotiation of United’s contract with ENH because, since the 2000 contract, ENH had been an “outlier” hospital with “much higher than the average reimbursement.” Foucre, Tr. 888.

400. United was concerned in part because the 2000 contract relied primarily on a discount off charges payment methodology, resulting in higher and higher reimbursements from United, which witnessed “alarmin[g] escalating costs in [ENH’s] billed charges” that were “outside of the norms for the market.” Foucre, Tr. 898, 889.

VOLUME 144 Initial Decision 401. In 2002, after exchanging proposals and counter-proposals a second time, United had made no progress towards achieving any of its business goals and considered terminating its existing contract with ENH. Foucre, Tr. 898-900.

402. United was also concerned that in 2002, “from quarter to quarter, the [chargemaster] increases were still occurring. It was not a one-time event.” Foucre, Tr. 1091, 1093, 1096, in camera; CX 2381 at 4, in camera; CX 6277 at 3, in camera. 403. { } Foucre, Tr. 1103-04, in camera. 404. Having had no success in lowering ENH’s prices, United pursued the more modest goal of asking ENH to stop increasing prices so much. Foucre, Tr. 906-09. { } CX 426 at 1, in camera. 405. The new contract between ENH and United was signed on April 14, 2004, with an effective date of June 1, 2004. Foucre, Tr. 887-88; CX 5176 at 1, 12.

406. { } Foucre, Tr. 1103, in camera.

407. { } Foucre, Tr. 1103-04, in camera. 408. Even today, with Lake Forest, Rush North Shore, St. Francis, and other neighboring hospitals in their network, United believes it cannot satisfy its customers without ENH. Foucre, Tr. 901-02, 925-26, 931-34.

(ii) PHCS 409. Prior to the merger, PHCS obtained competitive pricing from Evanston and Highland Park because PHCS “could choose EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 103 Initial Decision between the two and work them against each other.” Ballengee, Tr. Tr. 167.

410. On December 1, 1999, ENH notified PHCS of the impending merger and sought to assign Highland Park’s rates. CX 171 at 1. In response to that letter, PHCS wanted to renegotiate the rates. CX 1539 at 2; CX 172 at 1.

411. Bain advised ENH that it had “significant leverage in negotiations with PHCS as they have strong North Shore presence and need us in their network.” CX 1998 at 44. Bain indicated that Highland Park’s premerger terms with PHCS were significantly more favorable than Evanston’s terms. Hillebrand, Tr. 1892-93; RX 684 at BAIN 43, in camera.

412. ENH justified the request for an increase by indicating that it was one system which controlled the marketplace. Ballengee, Tr. 176-77, 194.

413. The “best scenario” for PHCS customers, strictly looking at dollars, was to eliminate ENH and redirect enrollees to the surrounding hospitals, such as Lake Forest, Advocate Lutheran General, and St. Francis. Ballengee, Tr. 244-48, in camera; CX 46 at 1, in camera.

414. PHCS believed, however, that customers did not want to “buy the network if they did not have [ENH in] it.” Ballengee, Tr. 181, 183-84.

415. PHCS states in contemporaneous documents that ENH’s proposal had a rate structure similar to Highland Park’s premerger contract and that PHCS’s goal was contract terms between Evanston and Highland Park’s previous terms. CX 115 at 1. 416. PHCS had previously eliminated the University of Chicago from its network and relied instead on the other teaching hospitals. Ballengee, Tr. 155.

VOLUME 144 Initial Decision 417. As an inducement to ENH, PHCS offered to exclude from its network hospitals like St. Francis, Rush North Shore, and Condell in return for lower prices. Ballengee, Tr. 178-79, 181-82. ENH declined the offer, except to offer a nominal discount for the exclusion of Advocate Lutheran General. Ballengee, Tr. 182; Hillebrand, Tr. 1746-47.

418. PHCS agreed to the { } CX 117 at 1, in camera; CX 5072 at 23, in camera; Ballengee, Tr. 252, 255, in camera; Hillebrand, Tr. 1893; CX 116 at 2, in camera. { } Ballengee, Tr. 258-61, in camera; CX 5072 at 23, in camera; CX 117 at 1, in camera. 419. PHCS negotiated more favorable terms than it had with Highland Park before the merger, although the rates were significantly higher than its premerger contract with Evanston. Ballengee, Tr. 175-76.

(iii) One Health (Great West) 420. Great West Healthcare (“Great West”) was formerly known as One Health. Neary, Tr. 581.

421. In December 1999, ENH contacted One Health to request the renegotiation of its hospital contract. Neary, Tr. 595. 422. Bain noted the “substantial difference” between One Health’s Highland Park and Evanston rates. CX 75 at 9-10; Neary, Tr. 604. Bain advised ENH to “[a]chieve [Highland Park] terms or better” in its negotiations with One Health. CX 1998 at 43. 423. Having last renegotiated the Highland Park and Evanston contracts in 1996 and 1995, respectively, One Health “agreed that it it had been several years since the contracts had been renegotiated and that it was appropriate to ... increase some of the rates.” Neary, EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 105 Initial Decision Neary, Tr. 608. One Health-was willing to give a price increase { } } Neary, Tr. 762-63, in camera; CX 2085 at 1-6, in camera. 424. In the first half of 2000, ENH and One Health did not reach an agreement on the renegotiation of the PPO and HMO contracts. Neary, Tr. 598, 609-10; Dorsey, Tr. 1438. One Health accepted ENH’s notice of termination. CX 266 at 1. 425. One Health’s contract with ENH terminated on August 31, 2000. Neary, Tr. 610-11; Hillebrand, Tr. 1707-08, 1898; CX 5062 at 1.

426. One Health made provisions for women “who were in the third trimester of pregnancy” at the time of the contract termination. Neary, Tr. 619-20. While One Health was able to negotiate a continuation of benefits for those expecting mothers, ENH charged the health plan rates that were higher than contract rates that were in place under the 1996 premerger One Health contract. Neary, Tr. 620, 637; CX 5063 at 1.

427. One Health customers complained about not having access to ENH, although One Health pointed to Lake Forest, Northwest Community, Advocate Lutheran General, Rush North Shore, and St. Francis as substitutes. Dorsey, Tr. 1451-52, 1459; Neary, Tr. 611, 617.

428. In the months following the termination of the ENH contract, One Health’s monthly membership reports began to reflect a “loss of membership within [the] network.” Dorsey, Tr. 1452, 1488; Neary, Tr. 617.

429. Before discussions between ENH and One Health resumed resumed in early October 2000, Great West received a written notice notice of termination, effective December 31, 2000, from Lake Forest and its medical group. RX 949; RX 950. Since Lake Forest was the primary alternative to Highland Park, it would have been “very problematic” for Great West to have lost Lake Forest from the VOLUME 144 Initial Decision the network at the same time Great West had no contract with ENH. ENH. Dorsey, Tr. 1484.

430. One Health returned to ENH prepared to accede “essentially regardless of what the ultimate price was.” Neary, Tr. 618-19; Dorsey, Tr. 1439-42.

431. One Health accepted a new agreement with an effective date of January 1, 2001. Dorsey, Tr. 1439-42; CX 5067 at 4; CX 266 at 1.

432. { } Hillebrand, Tr. 1947, in camera; compare CX 5067 at 15, in camera, CX 5059 at 17, and CX 5065 at 17. 433. { } Neary, Tr. 765-66, in camera; Hillebrand, Tr. 1944, in camera; CX 5064 at 17, in camera.

(iv) Aetna 434. Aetna “would have walked away” from Evanston if faced with a significant price increase before the merger. Mendonsa, Tr. 530, in camera. “[T]here probably would have been a walk-away point with the two independently. But with the two together, that was a different conversation.” Mendonsa, Tr. 520, in camera. 435. With the merger of “three extremely important hospitals negotiating together in a very important geography,” Aetna was “extremely concerned.” Mendonsa, Tr. 530, in camera. 436. Bain identified Highland Park’s rates for Aetna’s PPO and POS products as higher than Evanston’s rates for those products. RX 762 at ENHL TC 9936, in camera. Evanston’s contract with Aetna was nearly four years old in November 1999, so Bain recommended renegotiation of the Aetna contract as a priority. CX 75 at 10; CX 5001 at 2.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 107 Initial Decision 437. Aetna had not renegotiated its contract with Evanston since 1996 and expected ENH to make a proposal to renegotiate. Based on the 3% increase per year in Medical CPI between 1996 and 1999, Aetna calculated an appropriate increase compounded over three years to be { } Mendonsa, Tr. 533-34, in camera. 438. During the 2000 negotiations, ENH originally sought a discount off charges arrangement for PPO and POS plans. Hillebrand, Tr. 1896; RX 769 at ENH JL 2818-19, in camera. Aetna, however, did not agree to that payment methodology. Hillebrand, Tr. 1896.

439. ENH and Aetna ultimately agreed { } CX 5008 at 5-6, in camera; Hillebrand, Tr. 1896.

440. { } RX 855 at ENHL BW 11393, in camera; CX 5007 at 5. { } CX 5008 at 7, in camera.

441. Aetna agreed { } Mendonsa, Tr. 539, in camera; Hillebrand, Tr. 1948, in camera; CX 2447 at 1, in camera. 442. Aetna’s increased rates under the post-merger contract with ENH became effective June 1, 2000. CX 5008 at 1. 443. { } Mendonsa, Tr. 561, 573, in camera. 444. { } Mendonsa, Tr. 544, 568-69, in camera. 445. { } Mendonsa, Tr. 517-18, 530, in camera. 446. Aetna believed it “couldn’t walk away” from post-merger ENH because it would have “devastated us,” “killed our marketing,” and “shut down” Aetna’s marketing to local employers. Mendonsa, Tr. 518, 520, 530, in camera.

(v) Unicare VOLUME 144 Initial Decision 447. In 2000, Unicare acquired Rush Prudential, another managed care organization. Holt-Darcy, Tr. 1413. Prior to the merger, Rush Prudential had contracted with both Evanston and Highland Park, and Unicare had contracted with Evanston. Holt- Darcy, Tr. 1505-06, in camera.

448. { } Holt-Darcy, Tr. 1549-50, 1598, 1599-1601, in camera; CX 216 at 1; CX 5085 at 1; CX 5091 at 1. 449. With the merger, ENH proposed an unusual “all-or-nothing deal” in which there would be one rate for all three hospitals, regardless of the level of services at each facility – like the “Three Musketeers, all for one and one for all.” Holt-Darcy, Tr. 1529, in camera.

450. { } CX 215 at 1; CX 216 at 15, in camera; CX 5076 at 10; CX 5085 at 1; CX 5091 at 1. { } CX 124 at 2-3, in camera. { } Holt-Darcy, Tr. 1570-72, in camera.

451. Even if Unicare representatives had expected an increase in ENH contract rates after the merger – which they did not – the rates proposed by ENH in 2000 were above what Unicare considered to be a “reasonable” increase, { } Holt-Darcy, Tr. 1503-04, in camera. { } Holt-Darcy, Tr. 1504 in camera.

452. The result for Unicare { } Holt-Darcy, Tr. 1537, 1541, 1564, in camera.

453. { } Holt-Darcy, Tr. 1543, in camera. { } Holt-Darcy, Tr. 1542-43, in camera.

454. { } CX 5075 at 17-18, in camera; Holt-Darcy, Tr. 1582, in camera.

455. According to Unicare, ENH had indicated that it could obtain higher prices because it had “a lot more leverage now that EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 109 Initial Decision they have three hospitals in their service area” and ENH had a “stronger presence” in the area, meaning ENH had “basically sewn sewn up the North Shore geography.” Holt-Darcy, Tr. 1546, 1559- 1559-60, in camera; CX 129 at 1, in camera. 456. Unicare would be in a bind without ENH, now a “key provider” in the North Shore. Holt-Darcy, Tr. 1552-53, in camera. ENH’s “contiguous service area” made it “hard, painful, for customers to see [ENH] leave the network.” Holt-Darcy, Tr. 1603, in camera.

(d) ENH Highlighted the Managed Care Price Increases as a Merger Accomplishment 457. In his January 6, 2000 update to the ENH executive committee, Hillebrand reported that “as a result of combining the medical staffs and Hospitals of the merger, [ENH] was able to renegotiate a managed care contract that resulted in an additional $3.5 million benefit” and that “other managed care contracts will be renegotiated over the next 100 days.” CX 5 at 5; Newton, Tr. 369- 70.

458. The February 3, 2000 ENH board meeting minutes state: “Hillebrand commented on the recent re-renegotiation of managed care contracts and the 'added value’ as a result of combining the medical staffs and hospitals. Other managed care contracts are in the process of being re-negotiated.” CX 6 at 7. 459. On March 14, 2000, Hillebrand drafted ENH’s 2001-2003 Strategic Plan. In the draft of the Strategic Plan, Hillebrand stated: Through our growth initiatives, we will expand our presence in our marketplace in order to provide leverage to our market position as we negotiate relationships with the purchasers of care. Our goal VOLUME 144 Initial Decision will be to receive superior pricing for our services and to become indispensable to the purchaser of care care as they sell their product in our marketplace. CX 2070 at 3.

460. The June 16, 2000 Highland Park health care services committee meeting minutes state:

Neaman reviewed the list of merger accomplishments. Important successes have been accomplished in managed care contracting. There has been a $12 million improvement on the Hospital side and $8 million to physicians’ practices to date. The total improvements as a result of the merger are $29.5 million, which greatly exceeds the Board approved $19 million goal over three years. CX 12 at 2.

461. On July 3, 2000, Neaman issued a memorandum with the subject “July 4, 2000 -- Interdependence Day” which summarized the first six months since the merger. In the memorandum, Neaman stated:

The major economic accomplishments in June were the successful re-negotiation of two of our HMO agreements..., that will collectively produce some $6 million of additional revenues on an annualized basis. This brings the total managed care renegotiation benefits to some $16 million/year to the Institution. This figure does not include some $10 million+ additional managed care monies going to our physicians.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 111 Initial Decision CX 13 at 1; CX 12 at 2; Neaman, Tr. 1200. 462. In the July 3, 2000 “Interdependence” memorandum, Neaman stated:

As we begin the July 4th holiday, it is safe to say that our success in the merger integration effort is not a product of our “independence,” but of our “interdependence.” Neither Evanston nor Highland Park alone could achieve these results. Our three Hospitals, together with our 1500 physicians as a “fighting unit,” appear to have helped provide at least a small advantage for an interim period. CX 13 at 1.

463. At a September 27, 2000 meeting of the ENH board’s finance committee, Neaman emphasized the link between the merger and the managed care renegotiations. Neaman stated that “the larger market share created by adding Highland Park Hospital has translated to better managed care contracts.” CX 16 at 1. 464. Neaman’s October 2, 2000 “Final Report – Merger Integration Activities” memorandum to the ENH board reported that: “Some $24 million of revenue enhancements have been achieved – mostly via managed care renegotiations. (This figure does not include some $13 million of additional managed care revenues to participating physicians.) Our net income from operations will go from a budgeted $4 million to in excess of $20 million for Fiscal Year 2000.” CX 17 at 1. In addition, “[s]ome $12 million of cost improvements have been achieved -- mostly from corporate overhead areas.” CX 17 at 1.

465. Neaman’s October 2, 2000 Report reiterated: “As stated previously, none of this could have been achieved by either Evanston or Highland Park alone. The ‘fighting unit’ of our three VOLUME 144 Initial Decision hospitals and 1600 physicians was instrumental in achieving these ends.” CX 17 at 2.

466. None of the initial post-merger price increases obtained by ENH from health plans were reduced in subsequent years, with the exception of a { } Hillebrand, Tr. 1709-10, 1725-26; Neaman, Tr. 960-61, 1269-71.

467. Highland Park could not have raised its prices to health plans absent the merger. According to Chan, all the rates Highland Park had in place in July 1, 1999, were the best that Highland Park could accomplish at that time without threatening termination. Chan, Tr. 820, in camera; CX 1099, in camera.

468. Spaeth also testified that at the time of the merger Highland Park would not have been successful in raising its rates because the hospital could not sustain a strategy where it kept losing contracts. Spaeth, Tr. 2178-79. Spaeth did not see an opportunity to raise the rates before the merger. Spaeth, Tr. 2172-73. (2) Empirical Analysis Establishes that ENH Prices Rose Relative to Other Hospitals (a) Introduction to the Data and Methodology 469. Complaint Counsel’s economic expert, Dr. Deborah Haas- Haas-Wilson, Professor of Economics at Smith College, used four different data sources in her empirical analysis to examine whether whether prices increased at ENH after the merger. The four data sources were: (1) managed care claims data; (2) data from the Universal Dataset from the Illinois Department of Public Health (“IDPH Universal Dataset”); (3) data from the economic consulting consulting firm NERA, submitted to the FTC on behalf of ENH; and EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 113 Initial Decision and (4) data submitted directly by ENH in response to an FTC Civil Civil Investigative Demand (“CID”). Haas-Wilson, Tr. 2495-500. 470. As Noether indicated, however, “there were a number of problems with the data that made the measure of price certainly less than fully accurate.” Noether, Tr. 6051, in camera. 471. Noether concluded that analysis of the claims data could be used in “forming [her] opinion and reaching [her] conclusions,” but should be considered “in the context of all the other evidence in the case.” Noether, Tr. 6052, in camera.

472. Haas-Wilson also noted the strengths and weaknesses of the four data sources and indicated that she had to “process the data to get it into a form that you can actually use for research.” Haas- Wilson, Tr. 2496-500.

473. Haas-Wilson found that, regardless of the data source that is used or the methodology used to “clean” or manipulate the data, all the evidence shows that following the merger with Highland Park, ENH raised the prices of inpatient acute care hospital services to managed care organizations. Haas-Wilson, Tr. 2500-01. 474. While all experts agree that ENH experienced relative price increases in the 2000 time frame, Respondent’s economic expert, Dr. Jonathan B. Baker, Professor of Law at American University and Senior Consultant, Charles River Associates, contends that the relative price increases were smaller than those calculated by Haas- Wilson. Baker, Tr. 4617-20, 4646, 4795-96, in camera; Haas- Wilson, Tr. 2637, in camera.

475. Haas-Wilson, further, concluded that the merger eliminated the competition between the two competitors by excluding an alternative provider available to managed care organizations. Haas- Wilson, Tr. 2472-73.

VOLUME 144 Initial Decision (i) Relative Price Changes, Not Relative Prices, Is the Appropriate Methodology to Test for Market Power 476. Hospital services are a differentiated product. Haas-Wilson, Tr. 2492-93; Noether, Tr. 5910. Consumers are willing and able to pay higher prices for certain aspects of product differentiation. Because prices can vary in the market for a differentiated service for many different reasons, one may not conclude anything about market power by merely using a cross-sectional analysis of hospital prices at a single point in time. Haas-Wilson, Tr. 2492-93. 477. In contrast, by looking at price changes over time, one can compare the price change at one hospital to the price change at another hospital. Using such an approach, one can conclude that there is a change in market power if there is a price increase after having ruled out the other possible explanations for greater price increases at one hospital versus another. Haas-Wilson, Tr. 2495. 478. Whether ENH’s prices increased faster than other hospitals is determined by using a methodology called difference in differences. The first step in the difference in differences analysis is to calculate the difference in price at ENH by subtracting the premerger price at ENH from the post-merger price at ENH, and calling that the “ENH difference.” Haas-Wilson, Tr. 2546-47, in camera.

479. The second step in the difference in differences analysis is to repeat the process for the comparison hospitals. The difference for the comparison hospitals is called the “control group difference.” Haas-Wilson, Tr. 2546-47, in camera.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 115 Initial Decision 480. The third step in the difference in differences analysis is to take each difference as the percent of the premerger price, and then examine whether or not the ENH post-merger percentage increase in price is the same or different than the control group post-merger percentage increase in price. Haas-Wilson, Tr. 2546-48, in camera. (ii) Control Groups 481. Haas-Wilson used three control groups: (1) all general acute care hospitals in the Chicago Primary Metropolitan Statistical Area (“PMSA”) (the “Chicago PMSA Hospitals” control group); (2) all general acute care hospitals in the Chicago PMSA, that were not involved with a merger with another hospital between 1996 and 2002 (the “Non-Merging Chicago PMSA Hospitals” control group); and (3) all the general acute care hospitals in the Chicago PMSA that were involved in some teaching activity during the study period (the “Chicago PMSA Teaching Hospitals” control group). Haas-Wilson, Tr. 2548-49, in camera.

482. Using multiple control groups provides a “specifications test,” so that if one finds similar results using multiple control groups, that gives one increased confidence in the results. Haas- Wilson, Tr. 2549, in camera.

483. It is important that the hospitals in the control groups experience similar changes in cost, regulation, and demand. Haas- Wilson, Tr. 2548, in camera.

484. The first control group, the Chicago PMSA Hospitals control group, was chosen because those hospitals should be subject to similar changes in costs, demand, and regulation as ENH. Haas- Wilson, Tr. 2549, in camera.

485. The second control group, the Non-Merging Chicago PMSA Hospitals control group, was selected because theory and VOLUME 144 Initial Decision empirical work suggest that cost and pricing might be different at hospitals involved with mergers versus those that are not involved with mergers. Haas-Wilson, Tr. 2549-50, in camera. 486. The third control group, the Chicago PMSA Teaching Hospitals control group, was selected because empirical literature suggests that costs and therefore prices might be different at hospitals that are engaged in teaching activity versus those that are not. Haas-Wilson, Tr. 2550, in camera. The “Teaching Hospital” control group ended up including nearly fifty hospitals, half of the hospitals in the Chicago PMSA. Noether, Tr. 6110-11, in camera. 487. Haas-Wilson rejected the concept of picking hospitals that “looked like” Evanston to use as her control group, because this would have required making arbitrary decisions on which neither theory nor previous empirical work provided guidance. Haas- Wilson, Tr. 2550-51, in camera.

488. Any attempt to match hospitals with ENH to form a control group that “looked like” ENH would have to account for the fact that Evanston and Highland Park had different characteristics premerger. A control group that looked like Evanston may not be the appropriate control group to compare post-merger Evanston and Highland Park. Haas-Wilson, Tr. 2550-51, in camera. 489. Haas-Wilson’s results were statistically significant. The term “statistically significant” is a term from statistics and econometrics that indicates how much confidence one has in the results of one’s hypothesis test or how much confidence one has in the conclusions one makes based on those results. Haas-Wilson, Tr. 2553, in camera.

490. Statistical significance is expressed as levels of significance. One discusses the 1% level or a 5% level or a 10% EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 117 Initial Decision level, where a 1% level would be the highest level of significance. A A 5% or 10% level are also quite high levels of significance, but not not as high as a 1% level of significance. Haas-Wilson, Tr. 2553-54, 2553-54, in camera.

(b) Claims Data Submitted by Managed Care Organizations 491. Although seven managed care organizations produced claims data, the data was only usable for four managed care organizations: United, Blue Cross Blue Shield, Aetna, and Humana. Haas-Wilson, Tr. 2510, in camera; Noether, Tr. 6049-50, 6094, 6074, 6055, 6069, in camera.

492. The managed care claims data was collected not for research, but to enable managed care insurers to pay hospitals. Therefore, the data had to be processed into a usable form. Haas- Wilson, Tr. 2497-99; see also Noether, Tr. 6052-53, in camera (data came in a disaggregate fashion).

493. In addition, Haas-Wilson analyzed data from One Health. However, she admitted that this data “does not allow me to look at the total reimbursement to the hospital for inpatient care. It includes only the amount paid to the hospital by the insurance company. It does not include any individual consumer co-pay.” Haas-Wilson, Tr. 2576, in camera. Thus, the data could not be compared to the data provided by the other managed care organizations. Haas-Wilson, Tr. 2576-77, in camera.

494. The claims data received from One Health does not contain any pre-2000 data points. Noether, Tr. 6050, in camera. Haas-Wilson did not testify regarding what time period she used for the premerger period for One Health. Haas-Wilson, Tr. 2511-12, in camera (discussing DX 7010). Thus, it is not clear what time period was used by Haas-Wilson to perform her analysis and the One Health data is found to be unreliable.

VOLUME 144 Initial Decision 495. Haas-Wilson analyzed the managed care claims data by type of plan within payor. According to economic theory and institutional relationships, there was more potential for price increases at some types of plans relative to other types of plans. In particular, when a plan has a more narrow network (including fewer hospitals) that gives the managed care organization a better bargaining position, because there are fewer hospitals in the network and it is easier to exclude hospitals from the network. Haas-Wilson, Tr. 2510-11, in camera.

496. Haas-Wilson acknowledged that managed care organizations negotiate trade-offs pertaining to the various plans – e.g., a lower price for the HMO plan in return for a higher price for the PPO plan, and vice versa. Haas-Wilson, Tr. 2853, in camera; Mendonsa, Tr. 557, in camera; Holt-Darcy, Tr. 1541, 1586-87, in camera; Hillebrand, Tr. 1861-62, 2019; RX 844 at ENH JL 2023, in camera.

497. The premerger time period varied with each payor, because it was calculated from the beginning of 1998 through the contract effective date (“CED”) of the first contract negotiated by ENH with that payor after the merger. Haas-Wilson, Tr. 2511, in camera. 498. Haas-Wilson concluded that “for most payers and plans there were large post-merger price increases at ENH.” Haas-Wilson, Tr. 2518, 2524-25, in camera.

(i) United 499. The premerger period for United is from the { } The postmerger period for United is { } Haas-Wilson, Tr. 2511-12, in camera.

500. The United data had some limitations including only a sparse number of cases premerger. Baker, Tr. 4621-22, in camera. In EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 119 Initial Decision In addition, there were more mothers than newborns in the obstetrics obstetrics claims data, which was about 40% of the claims (the “missing baby” problem). Haas-Wilson used the data as provided while Noether added in babies to make up for the “missing baby” problem. Baker, Tr. 4625-26, 4628, 4806-07, in camera; Noether, Tr. Tr. 6053-55, in camera. Professor Baker could not fully correct the the obstetrics problem, so he performed his analysis two ways, including obstetrics and excluding obstetrics. Baker, Tr. 4628, in camera.

501. Haas-Wilson calculated the post-merger increase in inpatient price per day and per case. She then compared these results to the three control groups. Haas-Wilson, Tr. 2557-59, 61-62, in camera; CX 6279 at 3, 8-9, in camera.

502. The results are statistically significant unless otherwise noted. CX 6279 at 8-9, 19, in camera.

(aa) MO/HMO+ 503. For United HMO/HMO+ patients, the post-merger increase in inpatient price per day was { } This means that, according to United’s data, the average price per day across United HMO/HMO+ patients at ENH in the post-merger period was { } more than the average price per day across United HMO/HMO+ patients at Evanston in the premerger period. Haas-Wilson, Tr. 2516, in camera; CX 6279 at 3, in camera.

504. For United HMO/HMO+ patients, the post-merger increase in inpatient price per case was { } Haas-Wilson, Tr. 2516, in camera; CX 6279 at 3, in camera.

505. For United’s HMO/HMO+ plan, the price increase at ENH ENH in the price per day was { } greater than the average price price increase across all Chicago PMSA Hospitals. see CX 6279 at at 8, in camera. For United’s HMO/HMO+ plan, the price increase increase at ENH in the price per case was { } greater than the VOLUME 144 Initial Decision average price increase across all Chicago PMSA Hospitals. see CX CX 6279 at 9, in camera.

506. For United’s HMO/HMO+ plan, the price increase at ENH in the price per day was { } greater than the average price increase across Non-Merging Chicago PMSA Hospitals. see CX 6279 at 8, in camera. For United’s HMO/HMO+ plan, the price increase at ENH in the price per case was { } greater than the average price increase across Non-Merging Chicago PMSA Hospitals. see CX 6279 at 9, in camera.

507. For United’s HMO/HMO+ plan, the price increase at ENH in the price per day was { } greater than the average price increase across Chicago PMSA Teaching Hospitals. see CX 6279 at 8, in camera. For United’s HMO/HMO+ plan, the price increase at ENH in the price per case was { } greater than the average price increase across Chicago PMSA Teaching Hospitals. see CX 6279 at 9, in camera.

(bb) POS/EPO 508. For United POS/EPO (exclusive provider organizations) patients, the post-merger increase in inpatient price per day was { } Haas-Wilson, Tr. 2516, in camera; CX 6279 at 3, in camera. 509. For United POS/EPO patients, the post-merger increase in inpatient price per case was { } Haas-Wilson, Tr. 2516, in camera; CX 6279 at 3, in camera.

510. For United’s POS/EPO plan, the price increase at ENH in the price per day was { } greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 8, in camera. For United’s POS/EPO plan, the price increase at ENH in the price per case was { } greater than the average price increase across all Chicago PMSA Hospitals. see CX 6279 at 9, in camera. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 121 Initial Decision 511. For United’s POS/EPO plan, the price increase at ENH in the price per day was { } greater than the average price increase across Non-Merging Chicago PMSA Hospitals. See CX 6279 at 8, in camera. For United’s POS/EPO plan, the price increase at ENH in the price per case was { } greater than the average price increase across Non-Merging Chicago PMSA Hospitals. See CX 6279 at 9, in camera.

512. For United’s POS/EPO plan, the price increase at ENH in the price per day was { } greater than the average price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 8, in camera. For United’s POS/EPO plan, the price increase at ENH in the price per case was { } greater than the average price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 9, in camera.

(cc) PPO/Indemnity 513. For United PPO/Indemnity patients, the post-merger increase in inpatient price per day was { } Haas-Wilson, Tr. 2516- 17, in camera; CX 6279 at 3, in camera.

514. For United PPO/Indemnity patients, the post-merger increase in inpatient price per case was { } Haas-Wilson, Tr. 2516- 17, in camera; CX 6279 at 3, in camera.

515. For United’s PPO/Indemnity plan, the price increase at ENH in the price per day was { } greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 8, in camera. For United’s PPO/Indemnity plan, the price increase at ENH in the price per case was { } greater than the average price increase across all Chicago PMSA Hospitals. See Haas-Wilson, Tr. 2558-59, in camera; CX 6279 at 9, in camera. 516. For United’s PPO/Indemnity plan, the price increase at ENH in the price per day was { } greater than the average price VOLUME 144 Initial Decision price increase across Non-Merging Chicago PMSA Hospitals. See CX 6279 at 8, in camera. For United’s PPO/Indemnity plan, the price increase at ENH in the price per case was { } greater than the the average price increase across Non-Merging Chicago PMSA Hospitals. See Haas-Wilson, Tr. 2561-62, in camera; CX 6279 at 9,in camera.

517. For United’s PPO/Indemnity plan, the price increase at ENH in the price per day was { } greater than the average price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 8, in camera. For United’s PPO/Indemnity plan, the price increase at ENH in the price per case was { } greater than the average price increase across Chicago PMSA Teaching Hospitals. See Haas- Wilson, Tr. 2561-62, in camera; CX 6279 at 9, in camera. (dd) Summary 518. With respect to the United data, Haas-Wilson concluded from her regression analysis that the price increases at ENH were larger than the price increases at comparison hospitals, and that was true no matter how she measured resource intensity or which comparison group she used. Haas-Wilson, Tr. 2626-28, in camera; CX 6279 at 19, in camera.

519. For United, since the regression results take into account variations in patient mix, customer mix, and teaching intensity across hospitals over time, changes in these variables cannot explain all of the relatively larger price increases at ENH in the post-merger period compared to control group hospitals. Haas-Wilson, Tr. 2627- 28, in camera; CX 6279 at 19, in camera. 520. For United, using the control group of all Chicago PMSA Hospitals, and taking into account changes in patient mix, customer customer mix, and teaching intensity, the post-merger price increases EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 123 Initial Decision increases at ENH were { } greater than at the average control group group hospital. See CX 6279 at 19, in camera. 521. For United, using the control group of Non-Merging Chicago PMSA Hospitals, and taking into account changes in patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were { } greater than at the average control group hospital. See CX 6279 at 19, in camera.

522. For United, using the control group of Chicago PMSA Teaching Hospitals, and taking into account changes in patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were { } greater than at the average control group hospital. See CX 6279 at 19, in camera.

(ii) Aetna 523. The premerger period for Aetna is from { } The postmerger period for Aetna is from { } Haas-Wilson, Tr. 2512, in camera.

524. The results are statistically significant unless otherwise noted. CX 6279 at 8-9, 19, in camera.

(aa) HMO 525. For Aetna HMO patients, the post-merger increase in inpatient price per day was { } CX 6279 at 3, in camera. 526. For Aetna HMO patients, the post-merger increase in inpatient price per case was { } CX 6279 at 3, in camera. 527. For Aetna’s HMO plan, the price increase at ENH in the price per day was { } greater than the average price increase across across all Chicago PMSA Hospitals. See CX 6279 at 8, in camera. For Aetna’s HMO plan, the price increase at ENH in the price per VOLUME 144 Initial Decision case was { } greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 9, in camera. 528. For Aetna’s HMO plan, the price increase at ENH in the price per day was { } greater than the average price increase across Non-Merging Chicago PMSA Hospitals. See CX 6279 at 8, in camera. For Aetna’s HMO plan, the price increase at ENH in the price per case was { } greater than the average price increase across Non-Merging Chicago PMSA Hospitals. See CX 6279 at 9, in camera.

529. For Aetna’s HMO plan, the price increase at ENH in the price per day was { } greater than the average price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 8, in camera. For Aetna’s HMO plan, the price increase at ENH in the price per case was { } greater than the average price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 9, in camera. (bb) PPO 530. For Aetna PPO patients, the post-merger increase in inpatient price per day was { } CX 6279 at 3, in camera. 531. For Aetna PPO patients, the post-merger increase in inpatient price per case was { } CX 6279 at 3, in camera. 532. For Aetna’s PPO plan, the price increase at ENH in the price per day was { } greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 8, in camera. For Aetna’s PPO plan, the price increase at ENH in the price per case was { } greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 9, in camera. 533. For Aetna’s PPO plan, the price increase at ENH in the price per day was { } greater than the average price increase across EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 125 Initial Decision across Non-Merging Chicago PMSA Hospitals. This result is not statistically significant. See CX 6279 at 8, in camera. For Aetna’s PPO plan, the price increase at ENH in the price per case was { }greater than the average price increase across Non-Merging Chicago PMSA Hospitals. See CX 6279 at 9, in camera. 534. For Aetna’s PPO plan, the price increase at ENH in the price per day was { } greater than the average price increase across Chicago PMSA Teaching Hospitals. This result is not statistically significant. See CX 6279 at 8, in camera. For Aetna’s PPO plan, the price increase at ENH in the price per case was { } greater than the average price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 9, in camera.

(cc) Summary 535. For Aetna, using the control group of all Chicago PMSA Hospitals, and taking into account changes in patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were { } greater than at the average control group hospital. See CX 6279 at 18, in camera.

536. For Aetna, using the control group of Non-Merging Chicago PMSA Hospitals, and taking into account changes in patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were { } greater than at the average control group hospital. See CX 6279 at 18, in camera.

537. For Aetna, using the control group of Chicago PMSA Teaching Hospitals, and taking into account changes in patient mix, customer mix and teaching intensity, the post-merger price increases at ENH were { } greater than at the average control group hospital. See CX 6279 at 18, in camera.

VOLUME 144 Initial Decision (iii) Humana 538. The premerger period for Humana is from { } The postmerger period for Humana is from { } Haas-Wilson, Tr. 2511-12, in camera.

539. Haas-Wilson excluded payments under capitated plans from her analysis. Haas-Wilson, Tr. 2853, in camera; Noether, Tr. 6076- 77, in camera. { } Noether, Tr. 6076, in camera. 540. These results are statistically significant unless otherwise noted. CX 6279 at 8-9, 19,in camera.

(aa) ASO 541. For Humana ASO (administrative services only) patients, the post-merger percentage increase in inpatient price per day was { } CX 6279 at 3, in camera.

542. For Humana ASO patients, the post-merger increase in inpatient price per case was { } CX 6279 at 3, in camera. 543. For Humana’s ASO plan, the price increase at ENH in the price per day was { } greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 8, in camera. For Humana’s ASO plan, the price increase at ENH in the price per case was { } greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 9, in camera. 544. For Humana’s ASO plan, the price increase at ENH in the the price per day was { } greater than the average price increase increase across Non-Merging Chicago PMSA Hospitals. See CX 6279 at 8, in camera. For Humana’s ASO plan, the price increase at at ENH in the price per case was { } greater than the average price EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 127 Initial Decision price increase across Non-Merging Chicago PMSA Hospitals. See CX 6279 at 9, in camera.

545. For Humana’s ASO plan, the price increase at ENH in the price per day was { } greater than the average price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 8, in camera. For Humana’s ASO plan, the price increase at ENH in the price per case was { } greater than the average price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 9, in camera. (bb) HMO 546. For Humana HMO patients, the post-merger increase in inpatient price per day was { } CX 6279 at 3, in camera. 547. For Humana HMO patients, the post-merger increase in inpatient price per case was { } CX 6279 at 3, in camera. 548. For Humana’s HMO plan, the price increase at ENH in the price per day was { } greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 8, in camera. For Humana’s HMO plan, the price increase at ENH in the price per case was { } greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 9, in camera. 549. For Humana’s HMO plan, the price increase at ENH in the price per day was { } greater than the average price increase across Non-Merging Chicago PMSA Hospitals. See CX 6279 at 8, in camera. For Humana’s HMO plan, the price increase at ENH in the price per case was { } greater than the average price increase across Non-Merging Chicago PMSA Hospitals. This result is not statistically significant. See CX 6279 at 9, in camera. 550. For Humana’s HMO plan, the price increase at ENH in the the price per day was { } greater than the average price increase VOLUME 144 Initial Decision increase across Chicago PMSA Teaching Hospitals. See CX 6279 at at 8, in camera. For Humana’s HMO plan, the price increase at ENH ENH in the price per case was { } greater than the average price price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 9, in camera.

(cc) PPO 551. For Humana PPO patients, the post-merger increase in inpatient price per day was { } CX 6279 at 3, in camera. 552. For Humana PPO patients, the post-merger increase in inpatient price per case was { } CX 6279 at 3, in camera. 553. For Humana’s PPO plan, the price increase at ENH in the price per day was { } greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 8, in camera. For Humana’s PPO plan, the price increase at ENH in the price per case was { } greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 9, in camera. 554. For Humana’s PPO plan, the price increase at ENH in the price per day was { } greater than the average price increase across Non-Merging Chicago PMSA Hospitals. See CX 6279 at 8, in camera. For Humana’s PPO plan, the price increase at ENH in the price per case was { } greater than the average price increase across Non-Merging Chicago PMSA Hospitals. See CX 6279 at 9, in camera.

555. For Humana’s PPO plan, the price increase at ENH in the price per day was { } greater than the average price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 8, in camera. For Humana’s PPO plan, the price increase at ENH in the price per case was { } greater than the average price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 9, in camera.

VOLUME 144 Initial Decision (dd) Summary 556. With respect to the Humana data, Haas-Wilson concluded from her regression analysis that the price increases at ENH were larger than the price increases at comparison hospitals, and that was true no matter how she measured resource intensity or which comparison group she used. Haas-Wilson, Tr. 2626-27, in camera; CX 6279 at 19, in camera.

557. For Humana, since the regression results take into account variation in patient mix, customer mix, and teaching intensity across hospitals over time, changes in these variables cannot explain all of the relatively larger price increases at ENH in the post-merger period compared to control group hospitals. Haas-Wilson, Tr. 2626- 27, in camera; CX 6279 at 19, in camera. 558. For Humana, using the control group of all Chicago PMSA Hospitals, and taking into account changes in patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were { } greater than at the average control group hospital. See CX 6279 at 19, in camera.

559. For Humana, using the control group of Non-Merging Chicago PMSA Hospitals, and taking into account changes in patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were { } greater than at the average control group hospital. See CX 6279 at 19, in camera.

560. For Humana, using the control group of Chicago PMSA Teaching Hospitals, and taking into account changes in patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were { } greater than at he average control group hospital. See CX 6279 at 19, in camera.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 131 Initial Decision (iv) Blue Cross Blue Shield 561. Blue Cross Blue Shield of Illinois (“Blue Cross Blue Shield”) is the largest insurer in Chicago and accounts for approximately 20% of ENH’s managed care business. Foucre, Tr. 939; Hillebrand, Tr. 1859; Mendonsa, Tr. 481. 562. ENH had less leverage to increase its prices in contract negotiations with Blue Cross Blue Shield than with other payors. CX 67 at 36; Neaman, Tr. 1181-83. Blue Cross Blue Shield had a very strong bargaining position against ENH. Neaman, Tr. 1181-83; Haas-Wilson, Tr. 2638-42, in camera.

563. The premerger time period for Blue Cross Blue Shield’s HMO plan is from { } The post-merger period for Blue Cross Blue Shield’s HMO plan is from { } Haas-Wilson, Tr. 2511-12, in camera; CX 5046 at 1.

564. { } Haas-Wilson, Tr. 2511-12, in camera; CX 5057 at 1,in camera.

565. For Blue Cross Blue Shield HMO patients, the post-merger percentage increase in inpatient price per day was { } CX 6279 at 3, in camera.

566. For Blue Cross Blue Shield HMO patients, the post-merger percentage increase in inpatient price per case was { } CX 6279 at 3, in camera.

567. For Blue Cross Blue Shield POS patients, the post-merger percentage increase in inpatient price per day was { } CX 6279 at 3, in camera.

568. For Blue Cross Blue Shield POS patients, the post-merger percentage increase in inpatient price per case was { } CX 6279 at 3,in camera.

VOLUME 144 Initial Decision 569. For Blue Cross Blue Shield PPO patients, the post-merger percentage increase in inpatient price per day was { } CX 6279 at 3, in camera.

570. For Blue Cross Blue Shield PPO patients, the post-merger percentage increase in inpatient price per case was { } CX 6279 at 3, in camera.

571. The Blue Cross Blue Shield claims data does not show that prices to Blue Cross Blue Shield at ENH rose faster than prices at other hospitals in the Chicago PMSA following the merger between Evanston and Highland Park. CX 6279 at 18, in camera. 572. Using the same approach with the Blue Cross Blue Shield data, Haas-Wilson concluded that the price changes at ENH do not appear to be different in most cases than the price changes at the control group hospitals. Haas-Wilson, Tr. 2626, in camera. (c) Data from the IDPH Universal Dataset 573. The Illinois Department of Public Health (“IDPH”) Universal Dataset compiles data from all hospitals in Illinois. The data is very comprehensive. It includes data on all inpatient hospital stays at all hospitals in Illinois, regardless of the managed care organization. Haas-Wilson, Tr. 2500; Haas-Wilson, Tr. 2582-83, in camera.

574. The IDPH Universal Dataset includes the hospitals’ list prices for each procedure which reflect each hospital’s chargemaster. The Universal Dataset does not include information on the actual transaction prices, including managed care discounts and patient payments, that hospitals receive. Haas-Wilson, Tr. 2500. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 133 Initial Decision 575. In order to use the data from the IDPH Universal Dataset to calculate prices paid to managed care organizations, Haas-Wilson used a method that has been used by other health care economists to establish prices paid by managed care organizations. Haas-Wilson, Tr. 2527, in camera.

576. Haas-Wilson used the IDPH Universal Dataset with other data from the Medicare Cost Reports to derive an estimate of negotiated prices. Haas-Wilson, Tr. 2527-28, in camera. Medicare Cost Reports are reports that are required to be submitted by every hospital that participates in Medicare. Haas-Wilson, Tr. 2527, in camera. The Medicare Cost Reports show aggregate data on both net payments and gross payments by hospitals for inpatient and outpatient services. Haas-Wilson, Tr. 2529, in camera. 577. Using the Medicare Cost Reports, Haas-Wilson constructed a ratio of net receipts to gross billing amounts, and then multiplied that ratio by the billing information in the IDPH Universal Dataset (which is based on list prices) to get an estimate of the actual negotiated price. Haas-Wilson, Tr. 2529, in camera. 578. The ratio Haas-Wilson used included both inpatient and outpatient payments. Haas-Wilson, Tr. 2529, in camera. 579. While there is potential bias in such an approach, any bias would be small. If there was a bias, “it would work against finding a price increase.” Haas-Wilson, Tr. 2529-30, in camera. 580. The IDPH Universal Dataset does not identify the individual managed care organization that paid for a particular patient. Haas-Wilson, Tr. 2531-32, in camera. The IDPH Universal Dataset breaks down who paid for a particular patient only by categories of payors, such as: (1) all patients; (2) commercial and self pay; and (3) self administered as well as other categories. Haas- Wilson, Tr. 2532, in camera; CX 6279 at 7, in camera. VOLUME 144 Initial Decision 581. Haas-Wilson used the two calendar years 1998 and 1999 as the premerger period and the two calendar years 2001 and 2002 as the post-merger period in comparing premerger and post-merger prices. Haas-Wilson, Tr. 2530-31, in camera. 582. Haas-Wilson compared the price increases estimated from the IDPH Universal Dataset and the Medicare Cost Reports with the change in the Chicago medical care CPI for the period beginning in 1998 to the end of 2002. During that period, the Chicago medical care CPI increased 20.3%. Haas-Wilson, Tr. 2533, in camera. 583. Using the IDPH Universal Dataset in conjunction with the Medicare Cost Reports, for any of Haas-Wilson’s three control groups, and for any categorization of the different types of patients in the IDPH Universal Dataset, changes in patient mix, customer mix, and teaching intensity do not explain the relative price increases at ENH following the merger with Highland Park, when compared to control groups. All of the results show that the postmerger price increases at ENH were greater than the average price increases at comparison hospitals, even taking into account variations in patient mix, customer mix, and teaching intensity. Haas-Wilson, Tr. 2631-35, in camera; see CX 6279 at 20, in camera. 584. These results are statistically significant to the 1% level. CX 6279 at 10, 20, in camera.

585. Neither theory nor previous empirical research provided guidance on the best way to measure patient mix (capturing differences in resource use from both changes in case mix and severity of illness) across hospitals, so Haas-Wilson measured patient mix four different ways in the regression model: (1) the case case mix and severity of illness measure based on the APRDRGs; (2) the case mix and severity of illness measure based on the EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 135 Initial Decision APRDRGs in combination with a length of stay variable; (3) the case mix measure based on DRG weights; and (4) the case mix measure based on DRG weights in combination with the length of stay variable. Haas-Wilson, Tr. 2622-23, in camera. (i) All Patients 586. For all patients, the post-merger increase in inpatient price per day was 48%. CX 6279 at 7, in camera. 587. For all patients, the post-merger increase in inpatient price per case was 30%. CX 6279 at 7, in camera. 588. For all patients, the price increase at ENH in the price per day was 34% greater than the average price increase across all Chicago PMSA Hospitals. CX 6279 at 10, in camera. For all patients, the price increase at ENH in the price per case was 21% greater than the average price increase across all the Chicago PMSA Hospitals. CX 6279 at 11, in camera.

589. For all patients, the price increase at ENH in the price per day was 34% greater than the average price increase across Non- Merging Chicago PMSA Hospitals. See CX 6279 at 10, in camera. For all patients, the price increase at ENH in the price per case was 21% greater than the average price increase across Non-Merging Chicago PMSA Hospitals. See CX 6279 at 11, in camera. 590. For all patients, the price increase at ENH in the price per day was 34% greater than the average price increase across the Chicago PMSA Teaching Hospitals. See CX 6279 at 10, in camera. For all patients, the price increase at ENH in the price per case was 21% greater than the average price increase across the Chicago PMSA Teaching Hospitals. See CX 6279 at 11, in camera. VOLUME 144 Initial Decision 591. For all patients in the DPH Universal Dataset, using the control group of all Chicago PMSA Hospitals, and taking into account changes in patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were 14.2 to 16.8% greater than at the average control group hospital. The difference in the price increases at ENH and the control group hospitals is statistically significant. See CX 6279 at 20, in camera. 592. For all patients in the IDPH Universal Dataset, using the control group of Non-Merging Chicago PMSA Hospitals, and taking into account differences in patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were 15.2 to 17.0% greater than at the average control group hospital. The difference in the price increases at ENH and the control group hospitals is statistically significant. See CX 6279 at 20, in camera. 593. For all patients in the DPH Universal Dataset, using the control group of Chicago PMSA Teaching Hospitals, and taking into account differences in patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were 13.2 to 15.5% greater than at the average control group hospital. The difference in the price increases at ENH and the control group hospitals is statistically significant. See CX 6279 at 20, in camera. (ii) Commercial and Self Pay Patients 594. For commercial and self pay patients, the post-merger increase in inpatient price per day was 46%. CX 6279 at 7, in camera.

595. For commercial and self pay patients, the post-merger increase in inpatient price per case was 27%. CX 6279 at 7, in camera.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 137 Initial Decision 596. For commercially insured and self pay patients, the price increase at ENH in the price per day was 29% greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 10, in camera. For commercially insured and self pay patients, the price increase at ENH in the price per case was 15% greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 11, in camera. 597. For commercially insured and self pay patients, the price increase at ENH in the price per day was 29% greater than the average price increase across Non-Merging Chicago PMSA Hospitals. See CX 6279 at 10, in camera. For commercially insured and self pay patients, the price increase at ENH in the price per case was 16% greater than the average price increase across Non- Merging Chicago PMSA Hospitals. See CX 6279 at 11, in camera. 598. For commercially insured and self pay patients, the price increase at ENH in the price per day was 26% greater than the average price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 10, in camera. For commercially insured and self pay patients, the price increase at ENH in the price per case was 14% greater than the average price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 11, in camera. 599. For commercially insured and self pay patients in the IDPH Universal Dataset, using the control group of all Chicago PMSA Hospitals, and taking into account changes in patient mix, customer mix, and teaching intensity the post-merger price increases at ENH were 12.7 to 15.0% greater than at the average control group hospital. The difference in the price increases at ENH and the control group hospitals is statistically significant. See CX 6279 at 20, in camera.

600. For commercially insured and self pay patients in the IDPH IDPH Universal Dataset, using the control group of Non-Merging Chicago PMSA Hospitals, and taking into account changes in patient VOLUME 144 Initial Decision patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were 12.9 to 17.0% greater than at the average control group hospital. The difference in the price increases increases at ENH and the control group hospitals is statistically significant. See CX 6279 at 20, in camera. 601. For commercially insured and self pay patients in the IDPH Universal Dataset, using the control group of Chicago PMSA Teaching Hospitals, and taking into account changes in patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were 11.1 to 13.0% greater than at the average control group hospital. The difference in the price increases at ENH and the control group hospitals is statistically significant. See CX 6279 at 20, in camera.

(iii)Commercial, Self Pay, Self Administered, and HMO Patients 602. For commercial, self pay, self administered, and HMO patients, the post-merger increase in inpatient price per day was 46%. CX 6279 at 7, in camera.

603. For commercial, self pay, self administered, and HMO patients, the post-merger increase in inpatient price per case was 26%. CX 6279 at 7, in camera.

604. For commercially insured, self pay, HMO, and self administered patients, the price increase at ENH in the price per day was 29% greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 10, in camera. For commercially insured, self pay, HMO, and self administered patients, the price increase at ENH in the price per case was 14% greater than the average price increase across all Chicago PMSA Hospitals. See CX 6279 at 11, in camera.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 139 Initial Decision 605. For commercially insured, self pay, HMO, and self administered patients, the price increase at ENH in the price per day was 28% greater than the average price increase across Non- Merging Chicago PMSA Hospitals. See CX 6279 at 10, in camera. For commercially insured, self pay, HMO, and self administered patients, the price increase at ENH in the price per case was 15% greater than the average price increase across Non-Merging Chicago PMSA Hospitals. See CX 6279 at 11, in camera. 606. For commercially insured, self pay, HMO, and self administered patients, the price increase at ENH in the price per day was 27% greater than the average price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 10, in camera. For commercially insured, self pay, HMO, and self administered patients, the price increase at ENH in the price per case was 13% greater than the average price increase across Chicago PMSA Teaching Hospitals. See CX 6279 at 11, in camera. 607. For commercially insured, self pay, HMO, and self administered patients in the IDPH Universal Dataset, using the control group of all Chicago PMSA Hospitals, and taking into account changes in patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were 13.7 to 15.7% greater than at the average control group hospital. The difference in the price increases at ENH and the control group hospitals is statistically significant. See CX 6279 at 20, in camera. 608. For commercially insured, self pay, HMO, and self administered patients in the IDPH Universal Dataset, using the control group of Non-Merging Chicago PMSA Hospitals, and taking taking into account changes in patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were 14.2 14.2 to 17.9% greater than at the average control group hospital. The The difference in the price increases at ENH and the control group VOLUME 144 Initial Decision group hospitals is statistically significant. See CX 6279 at 20, in camera.

609. For commercially insured, self pay, HMO, and self administered patients in the DPH Universal Dataset, using the control group of Chicago PMSA Teaching Hospitals, and taking into account changes in patient mix, customer mix, and teaching intensity, the post-merger price increases at ENH were 11.9 to 13.5% greater than at the average control group hospital. The difference in the price increases at ENH and the control group hospitals is statistically significant. See CX 6279 at 20, in camera. 610. The IDPH Universal Dataset shows that prices to managed care organizations went up faster at ENH than at other hospitals after the merger with Highland Park. This result does not change with the different control groups and does not change with the different patient groups identified in the IDPH Universal Dataset. F. 591-93, 599-601, 607-09. All of the results show that the post-merger price increases at ENH were greater than the average price increases at comparison hospitals, even taking into account variations in patient mix, customer mix, and teaching intensity. F. 583. (d) Data Submitted by the Economic Consulting Firm NERA on Behalf of ENH 611. National Economic Research Associates (“NERA”), an economic consulting firm hired by ENH, submitted data to the FTC, on ENH’s behalf. Haas-Wilson, Tr. 2498. The NERA data reported actual negotiated prices for ENH’s fiscal years 1999 through 2001. Haas-Wilson, Tr. 2498.

612. The NERA data includes data on many commercial payors, payors, more payors than there are payors for which there was EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 141 Initial Decision claims data. Haas-Wilson, Tr. 2499; CX 6279 at 4 (showing data for for 13 payors), in camera.

613. Haas-Wilson used the fiscal year 1999 as the premerger period and fiscal year 2001 as the post-merger period for the NERA data in comparing premerger and post-merger prices. Fiscal year 2000 was not included in the analysis because it was considered a transition year, a period of time in which ENH was renegotiating many of its contracts with managed care organizations. Haas- Wilson, Tr. 2519, in camera.

614. The NERA data contained information only on ENH. It did not contain data on prices at other hospitals to use for comparison. Haas-Wilson, Tr. 2498-99. Therefore, Haas-Wilson compared the price increase per case estimated from the NERA data with the change in the Chicago medical care CPI for the period from the beginning of ENH’s fiscal year 1999 through the end of fiscal year 2001. During that period, the Chicago medical care CPI increased 11%. Haas-Wilson, Tr. 2520-22, in camera. 615. The NERA findings are reported per adult day and per adult case only. See CX 6279 at 4, in camera.

616. The NERA data showed “large price increases at ENH postmerger for many payers, and in some cases really large [price increases].” Haas-Wilson, Tr. 2519-20, in camera; CX 6279 at 4, in camera. For example, Haas-Wilson found that the percentage increase for PHCS using the NERA data was { } Haas-Wilson, Tr. 2522-23, in camera; see also Ballengee, Tr. 179 ({ }). (i) First Health 617. For First Health patients, the post-merger increase in inpatient price per day was { } CX 6279 at 4, in camera. VOLUME 144 Initial Decision 618. For First Health patients, the post-merger increase in inpatient price per case was { } Haas-Wilson, Tr. 2516, in camera; CX 6279 at 4, in camera.

(ii) Aetna 619. For Aetna patients, the post-merger increase in inpatient price per day was { } CX 6279 at 4, in camera. 620. For Aetna patients, the post-merger increase in inpatient price per case was { } Haas-Wilson, Tr. 2537, in camera; CX 6279 at 4, in camera.

(iii) Northwestern Students 621. For Northwestern Student patients, the post-merger increase in inpatient price per day was { } CX 6279 at 4, in camera. 622. For Northwestern Student patients, the post-merger increase in inpatient price per case was { } CX 6279 at 4, in camera. (iv) Blue Cross Blue Shield 623. For Blue Cross Blue Shield patients, the post-merger increase in inpatient price per day was { } CX 6279 at 4, in camera.

624. For Blue Cross Blue Shield patients, the post-merger increase in inpatient price per case was { } CX 6279 at 4, in camera.

(v) Cigna EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 143 Initial Decision 625. For Cigna patients, the post-merger increase in inpatient price per day was { } CX 6279 at 4, in camera. 626. For Cigna patients, the post-merger increase in inpatient price per case was { } CX 6279 at 4, in camera. (vi) PPONext 627. For PPONext patients, the post-merger increase in inpatient price per day was { } CX 6279 at 4, in camera. 628. For PPONext patients, the post-merger increase in inpatient price per case was { } CX 6279 at 4, in camera. (vii) Humana 629. For Humana patients, the post-merger increase in inpatient price per day was { } CX 6279 at 4, in camera. 630. For Humana patients, the post-merger increase in inpatient price per case was { } CX 6279 at 4, in camera. (viii) MultiPlan 631. For MultiPlan patients, the post-merger increase in inpatient price per day was { } CX 6279 at 4, in camera. 632. For Multiplan patients, the post-merger increase in inpatient price per case was { } CX 6279 at 4, in camera. (ix) Preferred Plan 633. For Preferred Plan patients, the post-merger increase in inpatient price per day was { } CX 6279 at 4, in camera. 634. For Preferred Plan patients, the post-merger increase in inpatient price per case was { } CX 6279 at 4, in camera. VOLUME 144 Initial Decision (x) PHCS 635. For PHCS patients, the post-merger increase in inpatient price per day was { } CX 6279 at 4, in camera. 636. For PHCS patients, the post-merger increase in inpatient price per case was { } Haas-Wilson, Tr. 2522-23, in camera; CX 6279 at 4, in camera.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 145 Initial Decision (xi) Unicare 637. For Unicare patients, the post-merger increase in inpatient price per day was { } CX 6279 at 4, in camera. 638. For Unicare patients, the post-merger increase in inpatient price per case was { } CX 6279 at 4, in camera. (xii) United 639. For United patients, the post-merger increase in inpatient price per day was { } CX 6279 at 4, in camera. 640. For United patients, the post-merger increase in inpatient price per case was { } Haas-Wilson, Tr. 2522, in camera; CX 6279 at 4, in camera.

(e) Data Submitted by ENH in Response to a Civil Investigative Demand Issued by the Federal Trade Commission 641. ENH submitted data in response to a CID issued by the Federal Trade Commission. The CID response data was similar to the NERA data. The CID response data reported actual negotiated prices for ENH’s fiscal years 1999 through 2002. Haas-Wilson, Tr. 2499-500.

642. The CID response data covered at least fourteen payors. CX 6279 at 5, in camera.

643. Haas-Wilson used the fiscal year 1999 as the premerger period and the fiscal year 2002 as the post-merger period for the CID data in comparing premerger and post-merger prices. Haas- Haas-Wilson, Tr. 2523, in camera. Fiscal years 2000 and 2001 were were not included in the analysis, because for this data set they were were both considered transition years, a period of time in which VOLUME 144 Initial Decision ENH was renegotiating many of its contracts with commercial payors. Haas-Wilson, Tr. 2523-24, in camera. 644. Haas-Wilson compared the price increase per case estimated from the CID data with the change in the Chicago medical care CPI for the period from the beginning of ENH’s fiscal year 1999 through the end of fiscal year 2002. During that period, the change in Chicago medical care CPI increased 14.3%. Haas-Wilson, Tr. 2526, in camera.

645. The CID data “showed for most commercial payers, there were large price increases at ENH” and “at some payers really large price increases.” Haas-Wilson, Tr. 2524-25, in camera; CX 6279 at 4-5, in camera.

(i) Beech Street/Capp Care 646. For Beech Street/Capp Care patients, the post-merger increase in inpatient price per day was { } CX 6279 at 5, in camera.

647. For Beech Street/Capp Care patients, the post-merger increase in inpatient price per case was { } CX 6279 at 5, in camera.

(ii) Cigna 648. For Cigna patients, the post-merger increase in inpatient price per day was { } CX 6279 at 5, in camera. 649. For Cigna patients, the post-merger increase in inpatient price per case was { } CX 6279 at 5, in camera. (iii) First Health EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 147 Initial Decision 650. For First Health patients, the post-merger increase in inpatient price per day was { } CX 6279 at 5, in camera. 651. For First Health patients, the post-merger increase in inpatient price per case was { } CX 6279 at 5, in camera. (iv) One Health (Great West) 652. For One Health patients, the post-merger increase in inpatient price per day was { } CX 6279 at 5, in camera. 653. For One Health patients, the post-merger increase in inpatient price per case was { } CX 6279 at 5, in camera. (v) Aetna 654. For Aetna patients, the post-merger increase in inpatient price per day was { } CX 6279 at 5, in camera. 655. For Aetna patients, the post-merger increase in inpatient price per case was { } Haas-Wilson, Tr. 2537, in camera; CX 6279 at 5, in camera.

(vi) Blue Cross Blue Shield 656. For Blue Cross Blue Shield patients, the post-merger increase in inpatient price per day was { } CX 6279 at 5, in camera.

657. For Blue Cross Blue Shield patients, the post-merger increase in inpatient price per case was { } CX 6279 at 5, in camera.

(vii) HFN 658. For HFN patients, the post-merger increase in inpatient price per day was { } CX 6279 at 5, in camera. VOLUME 144 Initial Decision 659. For HFN patients, the post-merger increase in inpatient price per case was { } CX 6279 at 5, in camera. (viii) Humana 660. For Humana patients, the post-merger increase in inpatient price per day was { } CX 6279 at 5, in camera. 661. For Humana patients, the post-merger increase in inpatient price per case was { } CX 6279 at 5, in camera. (ix) MultiPlan 662. For MultiPlan patients, the post-merger increase in inpatient price per day was { } CX 6279 at 5, in camera. 663. For MultiPlan patients, the post-merger increase in inpatient price per case was { } CX 6279 at 5, in camera. (x) PHCS 664. For PHCS patients, the post-merger increase in inpatient price per day was { } CX 6279 at 5, in camera. 665. For PHCS patients, the post-merger increase in inpatient price per case was { } CX 6279 at 5, in camera. (xi) Preferred Plan 666. For Preferred Plan patients, the post-merger increase in inpatient price per day was { } CX 6279 at 5, in camera. 667. For Preferred Plan patients, the post-merger increase in inpatient price per case was { } CX 6279 at 5, in camera. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 149 Initial Decision (xii) State of Illinois 668. For State of Illinois patients, the post-merger increase in inpatient price per day was { } CX 6279 at 5, in camera. 669. For State of Illinois patients, the post-merger increase in inpatient price per case was { } CX 6279 at 5, in camera. (xiii) Unicare 670. For Unicare patients, the post-merger increase in inpatient price per day was { } CX 6279 at 5, in camera 671. For Unicare patients, the post-merger increase in inpatient price per case was { } CX 6279 at 5, in camera. (xiv) United 672. For United patients, the post-merger increase in inpatient price per day was { } CX 6279 at 5, in camera. 673. For United patients, the post-merger increase in inpatient price per case was { } CX 6279 at 5, in camera. (f) Baker’s Analysis 674. Baker defined the premerger time period for his analysis as all observations before January 1, 2000 because that was the effective date of the merger. Baker, Tr. 4635, in camera. 675. Baker used the data provided by the managed care organizations to determine post-merger increases in inpatient price per case. Baker then compared these results to the post-merger increases in prices at control groups of eighteen hospitals provided by Noether. Baker, Tr. 4637-38, in camera; Haas-Wilson, Tr. 2548- 49, in camera.

VOLUME 144 Initial Decision 676. Baker found that, for United patients across all United plans, the raw, unadjusted post-merger increase in inpatient price per case at ENH was { }, while the post-merger price increase for United patients at his control group hospitals was { } Haas-Wilson, Tr. 2564-65, in camera.

677. Baker found that, for Aetna patients across all Aetna plans, the raw, unadjusted post-merger increase in inpatient price per case at ENH was { }, while the post-merger price increase for Aetna patients at his control group hospitals was { } Haas-Wilson, Tr. 2566-67, in camera; Baker, Tr. 4744-46, in camera. 678. Baker found that, for Humana patients across all Humana plans, the raw, unadjusted post-merger increase in inpatient price per case at ENH was { }, while the post-merger price increase for Humana patients at his control hospitals was { } Haas-Wilson, Tr. 2573, in camera; see Baker, Tr. 4747, in camera. 679. Baker found that the post-merger price increase for Blue Cross Blue Shield was exactly the same as the post-merger price increase for his control group – both were { } Haas-Wilson, Tr. 2569, in camera.

680. Baker did not calculate price changes for individual plans of managed care organizations. He looked at prices for the payors as a whole, and also aggregated over all of the payors. Baker, Tr. 4631- 32, in camera.

681. Baker controlled for case mix variation in his regression differently from Haas-Wilson. Baker Tr. 4648-49, in camera. Baker had a length of stay variable in his regression. Baker, Tr. 4800, in camera.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 151 Initial Decision 682. Baker used Noether’s control groups both for his price change analysis and his price level analysis. Baker, Tr. 4637-38, in camera.

683. Baker based his analysis on usable managed care claims data produced during discovery from four managed care organizations. This data reflects the prices actually paid by managed care organizations for ENH’s services. Baker, Tr. 4646-47, in camera.

684. Baker admitted that the pricing pattern of ENH’s prices to Humana, Aetna, and United was consistent with ENH obtaining market power through the merger with Highland Park. Baker, Tr. 4742-43, in camera.

685. Baker calculated an average price increase across all four payors whose claims data he used and he found that, for all United, Aetna, Blue Cross Blue Shield, and Humana patients across all plans, the post-merger increase in inpatient price per case at ENH was { }, while the post-merger price increase at his control group hospitals was { } Haas-Wilson, Tr. 2584-85, in camera. 686. Baker found that the raw, unadjusted price increase for United, Aetna, Blue Cross Blue Shield, and Humana patients, with inpatient and outpatient combined, at ENH from the premerger period to the post-merger period was { } Baker, Tr. 4639-40, in camera.

687. Baker found that ENH’s prices for inpatient and outpatient combined increased by { } more than the prices of the control group hospitals, without controlling for patient mix. Baker, Tr. 4640- 41, in camera.

688. Baker testified that his best estimate of ENH’s non-quality non-quality adjusted price increase at the time of the merger, as compared to a control group of hospitals, and adjusted for variation variation in case-mix across hospitals, for inpatient and outpatient VOLUME 144 Initial Decision services combined, is 11 to 12%. Baker, Tr. 4617-19, 4795-96, in camera.

689. Baker found that for his four payors combined, the postmerger price increases for inpatient services at ENH were 10.0% higher than the post-merger price increases on average at the comparison hospitals, taking into account the variation in the independent variables that he included in his regression model. Haas-Wilson, Tr. 2636-37, in camera; Baker Tr. 4645-46, in camera. When Baker excluded obstetrics, the estimated price increases at ENH for inpatient services were 9%. Baker, Tr. 4646, in camera. 690. Baker testified that his best estimate of ENH’s non-quality adjusted price increase at the time of the merger, as compared to a control group of hospitals, and adjusted for variation in case-mix across hospitals, for inpatient services only, is 9 to 10%. Baker, Tr. 4617-20, 4795-96, in camera.

691. There is no record evidence regarding Baker’s estimates of price changes at individual managed care organizations that were both case-mix adjusted and compared to a control group of hospitals. Baker, Tr. 4640, in camera.

692. Baker testified that examining the overall price changes, rather than looking at any individual managed care organization’s price change, is more appropriate because the market alleged by Complaint Counsel was the managed care market as a whole. Baker, Tr. 4648, in camera.

(3) Explanations of Price Increases Other than Market Power Ruled Out (a) Methodology 693. It is not feasible to directly test whether or not market power is the explanation behind the price increases at ENH. Haas- EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 153 Initial Decision Haas-Wilson, Tr. 2482. Because market power cannot be tested for for directly, “the best available method is to develop [a] list based on on theory and what theory would expect to result in a price increase increase and then use empirical tests based on available data to be able to either cross these items off the list or, if you’re not able with with your empirical test to cross them off, then see what you’re left left with at the end of the analysis.” Haas-Wilson, Tr. 2482. 694. It is not possible to test for all possible explanations of a price increase, so it is necessary to look for reasonable explanations that are grounded in economic theory. Haas-Wilson, Tr. 2481. 695. Haas-Wilson, drawing upon economic theory, came up with a list of eight potential explanations for the price increases at ENH after the merger other than market power or learning about demand. The “basis for including things in this list was economic theory and what economic theory suggested would be potential explanations for the large post-merger price increase at ENH.” Haas-Wilson, Tr. 2481.

696. The eight plausible explanations of the price increases at ENH, aside from market power or learning about demand, were: (1) cost increases that affect all hospitals; (2) changes in regulations that affect all hospitals; (3) increases in consumer demand for hospital services; (4) increases in quality at ENH; (5) changes in the mix of patients; (6) changes in the mix of customers; (7) increases in teaching intensity; and (8) decreases in outpatient prices. Haas- Wilson, Tr. 2482-88.

697. Haas-Wilson tested whether any of these potential explanations could explain the price increases at ENH and found that they could not. Haas-Wilson, Tr. 2481. VOLUME 144 Initial Decision (b) Changes in Costs, Regulations, Consumer Demand, Quality, and Outpatient Prices Can Be Ruled Out 698. Economic theory suggests that if there are increases in demand over a time period, one would expect those increases in demand in the Chicago area to increase prices at all hospitals in the Chicago area. Therefore, Haas-Wilson tested for whether increases in demand would explain why ENH’s prices increased. Haas- Wilson, Tr. 2484.

699. An example of what could cause an increase in demand that would subsequently affect prices is “[t]o the extent the elderly consume more hospital services than the young, to the extent the population is aging in the Chicago area, that would likely increase demand for hospital services in the Chicago area and could potentially explain, therefore, price increases at all hospitals in the Chicago area.” Haas-Wilson, Tr. 2484.

700. In her analysis, Haas-Wilson focused on price increases instead of price levels because the market for hospital services can be characterized as a market for a differentiated product as opposed to a product that would be characterized as homogenous. Consumers are willing and able to pay higher prices for certain aspects of product differentiation, e.g., convenient location or reputation. Thus, because prices can vary in the market for a differentiated service for many different reasons, one may not conclude anything about market power by merely using a cross-sectional analysis of hospital prices at a single point in time. Haas-Wilson, Tr. 2492. 701. In contrast, by looking at price changes over time, one can can compare the price change at one hospital to the price change at at another hospital. Using such an approach, one can conclude that that there is a change in market power if there are price increases after having ruled out the other possible explanations for greater EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 155 Initial Decision price increases at one hospital versus another. Haas-Wilson, Tr. 2495.

702. Haas-Wilson considered whether increases in costs, changes in regulation, and changes in demand for hospital services that would affect all hospitals could have been a possible explanation for the post-merger price increases at ENH. To test this hypothesis, she looked to see whether prices increased more at ENH than at comparison hospitals. If they did, general increases in costs, changes in regulation, and changes in demand for hospital services could not be a possible explanation for all of the post-merger price increases at ENH. Haas-Wilson, Tr. 2542-44, in camera. 703. Haas-Wilson was able to directly rule out five potential explanations of the price increases at ENH: (1) cost increases; (2) changes in regulations; (3) changes in consumer demand; (4) changes in quality; and (5) declines in outpatient prices. F. 704-26. (i) Changes in Costs 704. Economic theory suggests that when costs increase in competitive markets, one would expect to see prices increase. Therefore, Haas-Wilson tested for whether cost increases in the Chicago area would explain why ENH’s prices increased. Haas- Wilson, Tr. 2482.

705. An example of a kind of cost increase that could take place in an area that would lead to a price increase is a shortage of nurses in the area. If a hospital had to pay higher wages in order to hire nurses, that would be an increase in cost that would affect the hospital and all of the other hospitals in the area, and potentially lead to a price increase. Haas-Wilson, Tr. 2482-83. 706. Prices at ENH rose relative to the prices at other hospitals, hospitals, as explained above in Section II.D.1.b.2. These relative price increases rule out cost increases as an explanation of the price VOLUME 144 Initial Decision price increases observed at ENH. Haas-Wilson, Tr. 2562-63, 2565, 2565, 2573-74, 2579, 2583, 2586, in camera. 707. The relative price increases rule out any cause of the price increases that would affect all the hospitals in the control groups similarly. Haas-Wilson, Tr. 2562-63, 2565, 2573-74, 2579, 2583, 2586, in camera.

(ii) Changes in Regulations 708. Because a change in regulation that affected all hospitals in the Chicago area could potentially explain price increases at all hospitals in the Chicago area, Haas-Wilson tested for whether changes in regulations would explain why ENH’s prices increased. Haas-Wilson, Tr. 2483.

709. An example of a change in regulation that could affect the prices at hospitals is taken from California. In California, where they are particularly prone to earthquakes, there are regulations requiring hospitals to make sure their buildings are able to withstand earthquakes of certain levels. Such a regulation clearly would increase costs at all hospitals in California and would be expected to lead to higher prices. Haas-Wilson, Tr. 2483-84. 710. Prices at ENH rose relative to the prices at other hospitals, as explained above in Section II.D.1.b.2. These relative price increases rule out regulatory changes as an explanation of the price increases observed at ENH. Haas-Wilson, Tr. 2562-63, 2565, 2573- 74, 2579, 2583, 2586, in camera.

711. The relative price increases rule out any cause of the price increases that would affect all the hospitals in the control groups similarly. Haas-Wilson, Tr. 2562-63, 2565, 2573-74, 2579, 2583, 2586, in camera.

(iii) Changes in Consumer Demand EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 157 Initial Decision 712. Prices at ENH rose relative to the prices at other hospitals, as explained above in Section II.D.1.b.2. These relative price increases rule out cost increases, regulatory changes, and increases in demand as explanations of the price increases observed at ENH. Haas-Wilson, Tr. 2562-63, 2565, 2573-74, 2579, 2583, 2586, in camera.

713. The relative price increases rule out any cause of the price increases that would affect all the hospitals in the control groups similarly. Haas-Wilson, Tr. 2562-63, 2565, 2573-74, 2579, 2583, 2586, in camera.

(iv) Changes in Quality 714. If quality is increasing in general, that would lead to potentially higher prices at all hospitals. Haas-Wilson, Tr. 2485. If quality is increasing at one hospital relative to other hospitals, and the buyers of hospital services value that increase in quality, then that could potentially explain a greater price increase at the first hospital. Haas-Wilson, Tr. 2485.

715. Haas-Wilson relied upon findings by Complaint Counsel’s healthcare quality expert, Dr. Patrick S. Romano, Professor of Internal Medicine and Pediatrics at University of California at Davis, School of Medicine, that the post-merger increase in quality at ENH was not greater than the increases in quality at relevant control hospitals. Haas-Wilson, Tr. 2586-88, in camera. See also F. 853-68. Haas-Wilson did not conduct an independent analysis of non-clinical quality. Haas-Wilson, Tr. 2446-47; Haas-Wilson, Tr. 2586, in camera.

716. Increases in quality at ENH cannot explain the relatively larger price increases at ENH after the merger when compared to the price increases at other hospitals. Haas-Wilson, Tr. 2587-88, 2615, in camera.

VOLUME 144 Initial Decision (v) Changes in Outpatient Prices 717. Though economic theory does not predict that decreases in outpatient services prices would lead to increases in inpatient service prices, some managed care payors indicated that they would be concerned about what they paid for all the products that they were purchasing from a hospital. Haas-Wilson, Tr. 2487. 718. To the extent that a managed care organization is concerned about the total price, a managed care organization might be willing to pay higher prices for inpatient services if they were getting outpatient services at a lower price. It might be willing to trade one off for the other. Haas-Wilson, Tr. 2487-88. 719. Because decreases in prices of outpatient services is one of the potential explanations for the price increases that were observed at ENH after the merger, Haas-Wilson analyzed this possibility. Haas-Wilson, Tr. 2607, in camera.

720. Haas-Wilson tested whether changes in the prices of outpatient services at ENH differed from the change in prices of outpatient services at control hospitals over the same period of time. Haas-Wilson, Tr. 2607-08, in camera. Haas-Wilson used managed care data to test this hypothesis. Haas-Wilson, Tr. 2608, in camera. 721. There was no decrease in the prices of outpatient services to managed care organizations at the time of the increases in the prices of inpatient services. Haas-Wilson, Tr. 2614-15, in camera. 722. CX 6279 at 17 shows the post-merger increases in price per per case of outpatient care at ENH and at control hospitals with more than 100 cases of outpatient care in both the premerger and the the post-merger period. The table gives the result by payor and plan plan type for ENH and the three different groups of control EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 159 Initial Decision hospitals. Haas-Wilson, Tr. 2610, in camera; CX 6279 at 17, in camera.

723. Based on the empirical analysis, which used the managed care claims data, Haas-Wilson concluded that payors did not accept lower outpatient prices in return for higher inpatient prices. Haas- Wilson, Tr. 2614-15, in camera.

724. The finding that outpatient prices did not decline is consistent with Baker’s analysis. Baker estimated that the price increases at ENH for four managed care payors, relative to the control group, for inpatient and outpatient services combined was 11 to 12%. Baker, Tr. 4617-18, in camera. Looking at just inpatient services, Baker estimated that the price increases at ENH for four managed care payors, relative to the control group, was 9 to 10%. Baker, Tr. 4620, in camera. This implies that the price of outpatient services at ENH for Baker’s four payors increased more than the price of inpatient services. Baker, Tr. 4797, in camera. 725. Baker did not explicitly estimate the price change at ENH for outpatient services, because he could not adjust for case mix variation with the outpatient data. What he did to create an estimate of ENH’s price change for inpatient services was find a case mix ratio from the inpatient data, and apply that same ratio to the outpatient data. Baker, Tr. 4642, in camera. Baker agrees that the implication of his estimates is that outpatient prices did not decline. Baker, Tr. 4797, in camera.

726. Based on her empirical review of the managed care data, Haas-Wilson concluded that changes in the price of outpatient services were not a possible explanation for the post-merger ENH price increase. Haas-Wilson, Tr. 2615, in camera. VOLUME 144 Initial Decision (c) Changes in Patient Mix, Customer Mix, and Teaching Intensity Can Be Ruled Out Out (i) Regression Analysis Methodology 727. Haas-Wilson developed a multiple regression model to evaluate whether the remaining possible explanations (changes in patient mix, customer mix, or teaching intensity) were responsible for the post-merger ENH price increases. Haas-Wilson, Tr. 2615, in camera.

728. Multiple regression analysis is a statistical tool commonly used in econometrics that allows the researcher to study the impact of many variables simultaneously that may have an influence on the dependent variable of interest. Haas-Wilson, Tr. 2616, in camera. 729. Haas-Wilson employed a multiple regression model to measure the effect of the merger on the change in prices, while simultaneously taking into account changes in other variables changes in patient mix, customer mix, and teaching intensity. Haas- Wilson, Tr. 2616, 2619, in camera.

730. In Haas-Wilson’s multiple regression model, prices at ENH and control hospitals were the dependent variables, and patient mix (case mix and severity of illness), customer mix, and teaching intensity were included in the independent variables. Haas-Wilson, Tr. 2619-20, in camera.

731. Haas-Wilson used a difference in differences approach to see if the price increases at ENH after the merger were larger than the price increases at a control group of hospitals. Haas-Wilson, Tr. 2620, in camera.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 161 Initial Decision 732. Haas-Wilson used the same control groups for her multiple regression model that she used earlier in her difference in differences analysis of whether the price increases at ENH were greater than at the control group hospitals. Haas-Wilson, Tr. 2620, in camera. 733. The difference in differences model reported the actual percentage point price difference between the price increases at ENH and at comparison hospitals. So, the regression model reports the number of percentage points by which the prices at ENH exceed the comparison hospitals. Haas-Wilson, Tr. 2621, in camera. 734. Haas-Wilson used two data sources for her regression model: (1) the IDPH Universal Dataset in conjunction with the Medicare Cost Reports, and (2) the managed care claims data. Haas- Wilson, Tr. 2621-22, in camera. All of the results show that the postmerger price increases at ENH were greater than the average price increases at comparison hospitals, even taking into account variations in patient mix, customer mix, and teaching intensity. Haas-Wilson, Tr. 2631-35, in camera; see CX 6279 at 20, in camera. (ii) Changes in Patient Mix 735. Not all inpatient hospital stays require the same resources to treat. Patients with more complex conditions may require more resources than patients with less complex conditions. For two patients with the same condition, one may be sicker, requiring more resources to treat than the patient who is less sick. Haas-Wilson, Tr. 2485.

736. The mix of patients that a hospital has will influence the hospital’s prices. If the hospital has patients who require more resources to treat than other hospitals, that will impact the hospital’s prices. Haas-Wilson, Tr. 2486.

VOLUME 144 Initial Decision 737. If a hospital’s mix of patients is changing, such that the hospital is getting more complex cases or the patients are arriving sicker, one would expect that the hospital would be using more resources to treat those patients, and that would be a possible explanation for a price increase. Haas-Wilson, Tr. 2589, in camera. 738. If case mix or severity of illness is changing similarly across hospitals, it can not be an explanation of a relatively larger price increase at one hospital versus another. But if the mix of patients is changing over time across hospitals differently, then case mix or severity of illness could be a possible explanation of a higher price increase at one hospital versus another. Haas-Wilson, Tr. 2589- 90, in camera.

739. The case mix index is used by many people who analyze hospital data, and it is a measure of the complexity of the cases that are being treated at particular hospitals. It is constructed based on a system of weights related to diagnostic related groups (“DRG”). Haas-Wilson, Tr. 2594, in camera.

740. CX 6279 at 13 is a comparison of the post-merger change in case mix at ENH and at control hospitals using the managed care claims data. Haas-Wilson, Tr. 2592-93, in camera; CX 6279 at 13, in camera.

741. The managed care claims data suggested the patient mix was changing at ENH after the merger with Highland Park in a manner that may explain, at least in part, price increases at ENH. Haas-Wilson, Tr. 2590, 2595-96, in camera. Haas-Wilson used multiple regression to test the extent to which changing patient mix explains the price increases at ENH. Haas-Wilson, Tr. 2619-20, in camera.

742. CX 6279 at 14 is a comparison of the post-merger change in in case mix at ENH and at control hospitals, using the IDPH EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 163 Initial Decision Universal Dataset. Haas-Wilson, Tr. 2596-98, in camera; CX 6279 6279 at 14, in camera.

743. The IDPH Universal Dataset suggested the patient mix was changing at ENH after the merger with Highland Park in a manner that may explain, at least in part, the price increases at ENH. Haas- Wilson, Tr. 2598-99, in camera. Haas-Wilson used multiple regression to test the extent to which changing patient mix explains the price increases at ENH. Haas-Wilson, Tr. 2619-20, in camera. (iii) Changes in Mix of Customers 744. Mix of customers refers to the different types of organizations that pay for patients at a hospital, whether it is commercial insurance or public health insurance programs, such as the Medicare and Medicaid programs. Haas-Wilson, Tr. 2486. 745. If a hospital has more Medicare and Medicaid patients, that could provide a motivation for the hospital to raise its prices to patients of the managed care organizations, especially when payment under the public programs is reduced. Haas-Wilson, Tr. 2486.

746. Haas-Wilson used the Medicare Cost Reports data showing the percentage of patients receiving care at the hospital that are covered by Medicaid or Medicare. She used the percent of patients covered by Medicare or Medicaid as the measure of the mix of customers. Haas-Wilson, Tr. 2600, in camera. 747. Haas-Wilson tested the hypothesis that the change in the mix of customers at ENH and the change in the mix of customers at comparison hospitals over the relevant time period was the same. Haas-Wilson, Tr. 2600, in camera.

748. Haas-Wilson found that there were differences in the way the mix of customers was changing over time across hospitals. As a VOLUME 144 Initial Decision a result, she could not, at that point in her analysis, eliminate changes in the mix of customers as a possible explanation for the price increases at ENH. Haas-Wilson, Tr. 2600, 2602-03, in camera. camera. Haas-Wilson used multiple regression to test the extent to which changing customer mix explains the price increases at ENH. ENH. Haas-Wilson, Tr. 2619-20, in camera. (iv) Changes in Teaching Intensity 749. Teaching intensity is a measure of how much teaching activity is occurring at a hospital. Some hospitals participate in the training of residents and interns. Haas-Wilson, Tr. 2486-87. 750. There is empirical support for the proposition that hospitals that are involved in teaching activity have higher costs than hospitals that are not involved in teaching activity. Haas-Wilson, Tr. 2487. Therefore, those hospitals involved in more teaching may have higher costs than those involved with lesser amounts of teaching activity. Haas-Wilson, Tr. 2487.

751. Haas-Wilson tested the hypothesis that changes in teaching intensity at ENH over the relevant time period were the same as the changes in teaching intensity over the same time period at comparison hospitals. Haas-Wilson, Tr. 2603-04, in camera. 752. Teaching intensity was measured as the number of residents and interns per hospital bed at each hospital. Haas-Wilson, Tr. 2604, in camera.

753. Haas-Wilson included any hospital that had at least one intern or one resident. Haas-Wilson, Tr. 2869-70, in camera. Haas- Wilson used data from the Medicare Cost Reports to test the hypothesis regarding changes in teaching intensity. Haas-Wilson, Tr. 2604, in camera.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 165 Initial Decision 754. Haas-Wilson found that teaching intensity was changing across hospitals differently over time. As a result, she could not, without further analysis, eliminate changes in teaching intensity as a potential explanation for the price increases at ENH. Haas-Wilson, Tr. 2603-04, 2606, in camera. Haas-Wilson used multiple regression to test the extent to which changing teaching intensity explains the price increases at ENH. Haas-Wilson, Tr. 2619-20, in camera. 755. All of the results show that the post-merger price increases at ENH were greater than the average price increases at comparison hospitals, even taking into account variations in patient mix, customer mix, and teaching intensity. Haas-Wilson, Tr. 2631-35, in camera; see CX 6279 at 20, in camera.

2. Procompetitive Justifications a. Learning About Demand (1) Foundations for the Theory 756. During the due diligence work connected with the merger, Evanston learned about Highland Park’s managed care contracts and learned about Highland Park’s pricing information. Noether, Tr. 5973-74; Chan, Tr. 660-63, 711-12; Chan, Tr. 825, in camera; RX 620 at ENHL TC 17809, in camera; RX 652 at BAIN 9. 757. According to Noether, the learning about demand explanation is that before the merger with Highland Park, Evanston had poor information about the true demand for its services. Noether, Tr. 5968. Noether agreed, however, that a hospital merger could lead to market power at the same time the hospital learns more about demand for its services. Noether, Tr. 6142. VOLUME 144 Initial Decision 758. Haas-Wilson testified that the “empirical literature . . . suggests that costs and therefore prices ‘might’ be different at hospitals that are engaged in ‘teaching activity’ versus those that are not.” Haas-Wilson, Tr. 2550, in camera.

759. Respondent’s experts testified that premerger, Evanston priced itself more like a community hospital, rather than a major teaching hospital. Noether, Tr. 5968; Baker, Tr. 4654-55, in camera. 760. Respondent’s experts’ learning about demand theory proposes that once Evanston learned about the demand for its services, it modified its pricing to reflect this greater understanding and to price itself more like a teaching hospital. Noether, Tr. 5968- 69; Baker, Tr. 4654-55, in camera.

(2) ENH’s Contract Negotiations in the 1990's 761. Jack Sirabian handled Evanston’s managed care contracting negotiations from approximately 1990 to 2000. Sirabian, Tr. 5697- 98. Sirabian reported to Hillebrand with respect to managed care contracting. Sirabian, Tr. 5728-29; Hillebrand, Tr. 1700. 762. During the period in which Sirabian was responsible for contracting, he received positive evaluations from both Neaman and Hillebrand for his work at ENH. Sirabian, Tr. 5728. 763. When Bain provided contract negotiation advice in 1999 to Evanston, neither Bain nor Evanston management informed Sirabian that any of Evanston’s rates that were perceived to be unfavorable were the result of Sirabian’s poor contract negotiations in the 1990's. Sirabian, Tr. 5762.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 167 Initial Decision 764. Bain advised ENH that it “should recognize its position and not be afraid to ask to be paid fair market value” for its services. RX 2047 at 39-40 (Ogden, Dep.).

765. Sirabian received a bonus after the merger in 2000. Neaman, Tr. 1265-66; CX 31 at 1.

766. Hillebrand had and continues to have general oversight and supervisory responsibility for managed care contracting. Hillebrand, Tr. 1701-02; Neaman, Tr. 1220.

767. Hillebrand testified that Evanston’s negotiating stance with managed care organizations was equally aggressive before and after the merger. Hillebrand, Tr. 1731, 1733.

768. ENH’s CEO believes Hillebrand to be an effective negotiator, with a good understanding of the marketplace and ENH’s relationships with managed care organizations. The CEO never criticized Hillebrand about ENH’s premerger managed care contracts. Neaman, Tr. 1220.

769. Hillebrand was never accused of being soft or of not bargaining hard with managed care organizations. Hillebrand, Tr. 1727.

770. Hillebrand received a bonus after the merger in 2000. Neaman, Tr. 1221.

771. After the merger, Theresa Chan, who had negotiated contracts with managed care organizations on behalf of Highland Park, resigned and was not asked to remain. Hillebrand, Tr. 1730, 2044.

(3) Testimony of Managed Care Organizations 772. In its contract negotiations, ENH did not indicate to managed care organizations that ENH was attempting to match VOLUME 144 Initial Decision academic teaching hospitals’ pricing. Ballengee, Tr. 193-94 (PHCS); (PHCS); Neary, Tr. 621; Dorsey, Tr. 1447 (One Health). (a) One Health 773. In negotiating with hospitals to be in its network, One Health makes judgments about the hospitals’ level of services. Neary, Tr. 622.

774. One Health views academic teaching hospitals as teaching facilities that train physicians and as institutions that are part of a medical school. Such hospitals are on the cutting edge of medical technology, performing services that other general acute care facilities and community hospitals do not perform, such as transplant services, burn units, and higher levels of cardiac services. Neary, Tr. 622; Dorsey, Tr. 1443.

775. One Health believes academic hospitals in the Chicago area are: the University of Chicago, Rush-Presbyterian-St. Luke’s, Northwestern Memorial, Loyola University, and University of Illinois. Dorsey, Tr. 1443-44; Neary, Tr. 623. 776. One Health does not view any of the hospitals in ENH (Evanston, Glenbrook, and Highland Park) as academic teaching hospitals. Neary, Tr. 621; Dorsey, Tr. 1444. (b) PHCS 777. PHCS categorizes hospitals as community, tertiary, and advanced teaching hospitals. Advanced teaching hospitals offered the really high-level procedures, such as transplants, burn units, and hyperbaric centers. Ballengee, Tr. 159.

778. Premerger, PHCS viewed Highland Park as in the community hospital group and Evanston as a community and tertiary tertiary hospital, spanning both groups. Post-merger, PHCS EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 169 Initial Decision continued to view ENH as both a community and tertiary hospital. Ballengee, Tr. 158-59.

779. PHCS views the advanced teaching hospitals in the Chicago area as Northwestern Memorial, Rush-Presbyterian-St. Luke’s, University of Chicago, Loyola University, and University of Illinois. Ballengee, Tr. 189.

780. PHCS does not view ENH as an advanced teaching hospital. Ballengee, Tr. 189.

(c) United 781. United views an academic hospital as one that has a medical school as part of the hospital. Foucre, Tr. 935. 782. United believes Loyola University, University of Chicago, Northwestern Memorial, and Rush-Presbyterian-St. Luke’s are all academic hospitals. Foucre, Tr. 936.

783. United does not believe that Evanston, Glenbrook, and Highland Park are academic hospitals. Foucre, Tr. 936. (4) Evanston Could Not Have Learned Anything Significant About Demand from Highland Park (a) Differences Between Highland Park and Evanston 784. Evanston and Highland Park were different in a number of dimensions. Premerger, Highland Park was a community hospital, and Evanston had elements of both a community and tertiary hospital. Ballengee, Tr. 159.

785. Evanston offered a number of services that Highland Park Park did not. While Evanston and Highland Park offered many of VOLUME 144 Initial Decision the same services, about 11.6% of the patients at Evanston in 1999 1999 were being treated for DRGs for which Highland Park did not not treat four or more patients in a year. RX 1912 at 44, in camera. camera.

786. Evanston Hospital/ENH has been named by one publication as a top 15 teaching hospital and a top 100 hospital in the country. Neaman, Tr. 1197, 1290-91.

(b) Premerger, Highland Park Charged Lower Actual Prices Than Evanston 787. Sirabian testified that in approximately one third of thirtyfive or forty contracts with managed care organizations, Highland Park had higher contract rates than Evanston. Sirabian, Tr. 5717. 788. The negotiated rates that one observes in contracts typically are not the actual prices that health plans would pay to hospitals. Haas-Wilson, Tr. 2645, in camera.

789. Rates are just one factor that goes into determining prices. There are multiple factors in hospital contracts that determine the actual price or the reimbursement per case. Haas-Wilson, Tr. 2647, in camera.

790. In addition to per diem rates, contracts also specify stop loss provisions, which specify at what point the per diem no longer applies and instead the hospital gets reimbursed on a different basis specified in the contract. Haas-Wilson, Tr. 2647, in camera. 791. The contract itself also shows nothing about the hospital’s hospital’s chargemaster. Thus, if two hospitals have contracts that specify a 10% discount off charges, without knowing the respective respective chargemasters, knowing the discount off charges does not EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 171 Initial Decision not show which hospital had higher prices. Haas-Wilson, Tr. 2647- 2647-48, in camera.

792. The hospital with the higher negotiated rates is not necessarily the hospital with the higher prices. Haas-Wilson, Tr. 2645, in camera.

793. Evanston’s chargemaster was higher than Highland Park’s premerger. Chan, Tr. 743. See also CX 1373 at 14, in camera; RX 620, in camera (“The same contract terms that may be more favorable to [Highland Park] based on [Highland Park’s] charge data may turn out to be less favorable to ENH if rates were to apply to ENH’s charge data.”).

794. For each of the four managed care organizations that were covered in Noether’s back-up materials, the prices at Evanston were higher than the prices at Highland Park. Haas-Wilson, Tr. 2646, in camera.

795. Baker calculated the percentage price increase following the merger for four health plans, Aetna, Blue Cross, Humana, and United. He did the calculations in two ways: (1) comparing Evanston and Glenbrook’s premerger prices to the ENH post-merger prices; and (2) comparing Evanston, Glenbrook, and Highland Park’s combined premerger prices (Baker’s “constructed prices”) to the ENH post-merger prices. Baker, Tr. 4633, in camera. 796. When Baker’s constructed price (which includes the premerger prices at Highland Park) showed a larger price increase than his calculation for the price increase for just Evanston and Glenbrook, that necessarily means that the prices at Highland Park were lower than the prices at Evanston and Glenbrook premerger. See Baker, Tr. 4744-46, in camera.

797. Baker testified that looking at the prices actually paid by Aetna, Humana, and Blue Cross Blue Shield for inpatient services, VOLUME 144 Initial Decision services, the actual prices paid by those managed care organizations organizations to Highland Park were lower than the prices paid to Evanston in the premerger period. Baker, Tr. 4744-47, in camera. (5) Noether’s Control Groups Are Flawed 798. Noether looked at price levels and relied on a comparison of the price levels at ENH with the price levels of several major teaching hospitals in the Chicago area and with the price levels of community hospitals. Noether, Tr. 5991-92, 6000. 799. Noether drew conclusions about the manner in which ENH’s prices increased above the prices of her selected community hospitals toward the prices of her selected academic hospitals. See Noether, Tr. 6060, in camera.

800. The comparisons performed by Noether depend upon the hospitals that Noether selected for her two groups of hospitals. If the control group selected by Noether is not appropriate, the analysis using that control group could lead to a biased result. Haas-Wilson, Tr. 2697, in camera.

(a) Noether Began with an Arbitrary Group of Twenty Hospitals 801. Noether began with her list of twenty hospitals (eighteen plus Evanston and Highland Park) to develop what she called her academic hospital and community hospital control groups. Noether, Tr. 6154-55.

802. Noether testified that to determine the list of twenty hospitals, she selected hospitals after she “reviewed the evidence from a variety of sources in the record and developed a list based on [her] analysis of the information,” including hospitals which Noether testified were “in some way competitors to Evanston and/or Highland Park.” Noether, Tr. 5913-14, 6149-50. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 173 Initial Decision 803. There were no specific criteria or journal articles in economic literature used by Noether to decide which hospitals to include on her list of hospitals. Noether, Tr. 6149-50. 804. There was no single document that listed the hospitals as competitors. Noether made the decisions to pick and choose which hospitals she would include. Noether, Tr. 6149. 805. Noether’s academic hospital control group consists of six hospitals, in addition to Evanston: Advocate Lutheran General, Advocate Northside, Northwestern Memorial, Rush-Presbyterian-St. Luke’s, Loyola, and University of Chicago. Noether, Tr. 6000; RX 1912 at 60.

806. Noether’s community hospital group consists of twelve hospitals, in addition to Highland Park: Alexian Brothers, Louis A. Weiss, Northwest Community, Resurrection, St. Francis, Rush North Shore, Condell, Holy Family, Lake Forest, Swedish Covenant, Vista Health Saint Therese, and Vista Health Victory Memorial. Noether, Tr. 6000; RX 1912 at 60.

(b) Noether’s Division of Her List of Hospitals into an Academic Hospital Group and a Community Hospital Group Is Arbitrary 807. There is no official government designation of what hospitals are community hospitals or academic hospitals. Noether, Tr. 6155.

808. Noether used three criteria to select which of the twenty hospitals to include in her academic control group: teaching intensity (rate of residents to beds), number of staffed beds, and breadth of services (number of Diagnosis Related Groups (“DRGs”)). Noether, Tr. 5993-95.

VOLUME 144 Initial Decision 809. Medicare Payment Advisory Commission (“MedPAC”) is an advisory body to Congress on Medicare reimbursement criteria. MedPAC defines a major teaching hospital as a hospital with at least .25 residents per bed. Noether, Tr. 5995. 810. The MedPAC criteria for classification as a major teaching hospital have nothing to do with the number of DRGs that a hospital offers. Noether, Tr. 6155.

811. In determining the number of DRGs to use as a criterion to include hospitals in her academic control group, Noether counted a hospital as offering a DRG only if the hospital offered it four or more times in a year, an arbitrary cut-off. Noether, Tr. 5914-15. 812. Using Noether’s criterion of four cases, even a change from looking at a fiscal year as opposed to looking at a calendar year can cause the number of DRGs that Noether counts to change. For example, in fiscal year 1999 Highland Park was found to offer 208 DRGs, but in calendar year 1999 Highland Park was found offering 212 DRGs. RX 1912 at 44, in camera; RX 1912 at 60. 813. Noether listed the hospitals in order of the number of DRGs that they offered, and took the top third of the hospitals as having enough DRGs to be classified as academic hospitals, so that she only included hospitals with more than 370 DRGs. Noether, Tr. 6164-65. 814. There is no basis in the health care literature to require a hospital to be above a certain number of DRGs in order to be considered an academic hospital. Noether, Tr. 6165-66. 815. Only after considering evidence describing the different hospitals on her list and after looking over the list of hospitals, did Noether decide to include the top third, instead of the top quarter or EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 175 Initial Decision or top half of the hospitals as having enough DRGs to be included as as an academic hospital. Noether, Tr. 6166-67. 816. The last hospital to be included as having enough DRGs to be considered as an academic hospital was Rush-Presbyterian-St. Luke’s. Noether, Tr. 6167-68. Rush-Presbyterian-St. Luke’s is one of the four highest priced hospitals in Noether’s list of twenty hospitals. See RX 1912 at 147-52, in camera.

817. Similarly, the MedPAC criteria defining a major teaching hospital do not rely on size. Noether, Tr. 6155. All of the hospitals in Noether’s academic control group have more beds than ENH, some of them, significantly more (e.g. Advocate Northside, with 663 beds). Haas-Wilson, Tr. 2708-09, in camera; RX 1912 at 60. See F. 829.

818. Noether’s academic group included four of the most expensive hospitals in Chicago: { } RX 1912 at 147-52, in camera (Average Reimbursement per Case) RX 1912 at 147-49, in camera and (Average Reimbursement per Case, Excluding Obstetrics); RX 1912 at 150-52, in camera. Noether’s academic group of hospitals are priced higher than her community group of hospitals. RX 1912 at 60; RX 1912 at 147-52, in camera.

819. Noether’s academic control group excluded less expensive hospitals even though many of those excluded can handle most of the patients Evanston treated and treat more complex cases than ENH. See RX 1912 at 60; RX 1912 at 147-52, in camera. VOLUME 144 Initial Decision (c) Noether’s Academic Control Group Is Not Not an Appropriate Control Group 820. Noether’s academic control group is not an appropriate control group from the scientific perspective. Haas-Wilson, Tr. 2698, in camera.

(i) Case Mix and Services Provided 821. There is a difference between the case mix of four of the six hospitals included by Noether in her academic control group and the case mix at ENH.{ } all have case mix indexes that are much higher than ENH’s case mix index. Haas-Wilson, Tr. 2698-2700, in camera.

822. Quaternary services are different from other inpatient hospital services. These services, which include solid organ transplants and treatment for severe burns, require very specific human capital, specially trained nurses and doctors and very specialized physical capital. Haas-Wilson, Tr. 2701-02, in camera. 823. ENH differed from the hospitals in Noether’s academic control group in terms of quaternary services. ENH provides no solid organ transplants and no extensive burn cases, while four of the six hospitals in Noether’s academic control group offer solid organ transplants, and two of the six hospitals treat extensive burn injuries. Haas-Wilson, Tr. 2702, in camera; CX 6282 at 7-8, in camera; Neaman, Tr. 1378.

824. Each of the hospitals in Noether’s academic control group offers a broader range of services than ENH. The hospitals in Noether’s academic control group offer the following number of DRGs that ENH does not offer: { } RX 1912 at 44, in camera. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 177 Initial Decision 825. Noether excluded from her academic control group some hospitals that treated, on average, more complex cases than ENH, including: { } Haas-Wilson, Tr. 2594, in camera; Noether, Tr. 6168-72; RX 1912 at 25.

(ii) Teaching Intensity 826. Teaching intensity, as measured by the number of interns and residents per bed, is one way to see which hospitals are comparable to ENH. Haas-Wilson, Tr. 2708, in camera. 827. Four of the six other hospitals that Noether has in her academic control group have significantly more residents per bed than Evanston. Evanston has .3386 residents per bed, while Loyola University has .6060 residents per bed, Northwestern Memorial has .5670 residents per bed, Rush-Presbyterian-St. Luke’s has .7606 residents per bed, and University of Chicago has .7938 residents per bed. RX 1912 at 60.

828. The combined ENH has .29 residents per bed. O’Brien, Tr. 3542.

829. Size, in terms of number of beds, is a characteristic that one could use to compare other hospitals to ENH to see if they are similar. All of the hospitals in Noether’s academic control group have more beds than ENH, some of them, significantly more (e.g. Advocate Northside, with 663 beds). Haas-Wilson, Tr. 2708-09, in camera; RX 1912 at 60.

830. Noether excluded from her academic control group some hospitals that meet MedPAC’s definition of a teaching hospital (more than .25 residents per bed), including: Louis A. Weiss and St. Francis. Norther, Tr. 6170; RX 1912 at 60. VOLUME 144 Initial Decision (d) ENH Compared to Noether’s Proposed Geographic Market 831. Noether compared ENH’s prices to prices charged by other hospitals. Noether, Tr. 5992-93, 6000; RX 1912 at 146-50, in camera. { } RX 1912 at 147-52, in camera. { } RX 1912 at 147- 52, in camera.

832. { } RX 1912 at 148, 151, in camera. 833. { } RX 1912 at 147, in camera.

834. { } RX 1912 at 149, 152, in camera. 835. { } RX 1912 at 149, 152, in camera. 836. { } RX 1912 at 149, 152, in camera. 837. { } RX 1912 at 148, 151, in camera. b. Quality of Care (1) Price Increases to Managed Care Were Not Related to Improvements at Highland Park 838. The economic testimony in this case appears to view quality as part of the cost/price continuum. Baker testified that “quality improvements need to be considered in evaluating competitive effects because if quality gets better, the quality-adjusted price to the buyers declines.” Baker, Tr. 4604. Baker agreed that there is no need to adjust for quality of care if quality of care is changing at the same rate as other hospitals. Baker, Tr. 4799, in camera. 839. Haas-Wilson testified that “[i]f quality is increasing in general, that would lead to potentially higher prices at all hospitals, EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 179 Initial Decision hospitals, and if quality is increasing more at one hospital than at others, then that could potentially explain a greater price increase at at one hospital over others in the case where the buyers of hospital hospital services value that quality enhancement.” Haas-Wilson, Tr. Tr. 2484-85.

840. ENH did not justify its price increases to managed care based on improvements being made at Highland Park. F. 842-47. 841. Respondent did not present an explanation of how to value the “improvements” or how to compare them to the price increases. Chassin, Tr. 5447-48.

842. ENH’s COO, Hillebrand, admitted that he did not tell managed care organizations that the higher prices were justified by quality changes. Hillebrand, Tr. 1784.

843. ENH’s CEO, Neaman, admitted that he never saw any documents correlating the higher prices with the quality changes at Highland Park. Neaman, Tr. 1241-42.

844. The One Health representative testified that the topic of quality changes simply never came up during negotiations. E.g., Neary, Tr. 624.

845. The PHCS representative testified that even after implementing the changes, ENH did not advertise them to managed care organizations. Ballengee, Tr. 188, 200-03. 846. The PHCS representative testified that Highland Park’s quality of care has remained the same from before the merger to after the merger. Ballengee, Tr. 187.

847. The United representative testified that she had not been shown any evidence that the quality of care improved at Highland Park. Foucre, Tr. 926-27.

VOLUME 144 Initial Decision 848. Simultaneous with the execution of the Letter of Intent, on June 30, 1999, Evanston and Highland Park sent a press release to managed care organizations, area employers, elected officials, and the press describing the goals of the merger: “The merger will result in significant additional investments in clinical services at the Highland Park Hospital campus . . . . Our intent is to strengthen Highland Park Hospital’s capabilities in key clinical growth areas such as oncology, cardiac services, obstetrics, fertility, home health, behavioral health,” and listed specific projects such as the Kellogg Cancer Care Center. RX 563 at ENH TH 1568-76; Hillebrand, Tr. 1857-58.

849. The PHCS representative testified that managed care organizations will pay more to select hospitals that offer more complex services and with reputations for higher quality. Ballengee, Tr. 163-64.

850. Highland Park management and outside observers believed that the quality of care of Highland Park was “very good, if not excellent” at the time of the merger. Newton, Tr. 376. 851. Highland Park was also described as a “pretty good community hospital” that “delivers basic services at a very high level” and was perceived as an “excellent community hospital.” Neaman, Tr. 1306; Spaeth, Tr. 2098; CX 1868 at 7, 10. 852. Evanston and Highland Park “were both very good hospitals.” Ballengee, Tr. 160.

(2) No Evidence of Improvement in Overall Quality of Care Relative to Other Hospitals 853. In 1999, the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) evaluated hospitals including Highland Park and Evanston. See Spaeth, Tr. 2148-49. In 1999, Highland Park received a preliminary score of 95 and a final score EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 181 Initial Decision of 96 out of 100. Spaeth, Tr. 2122, 2148-49; CX 96 at 1; CX 2304 at at 3; RX 412 at ENHL PK 17794, in camera. In 2002, Highland Park received a score of 94 as part of ENH’s JCAHO survey. RX 1380 at ENH JH 11480.

854. Evanston received a preliminary score of 94 in 1999 and a final score of 95 in 2000. Neaman, Tr. 1198, 1231; CX 871 at 4; CX 6 at 5; RX 1380 at ENH JH 11480.

855. Most hospitals in this country use JCAHO scores to look at quality of care. Spaeth, Tr. 2154.

856. JCAHO scores are based on about 1200 very specific aspects of hospital activities that are called elements of performance. Chassin, Tr. 5156-57.

857. JCAHO is the nationally accepted norm for healthcare accreditation organizations. Other quality measurement tools are in their infancy and not viable options for managed care organizations to compare hospital quality. Ballangee, Tr. 186-87. 858. JCAHO accreditation is necessary to qualify for Medicare and many managed care plans. Ballengee, Tr. 151; Newton, Tr. 385. 859. Complaint Counsel’s expert testified that starting in the late 1990’s, there has been a nationwide trend of improved quality, with one major study finding an average per state inpatient improvement rate of 12% through 2001. Romano, Tr. 3000-01. Other studies also show that hospitals were improving their quality during the time from 1997 through 2004. Romano, Tr. 2999-3000; see also Noether, Tr. 6011.

860. The U.S. Agency for Healthcare Research and Quality (AHRQ) is the lead federal agency that is responsible for developing and promoting methods for quality of care research in the United States. Romano, Tr. 2969.

VOLUME 144 Initial Decision 861. Complaint Counsel’s expert, Romano, using AHRQ measures found { } at Highland Park relative to a control group. Romano, Tr. 3093-95, 3210-12, in camera; see also DX 7034A at 1. 862. Using the JCAHO measure, Romano found evidence of { } at Highland Park, although that evidence was not statistically significant. Romano, Tr. 3217, in camera; see also DX 7034A at 2. 863. In obstetrics, using the AHRQ measures, Romano found evidence of { } under the JCAHO measures. Romano, Tr. 3226-29, in camera; see also DX 7034A at 10-11.

864. JCAHO measure uses the more typical kind of risk adjustment process, which is logistic regression while the AHRQ measure uses a cruder risk adjustment based on DRGs. Chassin, Tr. 5184.

865. Press Ganey does survey work in hospitals regarding patients perception of their care, in the form of patient satisfaction surveys. Neaman, Tr. 1227; Romano, Tr. 2982-83; Romano, Tr. 3098, in camera.

866. Many of the Press Ganey questions concern amenities. Spaeth, Tr. 2093-94; Romano, Tr. 3339-40, 3342. 867. The response rate of the Press Ganey data is unclear. Romano, Tr. 3344-46. Respondent’s healthcare quality expert, Dr. Mark R. Chassin, Edmond A. Guggenheim Professor of Health Policy, Chairman of the Department of Health Policy of the Mount Sinai Medical School, made a rough estimate that the response rate was about twenty percent, which Romano admitted would be suboptimal. Romano, Tr. 3346; Chassin, Tr. 5244. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 183 Initial Decision 868. Complaint Counsel’s expert, Romano, was not aware of the Press Ganey survey methodology. Romano, Tr. 3344-45. (3) ENH’s Non-Merger Specific Improvements to Highland Park 869. In its 1998 Strategic Plan for Highland Park 1999-2002, Highland Park’s parent company, planned to: maximize the Northwestern Healthcare affiliation; implement a cardiovascular surgery program; implement a comprehensive oncology program; recruit physician specialists; enhance physician leadership throughout the organization with improved communication forums; promote a work environment that facilitates strong associate relations, open communication, teamwork, involvement, and standards of excellence; and improve workflow and scheduling systems in all departments with particular focus on radiology, cardiology, laboratory, and physical medicine to increase patient satisfaction. CX 1868 at 12-14, 17.

870. Recognizing the need to improve quality, on March 23, 1999, Lakeland’s strategic plan for 1999-2003, included among its goals to: enhance its core clinical competencies (cardiac surgery, oncology and specialty surgery); implement a cardiovascular surgery program; implement a comprehensive oncology program; identify and promote selected physician clinical leaders and enhance physician leadership; provide documented and measurable outcomes of quality which exceed those of the competition and establish national standards and provide a continuum of care for the patient across the delivery system including providing the highest quality clinical and non-clinical services; utilize the latest technology to support patient care; and promote a work environment that facilitates strong associate relations, open communication, teamwork, involvement, and standards of excellence to achieve success. CX 1908 at 9, 12-13, 18-20, 23. VOLUME 144 Initial Decision 871. Highland Park recognized the need for improvements as early as 1998 and in March of 1999, before the merger, outlined a strategic plan to improve its quality of care. CX 1868; CX 1908. 872. In March 1999, Highland Park’s finance committee approved more than $100 million for new projects through 2003. CX 1055 at 2; CX 1903 at 2-3; CX 545 at 3.

873. Highland Park’s long range capital budget identified $43 million for investment in strategic initiatives and master plan items such as cardiology services, ambulatory services, oncology, assisted living, and facility expansion and $65 million for hospital construction, routine capital, and information technology. CX 545 at 3.

874. The finance committee concluded that based on growth through new clinical services and existing cash and investments and cash flow, the hospital could “generate sufficient cash” to “restore the profitability” of Highland Park and fund the improvements. CX 1903 at 1; CX 545 at 4.

875. Prior to the merger, Highland Park always had the latest piece of equipment and if it needed to invest in new technology, it made those routine investments and purchased new technology. Newton, Tr. 384.

(a) Obstetrics and Gynecological Services 876. At the time of the merger the Obstetrics and Gynecological (“Ob/Gyn”) department was the largest patient care area at Highland Park. Chassin, Tr. 5196 877. ENH instituted nighttime and weekend coverage by obstetricians in Highland Park’s Ob/Gyn department. Chassin, Tr. 5204; Silver, Tr. 3779-80, 3783-84.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 185 Initial Decision 878. ENH installed a full-time chair for the Ob/Gyn department in the Spring of 2001. Chassin, Tr. 5204-05; Silver, Tr. 3841. 879. Nurse training models of care were improved. This process began before the merger and continued after the merger. Chassin, Tr. 5205.

880. ENH provided multidisciplinary clinical care at Highland Park, so that doctors, nurses, and all of the participants in the obstetric services worked together as a team. Chassin, Tr. 5206-07. 881. ENH instituted an Ob/Gyn preoperative surgery review program at Highland Park. Chassin, Tr. 5206; Silver, Tr. 3780-81. 882. ENH instituted physician discipline against a few of Highland Park’s Ob/Gyn physicians. Chassin, Tr. 5206-07; Silver, Tr. 3882-83, 3886, in camera.

883. Prior to the merger, Highland Park had invited the { } review of the hospital as part of its ongoing effort to improve quality of care. Romano, Tr. 3152-54, in camera; Spaeth, Tr. 2114-15; Chassin, Tr. 5498; RX 324 at ENHL PK 29688-89, in camera. 884. { } made a number of recommendations to improve the { } Romano, Tr. 3154-55, in camera; RX 324 at ENHL PK 29689, in camera.

885. Many changes were made in reaction to the { }, including the hiring of a { } in 1998. Romano, Tr. 3155, in camera; Spaeth, Tr. 2114-15; CX 98 at 2.

886. Highland Park’s efforts to implement { } recommendations were subsequently recognized by the Chicago Hospital Risk Pooling Program after a site visit and report issued in VOLUME 144 Initial Decision in November of 1999. Romano, Tr. 3155-58, in camera; CX 6265, in in camera.

887. The Chicago Hospital Risk Pooling Program made additional recommendations for improvement. CX 6265 at 17-30, in camera.

(b) Quality Assurance Programs 888. ENH changed the structure within the clinical departments of how oversight of physicians was conducted by replacing part-time and private practice chairs with full-time ENH clinician chairs. Chassin, Tr. 5211, 5224-25; Spaeth, Tr. 2253-54. 889. ENH took disciplinary action against a number of Highland Park physicians. Chassin, Tr. 5225-26.

890. ENH reviewed physician practices during periodic recredentialing. Chassin, Tr. 5226-27.

891. There were post-merger changes made in error reporting and adverse events reporting, although these changes took a fair amount of time to play out. Chassin, Tr. 5227-29. 892. Highland Park, premerger, had regularly initiated disciplinary actions against its physicians, including suspension, reduction, or removal of staff privileges. Newton, Tr. 382-83. 893. There are a number of examples of Highland Park’s review of adverse events prior to the merger. Chassin, Tr. 5514; RX 251 at ENHL PK 17839, in camera; RX 346 at ENHL PK 24708-11, in camera; CX 6296 at 3-6, in camera.

894. The { } was requested because of an adverse event in the { } Krasner, Tr. 3733-34.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 187 Initial Decision 895. The quality assurance improvements made by ENH at Highland Park after the merger reflect an emerging consensus in the field of quality assurance. Romano, Tr. 3159, in camera. (c) Quality Improvement Programs 896. After the merger, the critical pathways at ENH were aligned with the care maps being used at Highland Park, improving both. O’Brien, Tr. 3559-60; Chassin, Tr. 5257; CX 6286 at 4 (King, Dep.). 897. Critical pathways and care maps are protocols identifying the best practices for the treatment of patients. Romano, Tr. 3167-68, in camera; Silver, Tr. 3803-04.

898. Prior to the merger, Highland Park conducted an internal review of quality programs which highlighted areas for improvement. Chassin, Tr. 5256; RX 417.

899. Nothing in the record suggests that ENH’s critical pathways were better than the care maps used by Highland Park before the merger or that Highland Park would not have continued to develop other care maps after 1999 on its own. Silver, Tr. 3839; Romano, Tr. 3170-71, in camera.

900. The evidence does not clearly show whether the critical pathways are always being followed. Romano, Tr. 3170, in camera. 901. Critical pathways are always being revised and improved. O’Brien, Tr. 3561-62.

902. The quality improvements made by ENH at Highland Park after the merger reflect an emerging consensus in the field of quality improvements. Romano, Tr. 3159, in camera. (d) Nursing Staff VOLUME 144 Initial Decision 903. ENH improved communication and teamwork between nurses and physicians. Chassin, Tr. 5239-40. 904. After the merger, nurse training improved, some nurses received training at ENH, and nurse managers were rotated through all ENH hospitals. Chassin, Tr. 5239; O’Brien, Tr. 3535; Krasner, Tr. 3725-26; RX 1445 at ENHL PK 51620.

905. ENH eventually improved recruiting, vacancy, and turnover rates. RX 1445; O’Brien, Tr. 3671-72, in camera; Krasner, Tr. 3722- 24.

906. Highland Park had intergenerational nursing where grandmothers, mothers, and daughters were all nurses at the hospital. Newton, Tr. 383.

907. Highland Park had a “high quality nursing staff” in the 1990’s. Newton, Tr. 383.

908. In 1999, Highland Park adopted a comprehensive initiative to train, retain, and reward its nurses. CX 1908 at 23; CX 6264 at 1; Krasner, Tr. 3721; Newton, Tr. 410-11.

909. The nursing culture at Highland Park underwent a transition from a punitive and dysfunctional culture to a much more effective culture over a period of years beginning before the merger and continuing until 2004. Chassin, Tr. 5239, 5478-79; O’Brien, Tr. 3536-37.

910. The improvements to the nursing culture was an evolutionary process that took many years. Chassin, Tr. 5478-79. (e) Physical Plant 911. ENH built a new ambulatory care center which opened in February 2005, and which houses radiation medicine, nuclear EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 189 Initial Decision medicine, the Kellogg Cancer Care Center, and a new breast imaging Center. O’Brien, Tr. 3497-98; Chassin, Tr. 5288-89. 912. ENH built a new cardiac cath lab to support the interventional cardiology program; renovated and expanded the emergency department and psychiatry units; and added modern equipment in a variety of areas. Chassin, Tr. 5288-89. 913. ENH replaced the Highland Park patient care building’s electrical distribution and ventilation systems, plumbing, and waste pipes. Hillebrand, Tr. 1982.

914. ENH built a new central plant at Highland Park, including a new power plant that houses utilities such as electrical generators, backup generators, boilers, and air ventilation equipment. Hillebrand, Tr. 1979; O’Brien, Tr. 3514-15; CX 6304 at 14 (Livingston, Dep.).

915. ENH added an additional boiler, new air handlers for the ventilation system, replaced the electrical generator, and added a second emergency electrical generator. Hillebrand, Tr. 1979-80. 916. ENH began remodeling all of its patient units in December of 2003. O’Brien, Tr. 3511-12; Neaman, Tr. 1351-52. The process of remodeling patient rooms is continuing and scheduled at least through 2006. O’Brien, Tr. 3513.

917. ENH added a new parking garage and made improvements to the lobby corridor and entrance to Highland Park. O’Brien, Tr. 3513-15; Hillebrand, Tr. 1920-21, in camera; CX 6304 at 14 (Livingston, Dep.).

918. On April 15, 1999, the Illinois Department of Public Health Health and Healthcare Financing Administration conducted a survey survey of Highland Park’s physical plant and identified 144 physical physical plant deficiencies which needed to be corrected to continue VOLUME 144 Initial Decision continue to participate in Medicare. Chassin, Tr. 5285-86; RX 1379 1379 at ENH LH 11544.

919. On August 26, 1999, 26 items were removed from the list and 3 were added for a total of 121 deficiencies. RX 1379 at ENH LH 11544.

920. On December 9, 1999, a reinspection was conducted and 88 additional items were removed from the list leaving a total of 33 items. The plan for correction of these remaining items was submitted by Highland Park on December 28, 1999, prior to the merger, and these remaining items were corrected by ENH by August 1, 2000. RX 1379 at ENH LH 11544; Spaeth, Tr. 2258-59. (f) Oncology Service 921. Through the Kellogg Cancer Center at Highland Park, ENH implemented a multidisciplinary approach that brought together an oncology team consisting of the physician oncologist, nurse, pharmacist, psychologist, social workers, and nutritionists who were available to patients in one location. Chassin, Tr. 5369-70; Dragon, 4391.

922. ENH brought subspecialty oncologists to Highland Park so that patients would not have to travel for their consultations. Chassin, Tr. 5369-70.

923. The Kellogg Cancer Center moved into a new section of the ambulatory care center in March 2005. Dragon, Tr. 4389-90. 924. Before the merger, Highland Park already had undertaken numerous initiatives in ontology services and had a variety of options other than the merger to achieve these same ends. Spaeth, Tr. 2224-25; CX 91 at 2; CX 1869 at 4; Neaman, Tr. 1243. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 191 Initial Decision 925. Highland Park had considered joint comprehensive oncology programs with organizations other than ENH. CX 1868 at 13; CX 99 at 2; CX 1866 at 1, 5; Newton, Tr. 420. 926. In the 1990’s, Highland Park had created centers of excellence for oncology and breast cancer that it was continually improving until the time of the merger. CX 91 at 2; CX 1869 at 4; Newton, Tr. 291-92, 419-20.

927. These centers of excellence already had access to the necessary technology, physicians, and research protocols in place to develop a comprehensive oncology program, and Highland Park merely needed to develop the community perception of excellence. Newton, Tr. 291-92, 419-20.

928. To this end, Highland Park could have expanded its oncology services and research activities through an affiliation agreement with hospitals other than ENH and, in fact, it was exploring these options before the merger, including the possibility of a joint venture with ENH or another hospital for oncology services. Newton, Tr. 340-42, 417-20; Neaman, Tr. 1243; Hillebrand, Tr. 2044-45.

(g) Radiology, Radiation Medicine, and Nuclear Medicine 929. ENH purchased a linear accelerator for Highland Park. O’Brien, Tr. 3500.

930. ENH added two new CT scanners in Highland Park’s radiology department, upgraded radiation therapy equipment, and purchased a simulator. O’Brien, Tr. 3496, 3501-02; Chassin, Tr. 5362-63; RX 1896 at ENHL MO 7109.

931. ENH purchased a CT pet, which is a diagnostic tool, for the nuclear medicine department. O’Brien, Tr. 3496, 3501-02. VOLUME 144 Initial Decision 932. ENH extended RADNET, its radiology imaging system and PACS, its filmless radiology imaging system, to Highland Park. O’Brien, Tr. 3494; Romano, Tr. 3184-85, in camera. 933. ENH added additional radiology staff to improve turnaround times for reading radiology reports. O’Brien, Tr. 3493. 934. Highland Park had a premerger budget of $9.5 million to improve radiology services. CX 545 at 20. (h) Emergency Care 935. ENH improved both the physical layout and service components of Highland Park’s emergency department. Chassin, Tr. 5333-34.

936. ENH invested in a major facility expansion, improved physician and nurse staffing, and improved the fast track procedures in the emergency department. Harris, Tr. 4213-14; Newton, Tr. 470; Hillebrand, Tr. 1980-81.

937. Prior to the merger, the emergency department at Highland Park was “very good,” and was “on par, if not better” than Highland Park’s peers. Newton, Tr. 394-95.

938. Throughout the 1990’s, Highland Park had continually made improvements to its emergency care: it had implemented a fast-track program to improve turnaround times; it had added physician assistants to the emergency room; it had streamlined the radiology process; and it had reduced the time that it took for a patient to receive an EKG. Harris, Tr. 4266-70. 939. Highland Park planned to “expand the Emergency Department from a facilities standpoint.” Newton, Tr. 394; Harris, Tr. 4289-90; CX 98 at 2.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 193 Initial Decision 940. Highland Park could have made the changes to the emergency department absent the merger: for example, most emergency departments at hospitals like Highland Park are staffed through contracts with physician groups, and Highland Park simply could have “demanded” higher staffing of the emergency room as a condition of its contract. Romano, Tr. 3111-12, in camera; Harris, Tr. 4204-07.

(i) Laboratory Medicine 941. Prior to the merger, Highland Park operated Consolidated Medical Labs (“CML”), a joint venture with Lake Forest that consisted of a main lab located between the two hospitals with satellite labs at Highland Park and Lake Forest. Victor, Tr. 3638-40. 942. After the merger, ENH decided to close CML and expand the on-site laboratory at Highland Park, although certain tests are sent to the laboratory at Evanston. O’Brien, Tr. 3507-09; Victor, Tr. 3591-92.

943. ENH constructed new histology and cytology laboratories on-site, installed over $1 million in state-of-the-art lab equipment, and introduced more stringent quality controls. Victor, Tr. 3615-17, 3619-20.

944. CML afforded Highland Park’s lab “greater volume,” “access to greater human pathology,” and the “opportunity to provide a greater benchmark in terms of [the lab’s] performance.” Newton, Tr. 396-97. The lab operated “actually exceptionally well.” Newton, Tr. 396.

945. Highland Park could have easily implemented further changes in its laboratory in the absence of the merger. Romano, Tr. 3178, in camera.

VOLUME 144 Initial Decision 946. Many of the changes that ENH made after the merger were simply consistent with updates that all hospital laboratories made during that period in order to meet licensing and accreditation standards. Romano, Tr. 3179, in camera.

(j) Pharmacy 947. ENH installed approximately twenty Pyxis automated drug distribution machines at Highland Park in 2000. Kent, Tr. 4851, 4854-55.

948. ENH has decentralized the pharmacists. RX 1697 at ENHL PK 51635; Kent, Tr. 4864-65.

949. ENH added an additional pharmacist to dispense medications at night in the summer of 2003. Kent, Tr. 4846, 4849; RX 1697 at ENHL PK 51635.

950. The Pyxis system did not become available to hospitals until the late 1990’s, when there was a “trend” in which pharmaceuticals and medications were decentralized to be located in the unit itself. Newton, Tr. 397-98.

951. Pyxis costs about $20,000 per machine, and Highland Park could have installed the machines on its own. Newton, Tr. 399; Romano, Tr. 3180, in camera.

(k) Cardiac Surgery 952. ENH opened a cardiac surgery program at Highland Park in June of 2000. Spaeth, Tr. 2275-76; Neaman, Tr. 1381; RX 879 at ENH GW 3252.

953. Cardiac surgery is a necessary component of a full-service cardiology program. Chassin, Tr. 5290.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 195 Initial Decision 954. Cardiac surgery procedures include coronary artery bypass grafting, valve procedures, and surgery on the aorta. Rosengart, Tr. 4452.

955. Before the merger, Highland Park already had plans to open a cardiac surgery program with Evanston or another hospital. CX 1868 at 13; CX 1867 at 1; CX 91 at 2; CX 1869 at 4; Newton, Tr. 335-38.

956. Highland Park also considered a joint cardiac surgery program with Northwestern Memorial or Advocate Lutheran General. Newton, Tr. 338.

957. ENH runs successful joint cardiac surgery programs with Swedish Covenant and Louis A. Weiss. Romano, Tr. 3075, in camera; Rosengart, Tr. 4443-44.

958. Highland Park and Evanston had executed a contract for a joint cardiac surgery program before the merger. Newton 335-36; CX 2094.

959. The Certificate of Need Application for the Highland Park cardiac surgery program indicates that the collaboration necessary to implement the program did not depend on the merger. See CX 413 at 5.

(l) Interventional Cardiology 960. Interventional cardiology refers to the treatment of obstructions in coronary arteries (coronary disease) by dilating the plaques obstructing the arteries and inserting little wire tubes called stents to keep the arteries open. Chassin, Tr. 5303. 961. After the merger, ENH established an interventional cardiology program at Highland Park. Chassin, Tr. 5304-05. VOLUME 144 Initial Decision 962. ENH built a new cardiac catheterization lab at Highland Park that performs both diagnostic and interventional procedures such as angioplasties. Hillebrand, Tr. 1980; O’Brien, Tr. 3490. 963. Highland Park’s medical staff included physicians with the expertise to perform interventional cardiac procedures. Newton, Tr. 466.

964. Highland Park planned to expand the diagnostic capabilities of its existing cardiac catheterization lab and to provide emergent angioplasty in conjunction with the planned cardiac surgery program or even “without open heart on-site.” Newton, Tr. 337, 416-17. (m) Psychiatry 965. Before the merger and through the spring of 2001, Highland Park and Evanston each had separate inpatient psychiatric units that treated both adult and adolescent patients. O’Brien, Tr. 3516; RX 1754 at ENH RS 3086.

966. In the spring of 2001, ENH consolidated the adolescent inpatient services at Highland Park and the adult inpatient services at Evanston. O’Brien, Tr. 3517; Chassin, Tr. 5339; Neaman, Tr. 1358- 59; RX 1080 at ENHL PK 55405.

967. ENH hired several adolescent psychiatrists to staff the Highland Park adolescent unit. O’Brien, Tr. 3518. 968. ENH remodeled the psychiatric unit in December 2003, to include private patient rooms with a keyless entry system and secure furniture. O’Brien, Tr. 3518-19.

969. The post-merger segregation of psychiatric patients (adolescents at Highland Park and adults at Evanston) is a structural EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 197 Initial Decision structural change which has not been shown in the medical literature literature to improve outcomes. Romano, Tr. 3115-16, in camera. VOLUME 144 Initial Decision (n) Intensivist Program 970. ENH added an intensivist program to Highland Park after the merger. Ankin, Tr. 5041; RX 1099 at ENHE F35 340; O’Brien, Tr. 3529-30; Chassin, Tr. 5328.

971. An intensivist is a physician who specializes in the care of intensive care patients and who has more experience dealing with the complications of these critically ill people. Ankin, Tr. 5035-36; O’Brien, Tr. 3529.

972. Intensivists also have an administrative role in overseeing and coordinating the medical and nursing staff that provide care to critically ill patients. Ankin, Tr. 5036. 973. Intensivist programs in hospitals like Highland Park became popular only after 2000. Romano, Tr. 3113-14, in camera; Ankin, Tr. 5078.

974. Pulmonary Physicians of the North Shore, which provides the intensivist coverage at Highland Park, does so through a contractual arrangement. Ankin, Tr. 5103-04, in camera; CX 2176 at 1, in camera.

975. Pulmonary Physicians of the North Shore would consider contracting with a new owner of Highland Park. Ankin, Tr. 5104-05, in camera.

(4) ENH’s Merger Specific Changes to Highland Park (a) Electronic Medical Records 976. In 2001, ENH decided that its current medical records system was not sufficient to meet its needs and ENH began its search for a better system. Wagner, Tr. 3964. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 199 Initial Decision 977. In June 2001, the EPIC clinical information system was selected from a group of finalists. Wagner, Tr. 3965. 978. EPIC is a software system for managing patient records for both hospital and physicians and was selected, in part, for its ability to work with physician offices. Wagner, Tr. 3966-67. 979. EPIC includes a computerized physician order entry system and clinical decision support systems. O’Brien, Tr. 3520; Chassin, Tr. 5365.

980. The EPIC system was implemented at all three hospitals, at the faculty practice medical group, and at all the affiliated physician practices that were willing to participate. Wagner, Tr. 3967. 981. EPIC became functional at Highland Park in December 2003. Wagner, Tr. 4069- 70; Neaman, Tr. 1251. 982. ENH spent approximately $14 million to implement EPIC at Highland Park. O’Brien, Tr. 3523; Hillebrand, Tr. 1984; Neaman, Tr. 1355.

983. Comprehensive medical records systems like EPIC are an emerging technology and very few hospitals had such a system before 2000. Romano, Tr. 3161-62, in camera. 984. There are a number of electronic medical records systems other than EPIC, including Meditech and McKesson. Wagner, Tr. 4067-69.

985. Meditech, as deployed at Highland Park, was not paperless, could not be accessed remotely, and lacked ambulatory capability. O’Brien, Tr. 3521; Wagner, Tr. 4061-62.

VOLUME 144 Initial Decision 986. Meditech, the computer program used by Highland Park before the merger, was and is an “excellent” system that other hospitals continue to use today. Romano, Tr. 3165-66, in camera; Newton, Tr. 333-34.

987. The federal government has established a national initiative to develop universally accessible electronic healthcare records systems for all citizens. In 2004, the Office of National Healthcare Information Technology was created. Wagner, Tr. 3957; RX 1701 at 1.

(b) Medical Staff Integration and Academic Involvement 988. Family medicine is the only department at Highland Park that has residents and at the time of trial, there were only 6 residents. O’Brien, Tr. 3539; Romano, Tr. 3125, in camera. 989. Since the merger, physicians in pathology, radiology, emergency medicine, cardiology, cardiac surgery, and some parts of anesthesiology rotate through all three campuses. Chassin, Tr. 5598; O’Brien, Tr. 3540-41.

990. Following the merger, about sixty Highland Park physicians were able to obtain appointments at Northwestern Medical School. Chassin, Tr. 5376; O’Brien, Tr. 3540.

991. ENH provides Highland Park physicians with a $4,000 continuing medical education stipend. Harris, Tr. 4253. 992. The merger did not transform Highland Park into an academic hospital. Romano, Tr. 3117-18, in camera. 993. Merely being owned by a teaching hospital has not been shown in previous studies to be associated with improved processes processes and outcomes of care. Romano, Tr. 3118, in camera. There EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 201 Initial Decision There is no evidence that Highland Park benefitted simply by being being owned by a teaching hospital. Romano, Tr. 3124, in camera. camera.

c. Nonprofit Status (1) Respondent’s Nonprofit Status Did Not Affect Its Approach to Post-Merger Price Increases 994. As part of the merger with Highland Park, ENH decided to renegotiate contracts with the managed care organizations in 2000. Neaman, Tr. 1031; see F. 355-64.

995. When ENH set prices for the 2000 contract renegotiations with health plans, the fact that it was a non-profit entity did not weigh in as a reason not to take actions toward higher prices. Neaman, Tr. 1032-33.

(2) Respondent’s Nonprofit Status Did Not Affect Incentives for Management 996. On June 29, 1999, shortly before the letter of intent to merge was signed, Highland Park’s senior executives entered into enhanced compensation agreements that replaced their previous agreements. The new agreements “offered additional retention bonuses as well as enhanced severance agreements” at a cost of $8 million. CX 534 at 3.

997. ENH’s managers were given bonuses for meeting revenue targets from operations, giving managers the incentive to set supra competitive prices. Simpson, Tr. 1629.

998. ENH management planned to benefit from some of the money derived from raising hospital prices post-merger. The president of ENH proposed adding an additional $3 million into the the 2000 bonus pool attributable to the merger integration activities. VOLUME 144 Initial Decision activities. The board reduced this amount to $1 million, which ultimately was the amount distributed to the top fifty people. Neaman, Tr. 1263-64; CX 31 at 1.

999. Several of ENH’s senior executives received merit increases in their salaries in the range of 5 to 6% in 1998 to 1999 and a 10% increase from fall of 2000 to fall of 2001. These increases in compensation coincided with the completion of the merger integration efforts. Neaman, Tr. 1265-67; CX 2099 at 2-3. 1000. Various ENH executives also received substantially higher awards at the end of 2000 compared to the awards in 1998 and 1999. Neaman, Tr. 1267-69; CX 2099 at 8-9.

1001. ENH’s compensation contracts did not align management’s interests with consumers on the issue of price. Simpson, Tr. 1629.

(3) Respondent’s Board Did Not Get Involved in Pricing Issues 1002. ENH’s Board contains community representatives who provide oversight to the organization. Simpson, Tr. 1639. Approximately three-quarters of ENH’s Board are outside directors chosen from the community. Simpson, Tr. 1639. In addition to the ENH Board, the Healthcare Foundation of Highland Park also monitors ENH’s activities, specifically its commitments to Highland Park and the Highland Park community. RX 2037 at HFHP 1364; Styer, Tr. 4971, 4985.

1003. The ENH board did not actively monitor the pricing decisions of hospital management and did not try to ensure that price was set at a competitive level. Simpson, Tr. 1622, 1629. 1004. Spaeth attended meetings of the Highland Park board before the merger and of the ENH board after the merger. Spaeth, Tr. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 203 Initial Decision Tr. 2215. Over the years, including after the merger, Spaeth has never heard a board member or Neaman say that ENH should lower lower its rates to managed care organizations or make any comment comment regarding the rate at which the hospital was contracting with a particular payor. Spaeth, Tr. 2218-19. 1005. The ENH board is not involved in negotiations with managed care organizations, does not review contracts, and is not informed in advance of negotiating strategies. CX 6304 at 17-18 (Livingston, Dep.).

(4) Highland Park Healthcare Foundation 1006. In December 1999, Evanston Hospital and the Highland Park Foundation signed the agreement creating the Healthcare Foundation of Highland Park. RX 2037; Styer, Tr. 4977-78. The Healthcare Foundation of Highland Park came into being on January 1, 2000, as a result of the merger. Styer, Tr. 4951, 4971; Belsky, Tr. 4894; Spaeth, Tr. 2281.

1007. The Healthcare Foundation of Highland Park started with a corpus of roughly $100 million. Neaman, Tr. 1260. As of March 2005, the Healthcare Foundation had an $85 million corpus, down from its original $100 million, due to poor performance of investments in 2000 and 2001 and because the Foundation has given away more than $28 million. Styer, Tr. 4979-80. 1008. During the merger negotiations, Evanston attempted to minimize the amount of funds that Highland Park would contribute to the post-merger foundation. Kaufman, Tr. 5863. 1009. The Foundation Agreement establishing the Healthcare Foundation of Highland Park describes the Foundation’s mission as being to support Highland Park and healthcare in the general Highland Park community. RX 2037 at HFHP 1356; Styer, Tr. 4951, 4979; Neaman, Tr. 1373.

VOLUME 144 Initial Decision 1010. The Foundation Agreement creating the Healthcare Foundation of Highland Park obliged the Foundation to send to ENH the greater of 100% of its investment earnings or $8 million in 2000, the greater of 75% of its investment earnings or $6 million in 2001 and 2002, and the greater of 50% of its investment earnings or $4 million for every year thereafter. RX 2037 at HFHP 1362; Styer, 4980-81; Spaeth, Tr. 2281; Neaman, Tr. 1261; Belsky, Tr. 4898. The Foundation Agreement, in turn, obliges ENH to use the money it gets from the Healthcare Foundation to offset the costs of uncompensated care and other clinical programs at Highland Park selected at ENH’s discretion. RX 2037 at HFHP 1362; Styer, Tr. 4981.

1011. The majority of the Healthcare Foundation’s funds sent to ENH are used to support indigent or uncompensated care at Highland Park. Styer, Tr. 4981; H. Jones, Tr. 4179-80. 1012. The Healthcare Foundation of Highland Park also dispenses grants to charities in the Highland Park area. Styer, Tr. 4987-88. Since its creation, the Healthcare Foundation of Highland Park has given roughly $26 million back to Highland Park and another $3 to 4 million to organizations within the greater Highland Park community. Styer, Tr. 4974.

1013. In 2002, the Healthcare Foundation awarded $500,000 to the Lake County Health Department to establish a community healthcare clinic in the Highland Park/Highwood area to improve access to healthcare for underserved populations in southeast Lake County. RX 1238 at HFHP 2565.

d. Ease of Entry 1014. Illinois has a state Certificate of Need (“CON”) Law that governs future hospital entry or expansion. D. Jones, Tr. 1653-54, 1655; Spaeth, Tr. 2167.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 205 Initial Decision 1015. CON approval from the state’s Planning Board is required if a health care facility is going to engage in a transaction that is clinical in nature and exceeds either the capital expenditure or the major medical equipment threshold. D. Jones, Tr. 1655. 1016. The Planning Board, when reviewing a CON application for additional beds, considers whether the proposed beds are actually needed at the facility. D. Jones, Tr. 1656. 1017. Bed need is calculated with need formulas established by the board in its administrative rules. The Division of Health Statistics compiles the data and variables necessary to compute those bed needs for the Division of Health Systems Development. D. Jones, Tr. 1664.

1018. Based on the Planning Board’s current addendum to its inventory, there is no need for beds in the Evanston, Glenview, and Highland Park areas for services in medical/surgical, pediatrics, or intensive care units. D. Jones, Tr. 1665-66. 1019. If someone were to submit a CON application for the construction of a new hospital in Evanston today, the Department of Public Health’s report would most likely issue a negative finding regarding the bed need for a new facility by referencing the existing providers in the Evanston area, referencing the current bed need calculation for that area, and determining that additional beds are not needed based on the Planning Board’s inventory. D. Jones, Tr. 1666- 67.

1020. The state CON Board has denied hospitals beds where there is no bed need. If an area is overbedded, the likelihood that the State of Illinois would approve additional beds is minimal. Further, other hospitals might intervene to oppose the CON application. Spaeth, Tr. 2168-69.

VOLUME 144 Initial Decision 1021. There have been no CON applications for the construction of new hospitals in the area around Highland Park, Evanston, or Glenbrook over the past five years. D. Jones, Tr. 1664. 1022. In addition to a Certificate of Need, a person would need to get approval from other state agencies and local governments to build a new hospital. The Illinois Department of Health reviews facility plans, and a city council may need to provide zoning approval for the new hospital. Spaeth, Tr. 2169. 1023. The Illinois CON law is scheduled to be repealed on July 1, 2006. D. Jones, Tr. 1685. Unless the Illinois CON law is extended or new laws are enacted, the CON process will cease to exist in July 2006. D. Jones, Tr. 1685.

1024. Irrespective of the CON law, it takes about two and a half to three years to build a new hospital. Spaeth, Tr. 2169. 1025. In 1999, Condell filed a CON application for a major modernization and expansion of its hospital facilities, including its inpatient, ancillary and support services. RX 755 at CMC 5978. Since the merger, the Illinois Health Facilities Planning Board granted Condell Medical Center permits to add ten medical/surgical beds, eight ICU beds, and ten obstetric beds. D. Jones, Tr. 1683-84. 1026. In 2003, the Illinois Health Facilities Planning Board granted Lake Forest a permit to increase the number of medical/surgical beds by 10 beds. D. Jones, Tr. 1684. 1027. Since Evanston’s merger in 2000 with Highland Park, there has been no new hospital entry in the North Shore area (D. Jones, Tr. 1664), even though Evanston has raised prices. See F. 347- 755.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 207 Initial Decision e. Failing Firm (1) Highland Park Could Have Continued As a Stand Alone Competitor Without the Merger (a) Highland Park’s Management and Board Believed That Highland Park Was Financially Strong 1028. On March 23, 1999, Lakeland’s finance and planning committee approved the 1999-2003 Strategic Plan and Financial Plan. Spaeth, Tr. 2146; CX 1055 at 3.

1029. At the March 23, 1999 meeting, when members posed the question of the long-term financial viability, the Lakeland finance and planning committee concluded that Highland Park “can remain financially strong over the foreseeable future.” CX 1055 at 3; Newton, Tr. 432-34; Spaeth, Tr. 2147.

1030. Highland Park’s 1999-2003 financial plan set forth a “long range capital budget” that included $43 million for “strategic initiatives and master plan items,” including “ambulatory, assisted living and facility expansion.” The plan also set aside $65 million for “[h]ospital construction, routine capital and information technology” investments, and a small amount for Lakeland Health Ventures. The combined budget was in excess of $100 million. Newton, Tr. 430-31; CX 545 at 3; CX 1055 at 2. 1031. According to Highland Park’s 1999-2003 financial plan, “[c]ash and investments are forecasted to grow from $238 million in 1998 to $323 million in 2003.” CX 1055 at 3. 1032. Highland Park also forecasted that its investments would generate a return of $28 million in incremental net revenues in 2003. CX 1055 at 2.

VOLUME 144 Initial Decision 1033. The 1999-2003 Highland Park financial plan emphasized that “[e]existing cash and investments are available to fund strategic initiatives and generate new programs.” CX 545 at 3. 1034. At the April 30, 1999 Highland Park board meeting, the board members approved the 1999-2003 Strategic Plan and Financial Plans. CX 96 at 4; Spaeth, Tr. 2155. The board members did not express doubt about Highland Park’s ability to generate the $100 million required to fund the projects. Newton, Tr. 430-32. 1035. Highland Park’s 1999-2004 Financial Plan projected that cash and investments would increase by $48 million from 1999- 2004, and that long-term debt would be reduced by $24.3 million, excluding amortization. CX 1903 at 1.

1036. Highland Park’s 1999-2004 Financial Plan projected that it had sufficient cash flow for both the planned capital expenditures and the strategic initiatives. CX 1903 at 1. 1037. Highland Park’s 1999-2004 Financial Plan included planned capital expenditures of $79 million. These expenditures were comprised of “primarily routine capital for equipment and facility improvements, construction for renovation of patient care areas, information system enhancements and physician development.” CX 1903 at 1.

1038. Highland Park’s 1999-2004 Financial Plan also included an additional $28 million in planned expenditures for “Strategic/Master Plan Initiatives.” These initiatives included development of a cath lab, additional parking, and additional facilities for oncology and radiation therapy. CX 1903 at 1, 3. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 209 Initial Decision (b) Highland Park Had a Strong Balance Sheet 1039. Kenneth Kaufman is managing partner of Kaufman Hall & Associates, a financial consulting firm primarily servicing hospital systems. Kaufman, Tr. 5773. Kaufman and his firm provided financial and strategic consulting services to Highland Park prior to its merger with ENH and served as transaction counsel to Highland Park during the ENH merger negotiations. Kaufman, Tr. 5774, 5777-78.

1040. Kaufman advised the Highland Park board and management that “the financial condition of Highland Park was such that it did not require a financial reason to go forward with the merger.” Kaufman, Tr. 5840; CX 1923 at 2. 1041. At the end of 1998, Highland Park had a strong balance sheet. Kaufman, Tr. 5860.

1042. At the end of 1998, Highland Park had 444 days of cash on hand. CX 1912 at 1; Newton, Tr. 427-28. This was the equivalent of being able to run a fully functional hospital for 444 days without a penny of additional revenue. Kaufman, Tr. 5859-60. The 444 days of cash on hand did not include any premerger foundation assets. Kaufman, Tr. 5860.

1043. At the end of 1998, Highland Park had $133.6 million in cash assets available to contribute to the merged ENH. Kaufman, Tr. 5842; CX 1912 at 2. This $133.6 million did not include the premerger Highland Park Foundation’s assets. Kaufman, Tr. 5842; CX 1912 at 2-3.

1044. At the end of 1998, Highland Park and its affiliated corporations had a total of about $235 million in cash and unrestricted investments. The components of this total were the $102 $102 million earmarked for the independent, post-merger foundation foundation and $133.6 million in cash and unrestricted investments VOLUME 144 Initial Decision investments that Highland Park planned to contribute to the merged merged ENH. Kaufman, Tr. 5842, 5844.

1045. At the end of 1998, Highland Park and the foundation had $120 million in long-term debt. Kaufman, Tr. 5844; CX 1912 at 1. Highland Park’s bond issues in the 1990's accounted for this longterm debt. Kaufman, Tr. 5844. The assets of the obligated groups (the foundation and the hospital) backed up the long-term debt. Kaufman, Tr. 5846; CX 413 at 120.

1046. At the end of 1998, Highland Park had a debt service coverage ratio of 1.8 and a debt to capitalization ratio of 61%. CX 1912 at 1.

1047. When Kaufman calculated the debt indicators set forth in his February 1999 memorandum to Stearns and Spaeth, Kaufman did not include the assets of the foundation. Kaufman, Tr. 5846. Including the entirety of the obligated group’s assets in the financial calculations would cause the debt indicators to improve compared to indicators that only utilized the hospital’s assets. Kaufman, Tr. 5858. 1048. Highland Park projected that by 2003 the debt service coverage ratio would improve to 3.1 and the debt to capitalization ratio to 39%. CX 413 at 119.

1049. Highland Park and its affiliated corporations experienced a decline in long-term debt and an increase in cash and unrestricted investments position from 1998 to 1999. In particular, long-term debt declined from $120.5 million to $116.7 million. CX 693 at 17. Cash and unrestricted investments increased from $217.8 million to approximately $260 million. CX 693 at 16. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 211 Initial Decision 1050. At the end of 1999, Lakeland Health, Highland Park’s parent, had $140 million more in cash and unrestricted investments than long-term debt. CX 693 at 16-17.

1051. In 1999, Kaufman advised Highland Park that the hospital “has always Supported its credit position through exceptional liquidity.” CX 1912 at 2.

(c) Highland Park Was Backed by its Foundation’s Assets 1052. Premerger, Highland Park, through its parent, Lakeland Health, was backed by the assets of its foundation. These funds were available to support the hospital. Styer, Tr. 4954. The post-merger, independent foundation was established in order to compensate the local community of Highland Park for the loss of control following Highland Park’s merger with Evanston. Kaufman, Tr. 5855. 1053. The premerger Highland Park Foundation was “responsible for fund raising for and on behalf of Lakeland Health Services, Inc. (“Lakeland”), the Hospital [Highland Park] and their affiliates.” CX 6321 at 61.

1054. These raised funds were available to Highland Park. The foundation “maintains the funds received and distributes the funds based upon the needs of the affiliates, or, if restricted to a specific purpose, the directions of the donor.” CX 6321 at 61. As the former chairman of the premerger foundation testified, “[t]he funds from the premerger Foundation went to support the hospital, to fulfill needs.” Styer, Tr. 4954.

1055. Premerger, Highland Park executives “would bring [the foundation board] various projects that were ongoing in the hospital,” and the foundation members would select specific projects to fund, such as improvements to the hospital’s dialysis center. Styer, Tr. 4959-60.

VOLUME 144 Initial Decision (d) No Financial Need to Merge 1056. In the fall of 1998, Highland Park contemplated both a merger strategy as well as an independent, stand alone growth strategy. CX 1869 at 5-6; Spaeth, Tr. 2145-46 (plans set forth goals for “going forward without a merger”).

1057. Highland Park was prepared to proceed with the status quo, unaffiliated option if the ENH merger talks failed. Kaufman, Tr. 5838.

1058. Steams, Highland Park’s Chairman of the Board, testified that he believed that Highland Park was not in danger of exiting the market for at least ten years. CX 6305 at 5 (Steams, Dep.). 1059. If the merger with ENH had not closed, Highland Park had “the financial wherewithal to sustain [itself].” Highland Park management and board believed that “[t]here was no urgency to have an alternative immediately available.” CX 6305 at 11 (Steams, Dep.).

1060. Highland Park believed pursuing the stand alone, independent option in 1998-99 “was absolutely a viable alternative for Highland Park.” Newton, Tr. 319-20.

1061. Highland Park could remain independent due to a variety of factors. It had a quality medical staff with significant coverage over a range of about forty-five specialties. It had a broad primary care network and it was efficient in managed care activities. Newton, Tr. 320.

1062. At a March 23, 1999 meeting, the Lakeland finance and planning committee concluded that based on the 1999 strategic and and financial plans, Highland Park “can remain financially strong over the foreseeable future.” CX 1055 at 3; Spaeth, Tr. 2147. These EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 213 Initial Decision These plans were “developed assuming no affiliation with another provider were to occur.” CX 1055 at 1; Spaeth, Tr. 2145-46. 1063. Highland Park proposed a year 2000 budget in October 1999. The budget was prepared assuming no merger with ENH would take place; “therefore, no merger-related impact [was] included.” CX 397 at 1. The proposed budget for 2000 anticipated “dramatic improvement over 1999’s results.” CX 397 at 1. For example, the budget projected net revenue increases of more than $6.3 million in 2000 for the hospital. CX 397 at 3. 1064. The Highland Park board had assessed the financial position of the hospital and felt it was acceptable. Highland Park was not planning to file for bankruptcy before the merger. It never considered filing for bankruptcy. Spaeth, Tr. 2308. (2) Highland Park Was an Attractive Candidate for Other Merger Partners 1065. Highland Park viewed itself as an attractive partnership candidate and considered other partners besides ENH. In the fall of 1998, Highland Park contemplated a number of potential merger partners, besides Evanston, including Northwest Community, Lake Forest, and Condell. CX 1869 at 6.

1066. Highland Park had a strong balance sheet (F. 1039-51), was backed by its foundation’s assets (F. 1052-55), had an “attractive service area” (F. 339, 1069), and had no financial need to merge (F. 1056-64).

1067. If the ENH merger had not closed, Highland Park was prepared “to continu[e] to explore other options,” meaning “other partnership options.” CX 6305 at 11 (Stearns, Dep.). 1068. According to Highland Park’s chairman of the board, Highland Park “had at least some contact with other institutions and and . . . would have pursued those more aggressively had this -- the VOLUME 144 Initial Decision the merger with Evanston not gone through.” CX 6305 at 11-12 (Stearns, Dep.).

1069. Highland Park had “an attractive service area,” and therefore, it “would be attractive to other partnership candidates.” CX 6305 at 12 (Stearns, Dep.).

III. ANALYSIS AND CONCLUSIONS OF LAW A. Preliminary Issues 1. Jurisdiction The Complaint charges Respondent with violating Section 7 of the Clayton Act, 15 U.S.C. § 18. Section 7 of the Clayton Act states: No person ... shall acquire, directly or indirectly, the whole or any part of the stock or other share capital ... of another person ... where in any line of commerce or in any activity affecting commerce in any section of the country, the effect of such acquisition may be substantially to lessen competition, or tend to create a monopoly. 15 U.S.C. § 18. The word “person” includes corporations and associations existing under or authorized by the laws of any state. 15 U.S.C. § 12(a). Section 7 of the Clayton Act applies to asset acquisitions by nonprofit hospitals. FTC v. University Health, Inc., 938 F.2d 1206, 1209 (11th Cir. 1991).

Respondent Evanston Northwestern Healthcare Corporation (“ENH”) is a nonprofit corporation organized, existing, and doing business under the laws of Illinois. F. 1. ENH owns and operates three acute care hospitals: Evanston Hospital (“Evanston”), Glenbrook Hospital (“Glenbrook”), and Highland Park Hospital EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 215 Initial Decision (“Highland Park”). Prior to the merger, ENH was comprised of Evanston, Glenbrook, ENH Medical Group, ENH Research Institute, Institute, and ENH Homecare Services. F. 2. Throughout this Initial Initial Decision, except where noted, the premerger Glenbrook and and Evanston hospitals are referred to as “Evanston.” Prior to the merger, Highland Park was a nonprofit hospital and a subsidiary of of Lakeland Health Services (“Lakeland”), a nonprofit corporation corporation existing under the laws of Illinois. F. 18-19. In the merger agreement, finalized on October 29, 1999, the parties agreed that Lakeland and Highland Park would be merged into ENH and that Lakeland and Highland Park would no longer exist as separate corporations. F. 83-84. The merger was consummated on January 1, 2000. F. 85.

The Commission has express jurisdiction under Section 11 (b) of the Clayton Act to determine the legality of a corporate acquisition under Section 7. 15 U.S.C. § 21(b); United States v. Rockford Memorial Corp., 898 F.2d 1278, 1280 (7th Cir. 1990) (Commission’s jurisdiction to enforce the prohibitions of the Clayton Act includes the hospital industry); see also Hospital Corp. of Am. v. FTC, 807 F.2d 1381, 1386 (7th Cir. 1986). The Commission’s jurisdiction allows it to adjudicate the lawfulness of acquisitions that have already been completed. In re Chicago Bridge & Iron Co., 2005 WL 120878, Dkt. No. 9300, at 90 (Op. of FTC Commu January 6, 2005) (available at http://www.ftc.gov /os/adjpro/d9300/index.htm); In re Coca-Cola Co., 117 F.T.C. 795, 911 (June 13, 1994). See also United States v. E.I. du Pont de Nemours & Co., 353 U.S. 586, 597 (1957) (“[T]he Government may proceed at any time that an acquisition may be said with reasonable probability to contain a threat that it may lead to a restraint of commerce or tend to create a monopoly of a line of commerce.”). Accordingly, the Commission has jurisdiction over Respondent and the subject matter of this proceeding, pursuant to Sections 7 and 11 of the Clayton Act.

VOLUME 144 Initial Decision 2. Burden of Proof and Statutory Framework The parties’ burdens of proof are governed by Commission Rule 3.43(a), Section 556(d) of the APA, and case law. Federal Trade Commission Rules of Practice, Interim rules with request for comments, 66 Fed. Reg. 17,622, 17626 (April 3, 2001). Pursuant to Commission Rule 3.43(a), “[c]ounsel representing the Commission . . . shall have the burden of proof, but the proponent of any factual proposition shall be required to sustain the burden of proof with respect thereto.” 16 C.F.R. § 3.43(a). Under the APA, “[e]xcept as otherwise provided by statute, the proponent of a rule or order has the burden of proof.” 5 U.S.C. § 556(d). The APA establishes preponderance of the evidence as the standard of proof for formal administrative adjudicatory proceedings. Steadman v. SEC, 450 U.S. 91, 102 (1981). The preponderance of the evidence standard is also used in federal antitrust cases. E.g., Herman & McLean v. Huddleston, 459 U.S. 375, 389-90 (1983); Ramsey v. United Mine Workers, 401 U.S. 302, 308 (1971).

“To establish a prima facie case under Section 7 of the Clayton Act, [the government] must first define the relevant market, and then establish that the proposed merger will create an appreciable danger of anticompetitive consequences.” California v. Sutter Health Sys., 130 F. Supp. 2d 1109, 1118 (N.D. Cal. 2001) (citing United States v. Philadelphia Natl Bank, 374 U.S. 321, 362 (1962)). “[T]he test of a violation of § 7 is whether, at the time of the suit, there is a reasonable probability that the acquisition is likely to result in the condemned restraints.” E.I. du Pont, 353 U.S. at 607. “Congress used the words ‘may be substantially to lessen competition’ to indicate that its concern was with probabilities, not certainties.” Brown Shoe Co. v. United States, 370 U.S. 294, 323 (1962). “Thus, to satisfy section 7, the government must show a reasonable probability that the proposed transaction would substantially lessen competition in the future.” University Health, 938 F.2d at 1218. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 217 Initial Decision Under the framework established by the courts and the Commission, Complaint Counsel must first establish a prima facie case that the acquisition is unlawful. “Typically, this has been accomplished by showing that the transaction will significantly increase market concentration, which in turn establishes a ‘presumption’ that the transaction is likely to substantially lessen competition.” Chicago Bridge & Iron, Dkt. 9300, at 7 (citing U.S. DOJ and FTC, Horizontal Merger Guidelines, § 1.51 (1992, as amended 1997), reprinted in 4 Trade Reg. Rep. (CCH) ¶ 13, 104 (hereinafter “Merger Guidelines”)); FTC v. H.J. Heinz, Co., 246 F.3d 708, 715 (D.C. Cir. 2001); United States v. Baker Hughes, Inc., 908 F.2d 981, 982 (D.C. Cir. 1990). “[S]tatistics concerning market share and concentration are not conclusive indicators of anticompetitive effects, but they provide a meaningful context within which to address the question of the merger’s competitive effects.” FTC v. Warner Communications, Inc., 742 F.2d 1156, 1163 n.1 (9th Cir. 1984). “That the government can establish a prima facie case through evidence on only one factor, market concentration, does not negate the breadth of this analysis. Evidence of market concentration simply provides a convenient starting point for a broader inquiry into future competitiveness.” Baker Hughes, 908 F.2d at 984. Postacquisition evidence goes “directly to the question of whether future lessening of competition [is] probable,” and thus is appropriate to rely upon. United States v. General Dynamics, Corp., 415 U.S. 486, 506 (1974). Accordingly, Complaint Counsel may establish a prima facie case with concentration data and may introduce other types of evidence relating to market conditions to bolster their concentration data. Chicago Bridge & Iron, Dkt. 9300, at 7. If the government successfully establishes a prima facie case, “[t]he burden of producing evidence to rebut this presumption then then shifts to the defendant.” Baker Hughes, 908 F.2d at 982. Respondent “may rely on ‘nonstatistical evidence which casts doubt doubt on the persuasive quality of the statistics to predict future anticompetitive consequences,’” such as: ease of entry into the VOLUME 144 Initial Decision market, the trend of the market either toward or away from concentration, the continuation of active price competition, and weakness of the acquired firm. University Health, 938 F.2d at 1218 1218 (quoting Kaiser Aluminum & Chem. Corp. v. FTC, 652 F.2d 1324, 1341 (7th Cir. 1981)). In addition, evidence of improvements improvements that benefit competition, and hence, consumers, may may overcome the presumption arising from a prima facie case. See See University Health, 938 F.2d at 1223. If the respondent successfully rebuts the presumption of anticompetitive effects, “the “the burden of producing additional evidence of anticompetitive effect shifts to the government, and merges with the ultimate burden burden of persuasion, which remains with the government at all times.” University Health, 938 F.2d at 1218-19 (quoting Baker Hughes, 908 F.2d at 983); Sutter Health, 130 F. Supp. 2d at 1118. Accordingly, the proper “application of the burden-shifting approach requires the court to determine (1) the ‘line of commerce’ or product market in which to assess the transaction; (2) the ‘section of the country’ or geographic market in which to assess the transaction; and (3) the transaction’s probable effect on competition in the product and geographic markets.” United States v. Oracle Corp., 331 F. Supp. 2d 1098, 1110-11 (N.D. Cal. 2004); FTC v. Libbey, Inc., 211 F. Supp. 2d 34, 44-45 (D.D.C. 2002); FTC v. Staples, Inc., 970 F. Supp. 1066, 1072 (D.D.C. 1997); see also United States v. Phillipsburg Natl Bank & Trust Co., 399 U.S. 350, 359-66 (1970); United States v. Marine Bancorporation, 418 U.S. 602, 618-23 (1974).

B. Relevant Market Section 7 of the Clayton Act explicitly refers to “any line of commerce” and “any section of the country.” 15 U.S.C. § 18. Determination of the relevant market is a necessary predicate to a finding of a violation of the Clayton Act because the threatened monopoly must be one which will substantially lessen competition competition ‘within the area of effective competition.’” E.I. du Pont, EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 219 Initial Decision Pont, 353 U.S. at 593 (citation omitted). “The ‘area of effective competition’ must be determined by reference to a product market (the ‘line of commerce’) and a geographic market (the ‘section of the country’).” Brown Shoe, 370 U.S. at 324. Accordingly, an analysis of the antitrust implications of a challenged merger and whether a transaction violates Section 7 begins with an assessment assessment of the appropriate relevant market. FTC v. Freeman Hosp., 69 F.3d 260, 268 (8th Cir. 1995). Complaint Counsel bears the burden of proving a relevant market within which anticompetitive effects are likely as a result of the acquisition. FTC v. Tenet Health Care Corp., 186 F.3d 1045, 1052 (8th Cir. 1999); In re Adventist Health Sys./West, 117 F.T.C. 224, 289 (April 1, 1994). Indeed, Complaint Counsel must “show the rough contours of a relevant market” even when market power is established through direct evidence of anticompetitive effects. Republic Tobacco Co. v. North Atlantic Trading Co., Inc., 381 F.3d 717, 737 (7th Cir. 2004). As set forth below, substantial evidence in this case establishes that the relevant product market is general acute care inpatient services sold to managed care organizations and that the relevant geographic market encompasses the following hospitals: Evanston, Glenbrook, Highland Park, Lake Forest, Advocate Lutheran General, Rush North Shore, and St. Francis. 1. Product Market Complaint Counsel contends that the relevant product market is general acute care inpatient services sold to managed care organizations, which includes primary, secondary, and tertiary inpatient services, but excludes quaternary and outpatient services. CCB at 52-53. Respondent argues that because hospitals’ primary customers, managed care organizations, negotiate for all acute care hospital services, including both inpatient and outpatient services, the relevant product market also includes outpatient services. RB at 16-17.

VOLUME 144 Initial Decision a. Reasonable Interchangeability The relevant product or service market is “composed of products that have reasonable interchangeability for the purposes for which they are produced – price, use and qualities considered.” United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 404 (1956). This “cross elasticity of demand” represents product substitutability and the customer’s ability to choose among competing products. Id. at 393; H.J. Heinz, 246 F.3d at 718. The courts rely on various factors to determine how closely the products at issue compete. E.g., H.J. Heinz, 246 F.3d at 718-19; FTC v. Swedish Match, 131 F. Supp. 2d 151, 158-59 (D.D.C. 2000). “An element for consideration as to cross-elasticity of demand between products is the responsiveness of the sales of one product to price changes of the other.” E.I. du Pont, 351 U.S. at 400.

The Merger Guidelines delineate a product market by asking whether a hypothetical monopolist of the proposed product market could impose a “small but significant and nontransitory increase in price” (“SSNIP”) and not lose so much of its sales to alternative products that the price increase would be unprofitable. Merger Guidelines § 1.11; Swedish Match, 131 F. Supp. 2d at 160 (relevant question is whether the increase in the price of product B will induce substitution to product A to render product B’s “price increase unprofitable”). The SSNIP test typically utilizes a 5% price increase. Merger Guidelines § 1.11; Staples, 970 F. Supp. at 1076 n.8. Although the Merger Guidelines are not binding, courts have often adopted the standards set forth in the Merger Guidelines in analyzing antitrust issues. Sutter Health, 130 F. Supp. 2d at 1120. In order to define a relevant product market, a court must determine what services or products the customer, if faced with a price increase, could or would reasonably substitute for the products products in question. H.J. Heinz, 246 F.3d at 718 (citing Merger EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 221 Initial Decision Guidelines § 1.0); Eastman Kodak Co. v. Image Technical Services, Services, Inc., 504 U.S. 451, 481-82 (1992) (relevant market determined by the choices of products or services available to customers). The customers in this case are the managed care organizations that contract with hospitals for services. F. 110; see infra at Section III.B.2.a.

b. Hospital Context Inpatient hospital services may be treated as a “cluster of services” comprising acute inpatient care, rather than in terms of any individual service. Sutter Health, 130 F. Supp. 2d at 1119. This is necessary given a hospital’s chargemaster which, in this case, contains up to 20,000 individual service items and related procedures offered to patients. F. 176. “While the treatments offered to patients within this cluster of services are not substitutes for one another (for example, one cannot substitute a tonsillectomy for heart bypass surgery), the services and resources that hospitals provide tend to be similar across a wide range of primary, secondary, and tertiary inpatient services.” Sutter Health, 130 F. Supp. 2d at 1119. The cluster market concept has been accepted generally as the most realistic way to assess the actual competitive effects of hospital activity.

Courts reviewing hospital mergers consistently recognize acute inpatient care as the appropriate product market in hospital merger cases. E.g., Freeman Hosp., 69 F.3d at 268; University Health, 938 F.2d at 1210-11; Rockford Memorial, 898 F.2d at 1284; FTC v. Butterworth Health Corp., 946 F. Supp. 1285, 1290-91 (W.D. Mi. 1996); United States v. Long Island Jewish Med. Ctr., 983 F. Supp. 121, 139 (E.D.N.Y. 1997). See also In re Hospital Corp. of Am., 106 F.T.C. 361, 464-66 (Oct. 25, 1985), aff’d, Hospital Corp. of Am. v. FTC, 807 F.2d 1381 (7th Cir. 1986). The rationale for this conclusion is simply that “most hospital services cannot be provided by non-hospital providers; as to these, hospitals have no competition from other providers of medical care.” Hospital Corp. of Am., 807 F.2d at 1388.

VOLUME 144 Initial Decision In Section 7 hospital merger cases, the relevant market determination is restricted to acute inpatient care services, and not expanded to include outpatient services. E.g., Rockford Memorial, 898 F.2d at 1284; Butterworth Health Corp., 946 F. Supp. at 1290- 91. As explained by the Court of Appeals for the Seventh Circuit in Rockford Memorial:

For many services provided by acute-care hospitals, there is no competition from other sorts of provider. If you need a kidney transplant, or a mastectomy, or if you have a stroke or a heart attack or a gunshot wound, you will go (or be taken) to an acute-care hospital for inpatient treatment. The fact that for other services you have a choice between inpatient care at such a hospital and outpatient care elsewhere places no check on the prices of the services we have listed, for their prices are not linked to the prices of services that are not substitutes or complements. If you need your hip replaced, you can’t decide to have chemotherapy instead because it’s available on an outpatient basis at a lower price.

898 F.2d at 1284.

The evidence presented in this case is no less persuasive. The record establishes that, as a matter of medical practice and provision of services, there is an inherent inability to substitute outpatient services for inpatient services. F. 204-11. If a physician decides that a patient requires inpatient care, managed care organizations and hospitals do not and cannot switch the patient to outpatient care. F. 206. ENH’s expert concedes that inpatient and outpatient services are not functionally interchangeable. F. 211. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 223 Initial Decision The evidence in this case also demonstrates that prices for inpatient services are not restrained by prices for outpatient services. F. 207-08. ENH set inpatient rates independent of its outpatient rates and without concern that patients would switch to outpatient services. F. 209. Managed care organizations cannot substitute outpatient services for inpatient services if prices for the latter increase significantly. F. 208. Consistent with the decisions in Rockford Memorial, 898 F.2d at 1284 and Butterworth Health Corporation, 946 F. Supp. at 1291, and which excluded outpatient services because a price increase in inpatient services would not cause consumers to substitute services, there is not substantial evidence in this case to indicate that an increase in the price of inpatient care services would drive consumers to purchase outpatient services.

In defining the relevant product market, the Court acknowledges that some inpatient services can also be performed by specialized hospitals which may be located in the same geographic market. See Forsyth v. Humana, Inc., 114 F.3d 1467, 1477 (9th Cir. 1997). Such facilities might include psychiatric hospitals, rehabilitation hospitals, veterans’ hospitals, military hospitals, children’s and women’s hospitals, and nursing homes. See Hospital Corp., 106 F.T.C. at 436; Tenet Health Care, 186 F.3d at 1048 (excluding a veterans’ hospital from the product market). In this case, both parties agree that specialty hospitals, that may be specialized either in a particular service or for a particular category of patients, are excluded from the market. Complaint Counsel Proposed Order at M; RFF 380. There is no substantial evidence in this case that such speciality facilities were adequate to restrain the exercise of market power in the primary, secondary, and tertiary acute inpatient care markets. As such, they are properly excluded from the relevant product market. c. Demand Analysis VOLUME 144 Initial Decision Respondent argues that the relevant product market should be determined by using a demand-side analysis, which looks at the products sold by each merging firm, and that where a customer purchases several services together, it is those services taken as a whole that constitute the relevant product market. RB at 17. In the case on which Respondent primarily relies, FTC v. Staples, the relevant product market was determined by looking at the availability of substitute commodities and the responsiveness of sales of one product to price changes of another, and not just by whether customers demanded all the products sold by the merging parties. 970 F. Supp. at 1074-75. Indeed, the merging parties, Staples and Office Depot, each sold both consumable office supplies (products that consumers buy recurrently) and other office products, including business machines, computers, and furniture. Id. at 1069. The product market, however, was found to be the sale through office supply stores of only consumable office supplies; it did not include other products (e.g., computers, furniture) also sold by Staples and Office Depot. Id. at 1074. Thus, although the hospitals in the instant case sell services besides inpatient services, just as in Staples, those other services (outpatient) are not included in the relevant product market.

Further, the Court of Appeals for the Seventh Circuit has explicitly rejected an approach that defined the relevant product market as all the services provided by the merging parties and demanded by customers. The Court in Rockford Memorial held that inpatient and outpatient “services are not in the same product market merely because they have a common provider.” Rockford Memorial, 898 F.2d at 1284. The reasoning of the Seventh Circuit in Rockford Memorial applies with equal force here. Simply because the merging parties provide both inpatient and outpatient services does not compel a finding that outpatient services are included in the product market.

d. Summary EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 225 Initial Decision Although managed care organizations negotiate for all acute care hospital services, including both inpatient and outpatient services (RB at 16-17), the evidence clearly demonstrates that managed care organizations cannot substitute outpatient services for inpatient services. As such, outpatient services are not included in the relevant market. The evidence also demonstrates that quaternary services, which require the use of very specialized doctors, nurses, and equipment, and which are not offered at ENH (F. 200, 203), are also not included in the relevant market. Accordingly, Complaint Counsel has met its burden and demonstrated that the relevant product market is general acute care inpatient services sold to managed care organizations, which includes primary, secondary, and tertiary inpatient services.

2. Geographic Market a. Impact of Managed Care As a result of the restructuring of market forces, changing government policies, and technological innovations, the last two decades have brought tremendous change to the health care industry. Tenet Health Care, 186 F.3d at 1050; Long Island Jewish Med. Ctr., 983 F. Supp. at 124-25. During this transformation, hospital systems, previously unaffected by the influences of other markets, have begun to experience the competitive dynamics of the market place. United States v. Mercy Health Serv., 902 F. Supp. 968, 973-75 (N.D. Iowa 1995), vacated as moot, 107 F.3d 632 (8th Cir. 1997). During the 1990’s, these economic motivations led to a substantial wave of consolidations, forcing hospitals to reduce excess capacity while striving to improve the quality of care for patients. Long Island Jewish Med. Ctr., 983 F. Supp. at 124-25. These changes have also substantially affected the antitrust analysis of hospital mergers. Until the early 1980’s, most health insurance plans were “indemnity plans.” F. 153. Under indemnity plans, insurers routinely routinely contracted with all hospitals for services using the same VOLUME 144 Initial Decision formula for all contracts. The patient (or patient’s physician) had virtually complete discretion in choosing the hospital at which the patient would seek services. F. 155. The introduction of managed care, however, constituted a significant change from traditional indemnity insurance. See Mercy Health Serv., 902 F. Supp. at 973. One common feature of all managed care organizations is that – unlike indemnity insurers – a managed care organization exercises discretion in choosing the providers with which it contracts. F. 109, 109, 156. Managed care organizations thus introduced price competition among hospitals, and the managed care company – not not the doctor or patient – became the hospital’s customer for the terms, including price, under which managed care is delivered. F. 109-110; see also Sutter Health, 130 F. Supp. 2d at 1129; University University Health, 938 F.2d at 1213 n.13. Complaint Counsel’s economic expert, Dr. Deborah Haas- Wilson, refers to the price competition found in negotiations between hospitals and the managed care organizations as “first stage” competition. F. 106. In Haas-Wilson’s framework, second stage competition occurs when hospitals compete, primarily on nonprice factors, to attract patients to their hospitals. F. 111. Thus, hospitals initially engage in price competition in order to be included in a managed care organization’s hospital network. F. 109. The ultimate patient is not affected by price because the patient’s contribution, or co-payment, is generally the same regardless of which hospital in the hospital network is selected. F. 104. Second, hospitals compete with other hospitals in these networks through non-price factors, such as quality of care and amenities, in order to attract patients. F. 123.

In this case, the government challenges the merger because of its probable effects on price, i.e., on first stage managed care competition. Accordingly, it is the first stage managed care market that is of critical concern to the antitrust analysis, and it is the review of this market which will determine whether Respondent has market power to raise its prices to anticompetitive levels.

VOLUME 144 Initial Decision b. Overview The proper determination of geographic market is of critical importance in hospital merger cases and is a “necessary predicate” to ascertaining market concentration levels in the relevant market. E.I. du Pont, 353 U.S. at 593. Determination of the geographic market is highly fact sensitive and must be done on a market to market basis. Tenet Health Care, 186 F.3d at 1052; Long Island Jewish Med. Ctr., 983 F. Supp. at 140; see also Brown Shoe, 370 U.S. at 336 (“Congress prescribed a pragmatic, factual approach to the definition of the relevant market and not a formal, legalistic one.”); Freeman Hosp., 69 F.3d at 271 n.16 (“The Supreme Court has repeatedly emphasized that the definition of a geographic market is highly fact-driven and therefore different in each case.”). This determination must be based on a dynamic, “forward looking” analysis which considers not only where consumers have gone in the past for hospital services, but what “practical alternatives” they would have in the future. Freeman Hosp., 69 F.3d at 268-69; see also Tenet Health Care, 186 F.3d at 1055; Mercy Health Serv., 902 F. Supp at 978.

The Supreme Court has defined the relevant geographic market market as “the ‘area of effective competition . . . in which the seller seller operates, and to which the purchaser can practicably turn for for supplies.’” Philadelphia Natl Bank, 374 U.S. at 359 (quoting Tampa Elec. Co. v. Nashville Coal Co., 365 U.S. 320, 327 (1961)). A (1961)). A geographic market has also been described as the area “in “in which the antitrust defendants face competition.” Freeman Hosp., 69 F.3d at 268. “‘A properly defined market includes potential suppliers who can readily offer consumers a suitable alternative to the defendants’ services.” Long Island Jewish Med. Ctr., 983 F. Supp. at 136 (quoting Butterworth Health Corp., 946 F. F. Supp. at 1290). The properly defined market excludes those potential suppliers whose product is sufficiently differentiated or too too far away and who are unlikely to offer a suitable alternative. Long Island Jewish Med. Ctr., 983 F. Supp. at 136 (citation omitted). EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 229 Initial Decision omitted). Courts do not compel “scientific precision” in defining the the geographic market, although they do insist that any such market market be “well-defined.” Sutter Health, 130 F. Supp. 2d at 1120; FTC v. Cardinal Health, Inc., 12 F. Supp. 2d 34, 49 (D.D.C. 1998). 1998). Consequently, “[t]he geographic market selected must, therefore, both ‘correspond to the commercial realities’ of the industry and be economically significant.” Brown Shoe, 370 U.S. at at 336-37.

c. Positions of the Parties Complaint Counsel contends that the relevant geographic market is the area adjacent or contiguous to the three ENH hospitals. CCB at 53-55, see also Attachment 1 (DX 8173, map). Relying on the Merger Guidelines, Complaint Counsel argues that after the merger, ENH demanded large price increases – well above the 5% SSNIP test. See Merger Guidelines § 1.21. Complaint Counsel relies on evidence that managed care organizations tried to avoid ENH price increases through alternative hospital networks that did not include the ENH hospitals; that one managed care organization went so far as to terminate its contract with ENH but was later forced by market realities to negotiate a contract with ENH; and that managed care organizations found that they had to accept ENH’s price increases because they could not satisfy their customers, employers, without ENH in their networks. CCB at 54. In addition, Complaint Counsel points to testimony by ENH’s CEO and COO that when they approved price increases after the merger, managed care organizations’ ability to exclude the ENH hospitals from managed care plans was not a factor in their pricing decisions. CCB at 55. Complaint Counsel thus asserts that these market realities demonstrate that managed care organizations cannot “practicably” turn outside the ENH geographic triangle for substitute hospitals, and that ENH can raise prices by more than a SSNIP without losing so much in sales to hospitals outside its geographic triangle as to make the price increase unprofitable. CCB at 54-55. VOLUME 144 Initial Decision Respondent argues that the relevant geographic market should, at a minimum, include the three ENH hospitals plus Lake Forest, Advocate Lutheran General, Rush North Shore, St. Francis, Condell, and Resurrection. RB at 23. Respondent also contends that other hospitals outside this geographic market, such as Northwestern Memorial, Swedish Covenant, Holy Family, and the Vista hospitals, also place a competitive constraint on ENH. RB at 23, RFF ¶ 489. In determining her proposed geographic market, Respondent’s economic expert, Dr. Monica G. Noether, considered: geographic proximity; patient travel patterns; physician admitting patterns; and market participants’ views on competition. RB at 23. In addition, Respondent points to the rather expansive definitions of geographic market found in previous hospital merger cases. RB at 18. d. Prior Case Law Both parties acknowledge the string of government losses in hospital merger cases over the last decade. CCB at 57; RB at 18. In many of those cases, the government’s failure to prove a relevant geographic market within which a hospital merger would have anticompetitive effects was determinative. E.g., Tenet Health Care, 186 F.3d at 1053 (characterizing the FTC’s failure to produce sufficient evidence of a well-defined relevant geographic market as fatal to the government’s claim); Freeman Hosp., 69 F.3d at 272 (describing the FTC’s failure to meet its burden of establishing the relevant geographic market as dispositive); Mercy Health Serv., 902 F. Supp. at 987 (“The government has failed to establish the relevant geographic area and hence has failed to establish that the merger . . . will likely result in anticompetitive effects.”). These hospital merger challenges are distinguishable because they were decided in the context of prospective mergers, without the benefit of postacquisition evidence.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 231 Initial Decision At issue in these prior hospital merger cases was the probable anticompetitive effect of the merger, specifically whether managed care organizations could practicably defeat a price increase by eliminating the merged entity from their hospital networks and switching to a lower-cost alternative hospital network configuration, through steering or selective contracting. In Tenet Health Care, the court doubted that managed care organizations would “unhesitatingly accept a price increase rather than steer their subscribers to hospitals [outside the geographic market].” 186 F.3d at 1054 (managed care’s “economic interests” would be to resist a price increase). See also Sutter Health, 130 F. Supp. 2d at 1132 (managed care organizations likely to “steer” members away from merged entity’s price increases to other hospitals); Long Island Jewish Med. Ctr., 983 F. Supp. at 130, 144 (managed care representatives testified that if confronted with twenty percent price increase by merged entity, they would “drop” the hospital from their networks, as they had done in comparable situations). As noted, the courts in these cases made certain assumptions regarding managed care organizations’ behavior which depended in large part upon the competitive dynamics existing in each individual market. The post-merger evidence in this case, however, demonstrates that when ENH raised prices more than 5% after the merger, managed care organizations did not utilize alternative hospital network configurations to avoid the price increases. F. 372. Managed care organizations’ inability to selectively contract or steer patients to more distant hospitals to avoid ENH’s price increases is powerful evidence that a local market for hospital services exists in the geographic market and that patients want a local hospital in their managed care plan. F. 398, 408, 414, 446, 455. Given these business and economic realities, managed care testimony is more credible because their post-merger actions, prior to initiation of legal investigation or proceedings, support their testimony. VOLUME 144 Initial Decision Prior cases have traditionally relied on the Elzinga-Hogarty test and patient flow data to establish the geographic market for hospital services. E.g., Freeman Hosp., 69 F.3d at 264; Sutter Health, 130 F. Supp. 2d at 1120-21; Adventist Health Systems/West, 117 F.T.C. at 257, 292. The Elzinga-Hogarty test was developed by Kenneth G. Elzinga and Thomas F. Hogarty in the 1970’s to analyze patterns of consumer origin and destination and to identify relevant competitors of merging entities. Freeman Hosp., 69 F.3d at 264; Elzinga & Hogarty, The Problem of Geographic Market Delineation Revisited: the Case of Coal, 23 Antitrust Bull. 1 (1978); Elzinga & Hogarty, The Problem of Geographic Market Delineation in Antitrust Suits, 18 Antitrust Bull. 45 (1973). The test was developed for the beer and coal industries prior to the development of the Merger Guidelines. F. 212. In the hospital context, the Elzinga-Hogarty test has been used to examine current market behavior through an analysis of hospital service areas and historical patient flow data. Sutter Health, 130 F. Supp. 2d at 1120-21; F. 215. Dr. Kenneth G. Elzinga testified as Complaint Counsel’s expert at trial, however, that his Elzinga- Hogarty test is not appropriate for determining the relevant geographic market for hospital services. F. 216. Indeed, neither party relies on the Elzinga-Hogarty test, although Respondent argues that patient-flow data remains relevant to a geographic market determination. CCB at 53-55; RB at 18-31. As explained by Elzinga, the first problem with use of patient flow data and the Elzinga-Hogarty test is the “payor problem,” which recognizes that, in the hospital industry, managed care organizations pay for hospital services, but their enrollees are the ones who use the services. F. 217. Because patients do not set the price of hospital services, their willingness to travel tells us nothing about their sensitivity to price changes by the merging hospitals. F. 218. In other words, patient flow data is relevant to second stage competition for patients, but provides no useful information about first stage competition for managed care contracts.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 233 Initial Decision The second problem with patient flow analysis is that it incorrectly assumes that if some patients are willing to travel to distant hospitals, then others will also travel in response to a change in hospital prices, thereby incorrectly suggesting a broader geographic market. F. 219. Actually, a “silent majority” of people will not travel in response to a change in hospital prices, and those people can be subject to an anticompetitive price increase. F. 220. Similarly, based on perceptions of hospital services and quality in large urban centers from patients living in surrounding areas, the Elzinga-Hogarty test may overestimate the geographic market to include hospitals in surrounding towns, when in fact, few urban patients are willing to travel to surrounding hospitals for services. See F. 251, 257.

Patient flow data is used by managed care organizations and by hospitals themselves to determine service areas and core service areas. F. 221. Indeed, patient flow data may provide reliable information for hospitals engaging in second stage (non-price) competition for patients because it shows which hospitals patients actually utilize for services. F. 214. However, the question of which hospitals patients ultimately utilize for treatment is a different question than which hospitals patients want available in their managed care organizations’ hospital networks. Therefore, evidence regarding patient flow data, service areas, and the Elzinga-Hogarty test are not probative in determining the relevant geographic market. A key issue in determining the geographic market in this case is is identifying which hospitals managed care organizations need to have in their hospital networks in order to establish viable, competitive networks. This situation is similar to that in Republic Tobacco, where the ultimate consumer was not the purchaser. 381 F.3d at 738-39. In Republic Tobacco, the parties sold cigarette papers papers to distributors and wholesalers, not to retailers and customers. Id. The Seventh Circuit noted that “the evidence presented regarding where wholesalers can practicably sell their VOLUME 144 Initial Decision products (or in other words, where customers and retailers practicably turn for alternative sources of [the product]) is beside the the point when it comes to [geographic] market definition.” Id. Here, Here, the evidence establishes that people select managed care plans plans that include a local hospital – that is, a hospital that is close geographically and in travel time and a hospital where their physician admits patients. F. 226-28, 251, 253-54, 257, 261. Thus, patient flow data and service areas are not reliable in determining substitutability in first stage (price) competition for managed care contracts and are not considered in determining the geographic market. The factors utilized by Respondent’s expert are appropriate, with the exception of patient flow data, which most likely overestimates the geographic market to include certain outlying hospitals not otherwise shown to constrain ENH’s pricing to managed care. Therefore, factors such as market participant views, geographic proximity, travel times, and physician admitting patterns are considered in making the geographic market determination.

e. Market Participant Views Views of market participants are relevant to a determination of the proper geographic market, although they may not be sufficient, sufficient, alone, to establish the geographic market. Freeman Hosp., Hosp., 69 F.3d at 270; see also Tenet Health Care, 186 F.3d at 1054. 1054. Hospital services are a highly differentiated product. F. 102. The commercial realities of the highly competitive health insurance insurance industry in Chicago are that managed care organizations believe that they cannot successfully market a managed care plan without a local hospital. F. 226-27. For example, one managed care care representative stated that people “do not like to drive by a local local hospital and have to go to another hospital.” F. 226. Although Although all of the managed care representatives who testified indicated that selective contracting is used, most managed care plans plans only exclude a small minority of hospitals in the Chicago EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 235 Initial Decision market. F. 158-65. The fact that patients may ultimately travel great great distances for medical care does not alter the analysis. Thus, although patients may use hospitals outside of the geographic market, the evidence demonstrates that, in this market, these outlying hospitals do not constrain Respondent’s pricing and they are not hospitals to which managed care organizations can turn to construct viable hospital networks.

The inclusion of local hospitals in this particular geographic market is critical to hospital networks because, as ENH officials proclaimed, this is an area populated by “senior executives and decision-makers” and it would be “real tough” for any managed care organization and employer “whose CEOs either use [Evanston or Highland Park] to walk from [ENH] and 1700 of their doctors.” F. 227. Many executives live within this geographic market who “make decisions about health benefits for their employers, employees,” and have “immense influence and power with the health plans.” F. 227. According to Elzinga, this testimony is consistent with economic literature which finds that affluent consumers may be less willing to travel because they “impute a higher value to their time and consequently travel becomes more costly to them in the opportunity cost sense.” F. 228.

Prior to the merger, managed care organizations viewed Evanston and Highland Park as substitutes and price constraints for for purposes of building viable hospital networks in the local area. F. F. 229-33. Managed care representatives described the two hospitals hospitals as each other’s “main” competitors or “primary” alternative, thereby permitting managed care organizations to “trade “trade off one for the other” or “work them against each other” in contract negotiations. F. 229. Aetna could constrain Evanston’s prices by utilizing Highland Park (and others) in its network as an alternative (and vice-versa). F. 230. Unicare could exclude Evanston Evanston and satisfy the needs of local customers by offering a network that consisted of Highland Park and other hospitals offering offering services comparable to Evanston (and vice-versa). F. 234. VOLUME 144 Initial Decision PHCS knew that if rate negotiations were not “going well” at either either Evanston or Highland Park, PHCS could turn to the other as as the alternative and use this fact to work the negotiations favorably favorably its way. F. 231. The Unicare representative testified that she could have a viable network comprised of Highland Park, Advocate Lutheran General, Rush North Shore, and St. Francis or Evanston and Lake Forest. F. 234.

Lake Forest, Advocate Lutheran General, Rush North Shore, and St. Francis were the hospitals most consistently included by managed care organizations in their lists of hospitals that compete with ENH. F. 233-42. Aetna’s representative testified that Evanston competed locally with Rush North Shore and St. Francis and that Highland Park competed locally primarily with Lake Forest. F. 235. PHCS’s representative testified that premerger, Advocate Lutheran General, Rush North Shore, and St. Francis were significant competitors to Evanston, that Lake Forest was a significant competitor to Highland Park, and that for purposes of developing its network, she viewed the service and quality of Advocate Lutheran General, possibly Rush North Shore, and possibly Advocate Northside to be comparable to Evanston. F. 236-37. Great West’s representative testified that the main alternatives to ENH were: Advocate Lutheran General, St. Francis, Condell, and Northwestern Memorial. F. 239. Unicare’s representative testified that ENH competes with Lake Forest, Rush North Shore, St. Francis, and Advocate Lutheran General to varying degrees. F. 241. United’s representative testified that Evanston competes with Advocate Lutheran General, Rush North Shore, and St. Francis and that Highland Park primarily competes with Lake Forest and Condell. F. 242.

Moreover, contemporaneous documents from two of the managed care organizations are relevant in informing the Court’s geographic market determination. Contemporaneous documents are are entitled to significant weight. See United States v. United States States Gypsum Co., 333 U.S. 364, 396 (1948); see also United States EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 237 Initial Decision States v. International Business Machines Corp., 1974 WL 899, * 2 2 (S.D.N.Y. 1974); In re Adolph Coors Co., 83 F.T.C. 32, 326 (July (July 24, 1973). When PHCS notified its customers about the merger, it identified “other contract providers within the same geographical area as that of Highland Park Hospital and Evanston,” Evanston,” including: Lake Forest, Advocate Lutheran General, Rush North Shore, St. Francis, and Holy Family Medical Center. F. F. 238. Great West provided its subscribers with a list of hospitals that were in its network that could be alternatives to ENH, including: including: Lake Forest, Advocate Lutheran General, St. Francis, and and to the north, St. Therese and Victory Memorial (now the Vista hospitals). F. 240.

Highland Park, prior to the merger, considered its closest or primary competitor to be Lake Forest, although it was also “reasonably close” to Evanston, Advocate Lutheran General, Rush North Shore, and Condell. F. 244. Highland Park’s president indicated that he believed that managed care organizations could exclude Highland Park from a network and substitute: Evanston, Lake Forest, Advocate Lutheran General, Rush North Shore, St. Francis, and/or Condell. F. 245.

At trial, the CEO of Evanston testified that Condell and Lake Forest were competitors of Evanston, but testified that Highland Park was not a substantial competitor of Evanston. F. 243. This testimony by an interested party, however, is contrary to contemporaneous evidence which clearly demonstrates that Evanston considered Highland Park as a significant competitor throughout the premerger period. F. 243, 247. As such, his testimony on this point is accorded little, if any, weight. Gypsum Co., 333 U.S. at 396.

The contemporaneous evidence and market participants’ views views thus clearly demonstrate that managed care organizations cannot develop a viable managed care plan in this market without: Evanston, Glenbrook, Highland Park, Lake Forest, Advocate Lutheran General, Rush North Shore, or St. Francis in their hospital VOLUME 144 Initial Decision hospital network. F. 233-42. As previously noted, although patients patients may use hospitals outside of the geographic market, the evidence demonstrates that those hospitals do not constrain Respondent’s pricing to managed care organizations and are not hospitals to which managed care organizations can realistically turn turn to construct their local hospital networks. f. Geographic Proximity, Travel Times, and Physician Admitting Practices The evidence demonstrates that geographic realities matter to competition. Managed care organization testimony indicates that the distance an employee must travel is a critical component for employers who are evaluating health care benefit plans. F. 226-27. Because managed care organizations typically market their health care plans to employers, who are concerned about where their employees want to seek hospital care, managed care organizations themselves take into account patient preferences concerning hospital geography when building their hospital networks. F. 111, 114. Consequently, to the extent that employees value convenience, there is a derived demand by managed care organizations for hospitals that are convenient to their enrollees. F. 114-15, 118. Prior hospital merger cases recognize the relevance of patient travel patterns. Tenet Health Care, 186 F.3d at 1053-55 (patient travel patterns a relevant factor in defining geographic market and practical alternatives to the merged hospital); Butterworth Health Corp., 946 F. Supp. at 1292-93 (relying on travel patterns to define define geographic market and identify competitors). In addition to accounting for the physical distance between locations, courts routinely find travel times – which are affected by roads, traffic patterns, and natural impediments such as rivers or mountains – relevant to geographic market definition. See, e.g., Sutter Health, 130 F. Supp. 2d at 1126 (travel time is relevant to a dynamic analysis analysis of the geographic market); J&S Oil, Inc. v. Irving Oil Corp., EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 239 Initial Decision Corp., 63 F. Supp. 2d 62, 68 (D. Me. 1999) (“Simply put, the geographic market for retail gasoline depends on how far individuals individuals are willing and able to travel to purchase the product.”). product.”).

According to a 2001 Lake Forest customer survey report, consumers are willing to travel, on average, up to 16 minutes for emergency care and 35 minutes for an overnight hospital stay. F. 257. It is thus reasonable to presume that, when selecting a managed care plan, these customers would select a plan that includes a local hospital, ideally one within 16 minutes of their home or work. Although this may not be a scientific survey, it does give a glimpse into what consumers in this market consider to be reasonable travel times when selecting a managed care plan. As part of her proposed geographic market, Respondent’s expert, expert, Noether, computed the driving times from Evanston and Highland Park to other area hospitals. F. 256. These distance and driving time components of Noether’s methodology are appropriate appropriate factors to utilize to determine the relevant geographic market. The actual driving time will vary for each patient, depending depending on where he or she lives or works, and may be longer than Noether’s estimates. F. 256; see Attachment 1 (DX 8173, map). map). Adopting Noether’s methodology, it is clear that the hospitals hospitals included in the geographic market (discussed below), are are the closest hospitals to the triangle formed by Evanston, Glenbrook, and Highland Park, in both mileage and driving time: Lake Forest, 6.1 miles (13 minutes) from Highland Park; Advocate Advocate Lutheran General, 10.2 miles (21 minutes) from Evanston; Evanston; Rush North Shore, 3.7 miles (9 minutes) from Evanston; Evanston; and St. Francis, 3.0 miles (8 minutes) from Evanston. F. 266, 272, 281, 287. Together, the average driving distance of these these hospitals is 5.75 miles from the closer of Evanston or Highland Highland Park, while the average driving time is 13 minutes. F. 258. 258. With respect to the two hospitals that Noether proposed for inclusion in the geographic market, but which are found to be outside of the geographic market, Condell and Resurrection, the VOLUME 144 Initial Decision average distance from the closer of Evanston or Highland Park is 12.4 miles, while the average driving time is 24.5 minutes. F. 259. Another component of Noether’s methodology, physician admitting practices, is relevant to establishing the geographic market. The record demonstrates that when the merger was announced, several physicians who had been admitting patients primarily to Highland Park shifted “a lot” of their patients to Lake Forest. F. 269. Managed care organizations, therefore, would want a hospital network that includes Highland Park or Lake Forest for patients of physicians with admitting privileges at both hospitals. See F. 270-71. Such evidence is highly relevant to a dynamic analysis of the geographic market. There is insufficient evidence in the record, however, regarding physician admitting practices at the other relevant hospitals.

g. Hospitals Included in the Geographic Market The evidence does not support Complaint Counsel’s contention that the geographic market should be comprised exclusively of the three merging ENH hospitals and that no additional hospitals could constrain ENH’s pricing. However, the evidence also does not support the inclusion of all nine hospitals that Respondent’s expert selected for her proposed geographic market. Establishing a geographic market for a differentiated product such as hospital services is challenging. As Respondent’s expert stated “in the context of a differentiated product, it’s difficult to draw a bright line that hospitals inside the bright line are all competitors to each other, and then as soon as you cross that line, there’s no competitive pressure that’s exerted.” F. 103; see also E.I. du Pont, 351 U.S. at 392-93. Thus, neither party has proposed a geographic market which fully (and persuasively) addresses the particular market structure characteristics that define competition in this market. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 241 Initial Decision The Court must identify the market which best comports with the totality of the relevant evidence. Upon review of the record, it has therefore determined that the geographic market should properly include a total of seven hospitals: Evanston, Glenbrook, Highland Park, Lake Forest, Advocate Lutheran General, Rush North Shore, and St. Francis. This determination encompasses the three merging hospitals, as proposed by Complaint Counsel, plus an additional four hospitals. F. 262-92. This market includes seven of the nine hospitals, including ENH, in Respondent’s proposed geographic market, but excludes Condell and Resurrection. F. 293-303. The geographic market reflects the market reality, noted by the Seventh Circuit, that hospital services are essentially local. Rockford Memorial, 898 F.2d at 1284-85 (“For highly exotic or highly elective hospital treatment, patients will sometimes travel long distances, of course. But for the most part hospital services are local. People want to be hospitalized near their families and homes, in hospitals in which their own – local – doctors have hospital privileges.”). It is highly probable that the four non-ENH hospitals in the geographic market would have the ability to constrain prices at ENH, either now or in the future, and could be utilized by managed care organizations to create alternate hospital networks. These hospitals comprise the “area of effective competition” (Philadelphia Natl Bank, 374 U.S. at 359) to ENH and provide suitable alternatives for managed care organizations in building and marketing their health plan networks in the geographic market. The three ENH hospitals, Evanston, Glenbrook and Highland Park, have been described as forming a geographic triangle in the North Shore area of Chicago. The evidence establishes that the actual geographic market forms a parallel, but larger, triangle, proximal to and encompassing the ENH triangle. See Attachment 1 1 (DX 8173, map). Should ENH hospitals be excluded from a payor’s hospital network, a patient living within the ENH triangle would only have to drive past one hospital to reach a hospital within VOLUME 144 Initial Decision within the geographic market. The rationale for each hospital’s inclusion in the geographic market determination is discussed more more fully below.

(1) Evanston Evanston Hospital, located in Evanston, Illinois, has more than 400 beds. F. 1, 5. Evanston Hospital provides a wide array of inpatient and outpatient services, from basic hospital services (such as obstetrics) to more intensive services (such as cardioangiogenesis). F. 8. Evanston also offered obstetrical services, pediatric services, a skilled nursing facility, psychiatric care, neurosurgery, radiation therapy, cardiology services, orthopedics, trauma centers, and the Kellogg Cancer Care Center. F. 7. Evanston had .34 residents per bed in 1999. F. 6. (2) Glenbrook Glenbrook, located in Glenview, Illinois, is a community hospital that was developed, built, and opened by Evanston in 1977. F. 9. Glenbrook is located 12.6 miles and 26 minutes west of Evanston. F. 10. Glenbrook has approximately 125 to 150 beds. F. 11. Glenbrook has a Kellogg Cancer Care Center, center of excellence in orthopedics, and does a significant amount of work in neurology, particularly movement disorders. F. 13. Glenbrook Hospital provides inpatient and outpatient services, but it does not provide obstetrics services. F. 12.

(3) Highland Park Highland Park, located in Highland Park, Illinois, has approximately 150 to 200 beds. F. 20, 22. Highland Park is located located 13.7 miles and 27 minutes north of Evanston, along Lake Michigan. F. 21. Prior to the merger, Highland Park offered obstetrical services, including a level II perinatal center, pediatric services, diagnostic services, a skilled nursing facility, a fertility EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 243 Initial Decision center, psychiatric care, neurosurgery, radiation therapy, cardiology cardiology services, including an adult cardiac catheterization lab, an oncology program, and a level II trauma center. F. 24. Highland Highland Park had a medical staff of 562 physicians in 1999. F. 23. 23.

(4) Lake Forest Lake Forest is located 6.1 miles and 13 minutes northwest of Highland Park. F. 266. Lake Forest is a 142 bed hospital that does not provide any tertiary care and had no residents per bed in 1999. F. 267-28. It therefore provides similar services to those provided at Highland Park. In addition, there was a substantial overlap of physicians who had privileges and admitted patients to both Highland Park and Lake Forest prior to the merger. F. 269. Once the merger was announced, a number of these physicians actually shifted a significant volume of their admissions from Highland Park to Lake Forest. F. 269. Lake Forest was identified in contemporaneous PHCS and Great West correspondence to patients as a viable alternative to ENH. F. 270. Managed care representatives identified Lake Forest as a significant competitor to ENH. F. 271. The evidence thus strongly demonstrates that Lake Forest is a significant competitor to ENH and is appropriately included in the geographic market.

(5) Advocate Lutheran General Advocate Lutheran General is located 10.2 miles and 21 minutes minutes west and slightly south of Evanston. F. 272. Advocate Lutheran General is a 521 bed tertiary care hospital that is the largest largest hospital in the Advocate system. F. 273. Advocate Lutheran Lutheran General has a teaching relationship with University of Illinois at Chicago Health Services Center. F. 274. Advocate Lutheran General had .36 residents per bed in 1999. F. 275. In terms terms of range of services, Advocate Lutheran General is similar to to Evanston. F. 276. United’s representative stated that: “Lutheran General is the most comparable facility [to Evanston] from type of VOLUME 144 Initial Decision of services, quality of services, size of facility; however, it is the furthest away. It’s got a bit of geographical disadvantage, but it’s not not terribly far away.” F. 276. Before the merger, patients who went went to the emergency room at Highland Park or Lake Forest with a a heart attack were referred to Advocate Lutheran General for more more advanced care. F. 277. It is significant that ENH, during contract negotiations with PHCS, suggested giving a better rate to PHCS if PHCS excluded Advocate Lutheran General from its hospital network. F. 278. Moreover, Advocate Lutheran General was was identified in contemporaneous PHCS and Great West correspondence to patients as an alternative to ENH. F. 279. Managed care representatives identified Advocate Lutheran General General as a significant competitor to ENH. F. 280. Thus, under the the relevant criteria, Advocate Lutheran General – although a little little further away than the other hospitals in the geographic market market – is similar enough in range of services, according to predominant payors’ views, that it is considered a significant competitor to ENH and is appropriately included in the geographic geographic market.

(6) Rush North Shore Rush North Shore, owned by the Rush system, is located 3.7 miles and 9 minutes southwest of Evanston. F. 281. Rush North Shore has 150 to 200 beds and, as of February 2005, it was affiliated affiliated with Rush-Presbyterian-St. Luke’s. F. 282. The Rush- Rush-Presbyterian affiliation improved the breadth, quality, and the the perception of services offered at Rush North Shore. F. 282. Rush Rush North Shore is geographically close to Evanston, but does not not have the same tertiary facilities that exist at Advocate Lutheran Lutheran General. F. 283. Rush North Shore had .12 residents per bed in 1999. F. 284. Rush North Shore was identified in contemporaneous PHCS correspondence to patients as an alternative alternative to ENH. F. 285. Managed care organizations identified Rush North Shore as a significant competitor to ENH. F. 286. Given Given this evidence and the fact that Rush North Shore’s future EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 245 Initial Decision competitive position may increase as a result of its affiliation with the Rush-Presbyterian system, a dynamic, forward looking analysis analysis of its position in the market indicates that it is and will continue to be a significant competitor to ENH and is appropriately appropriately included in the geographic market. VOLUME 144 Initial Decision (7) St. Francis St. Francis is located 3 miles and 8 minutes south of Evanston on the same street. F. 287. St. Francis has 300 to 400 beds and, as of February 2005, was part of the Resurrection System. F. 288. St. Francis’s services range from cardiology and obstetrics to general surgery. F. 288. St. Francis is geographically close to Evanston, but does not have the same tertiary facilities that Advocate Lutheran General has and has less of a reputation as an equivalent facility. F. 289. St. Francis had .36 residents per bed in 1999. F. 290. St. Francis was repeatedly identified in contemporaneous PHCS and Great West correspondence to patients as an alternative to ENH. F. 291. Moreover, managed care organizations identified St. Francis as a competitor to ENH. F. 292. Thus, St. Francis is considered a significant competitor to ENH – geographically close, and a competitor on primary and secondary services, although without the same level of tertiary services available at Evanston, and is appropriately included in the geographic market. h. Hospitals Excluded from the Geographic Market The geographic market in this case has been described as a “moving target.” RB at 19. Indeed, neither party’s proposed geographic market is supported with scientific precision. The Complaint describes the geographic market as: “the densely populated corridor that runs for about 15 miles north-south along the the shore of Lake Michigan, and extends roughly ten miles west of of the Lake.” Complaint ¶ 17. Complaint Counsel later suggested that, hypothetically, the geographic market could be “expanded to encompass a larger geographic area in which additional hospitals are are located, such as Holy Family Medical Center, St. Francis Hospital, Lake Forest Hospital, Advocate Lutheran General Hospital, and Rush North Shore Hospital.” Complaint Counsel Interrog. Answers at 20. However, Complaint Counsel now contends EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 247 Initial Decision contends that the geographic market should only include the three ENH hospitals. CCB at 9, 54.

Similarly, Respondent proposed a minimum geographic market of nine hospitals, but qualified that determination with a list of additional hospitals that “could potentially” be in the market. Respondent argues that Holy Family, Swedish Covenant, the two Vista hospitals, and even teaching hospitals such as Northwestern Memorial should also be considered for inclusion in the geographic market. RB at 23. As noted, the Court adopts Respondent’s proposed minimum market, with the exception of Condell and Resurrection. F. 262-92.

Complaint Counsel’s proposed geographic market, comprised only of the three ENH hospitals, is found to be too limited and not sufficiently forward-looking. The Court is mindful that during the last three years ENH has been under investigation by the Commission, which may have acted as a constraint against ENH imposing even further price increases on managed care organizations. The geographic market recognizes that in the face of such future increases, there are alternate providers to which managed care organizations could turn for hospital services. Each of the hospitals proposed for the geographic market by the parties but found by the Court to be outside the geographic market are discussed in detail below.

(1) Condell Condell was included by Respondent in its proposed geographic geographic market and some market participants mentioned Condell Condell as generally competing with ENH. Condell is a 163 bed hospital and had no residents in 1999. F. 294, 296. The evidence as a as a whole does not warrant its inclusion in the geographic market. market. The market participants who commented specifically on Condell mentioned significant proximity issues, stating that it was “further west” than Lake Forest, which is the principal competitor VOLUME 144 Initial Decision north of Highland Park. F. 297. Condell is 12.7 miles and 24 minutes minutes (driving time) northwest of Highland Park. F. 293. Thus, the the drive time to Condell is substantially beyond the 16 minute drive drive time noted in the informal Lake Forest survey that people living within the area are willing to travel for emergency care. F. 257. Moreover, Condell does not offer any additional services which which are unavailable at Highland Park and Lake Forest. See F. 295. 295. Accordingly, Condell is not included in the geographic market. market.

(2) Resurrection Resurrection was also included in Respondent’s proposed geographic market. Resurrection is 12.1 miles or 25 minutes (driving time) southwest of Evanston Hospital. F. 298. Like Condell, the drive time to Resurrection is substantially beyond the 16 minute drive time noted in the Lake Forest survey that patients within the area are willing to travel for emergency care. F. 257. Resurrection had 350 staffed beds and .17 residents per bed in 1999. F. 299-300. The Resurrection system includes St. Francis, which is included in the geographic market, and there is conflicting testimony regarding whether the Resurrection system negotiated all of its hospitals as one contract or separately. F. 302. The Resurrection system is large and was described by one managed care organization as a “system which we really need to keep.” F. 301. Therefore, managed care organizations may value Resurrection Medical Center only because they value the system. In addition to significant proximity and travel time issues, none of the managed care representatives testified that Resurrection was a significant competitor to ENH. Thus, there is insufficient evidence to support including Resurrection in the geographic market.

(3) Holy Family Holy Family is 11.3 miles or 23 minutes (driving time) from Evanston Hospital. F. 305. Holy Family has 260 staffed beds and .02 EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 249 Initial Decision .02 residents per bed. F. 305. Although PHCS contemporaneous correspondence mentions Holy Family as an alternative to Evanston Evanston (F. 305), there is virtually no evidence in the record, including testimony of managed care representatives, which would would indicate that Holy Family constrains the prices of Evanston or or is in any way a significant competitor. F. 305. Moreover, as is the the case with Condell and Resurrection, proximity and travel times times mitigate against Holy Family being a significant competitor to to ENH. Given these substantial limitations, the evidence does not support including Holy Family in the geographic market. (4) Swedish Covenant Swedish Covenant is 6.8 miles or 19 minutes (driving time) south of Evanston, and as of February 2005, had 324 beds. F. 306. In 1999, Swedish Covenant had .13 residents per bed. F. 306. The managed care representatives did not mention Swedish Covenant as a significant competitor to ENH, nor is there sufficient evidence from ENH that it considered Swedish Covenant as a viable competitor or that Swedish Covenant otherwise constrained ENH’s prices. F. 306. The evidence does not, therefore, support including this hospital in the geographic market.

(5) Vista Hospitals The Vista hospitals include Vista Health St. Therese and Vista Health Victory Memorial, both located in Waukegan in northern Illinois, with Victory Memorial located “almost up to Wisconsin.” F. 307. The Vista hospitals are an average of 15.9 miles or 30 minutes (driving time) north of Highland Park. F. 307. Although Great West lists the Vista hospitals as an alternative in contemporary correspondence (F. 240, 307), given the outlying proximity issues of distance and travel times, and the almost complete lack of payor testimony and evidentiary support as to their competitive constraint on ENH, there is no foundation to include these northern Illinois hospitals in the geographic market.

VOLUME 144 Initial Decision (6) Teaching Hospitals Teaching hospitals in downtown Chicago, such as Northwestern Memorial, Rush-Presbyterian-St. Luke’s, and the University of Chicago, may compete with ENH for more sophisticated or tertiary services. F. 242, 248, 308. However, as previously noted, when selecting a managed care plan, employees and employers want a plan that includes a local hospital. This is true even though patients may be willing to travel further for “exotic” services. Rockford Memorial, 898 F.2d at 1284-85. The court in Long Island Jewish Medical Center concluded that there were two relevant geographic markets – one for primary and secondary care and the other for tertiary care – to account for evidence that “patients prefer to receive health care treatment relatively close to their homes,” but also that patients are willing to travel further for certain services such as specialty tertiary care. Long Island Jewish Med. Ctr., 983 F. Supp. at 141. Thus, although teaching hospitals may compete with ENH in the second stage for patients with more complex needs, they do not constrain ENH’s first stage prices to managed care organizations, and are thus not properly considered as part of the geographic market.

i. Summary The evidence establishes that when employers select a managed managed care plan, they prefer a plan that provides the most choice choice – specifically the choice, or option, of using a local hospital. hospital. F. 115, 118. Therefore, to create a viable hospital network, network, managed care organizations in this market must include local hospitals. The Court, guided by relevant case law, has defined defined the geographic market on the principle that such determination must undergo a dynamic “forward looking” approach approach to Clayton 7 analysis which considers the probable competitive responses from competing hospitals, managed care organizations, and, ultimately, consumers. Freeman Hosp., 69 F.3d F.3d at 268; Mercy Health Serv., 902 F. Supp. at 978. Based on the EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 251 Initial Decision the evidentiary record, it seems reasonable that in the face of probable, future anticompetitive pricing, managed care organizations organizations could create a network excluding the ENH hospitals and including the next proximal set of geographically close hospitals hospitals where consumers could go to seek “practical alternative” alternative” acute care inpatient hospital services. Freeman Hosp., 69 F.3d at 268-69. Thus, the hospitals included in the geographic market are: Evanston, Glenbrook, Highland Park, Lake Forest, Advocate Lutheran General, Rush North Shore, and St. Francis. This This geographic market determination best comports with the market realities and the evidentiary record. C. Probable Effects on Competition “The Supreme Court has adopted a totality-of-the- circumstances approach to [Section 7], weighing a variety of factors to determine the effects of particular transactions on competition.” Baker Hughes, 908 F.2d at 984. The “Supreme Court and appellate courts acknowledge the need to adopt a flexible approach in determining whether anticompetitive effects are likely to result from a merger.” Oracle Corp., 331 F. Supp. 2d at 1111. Courts require that the merger be “functionally viewed, in the context of its particular industry” and “only a further examination of the particular market – its structure, history and probable future – can provide the appropriate setting for judging the probable anticompetitive effect of the merger.” Brown Shoe, 370 U.S. at 321, 322 n.38; In re Weyerhauser Co., 106 F.T.C. 172, 278 (Sept. 26, 1985). 1. Anticompetitive Effects Having determined the relevant product and geographic markets, markets, the Court now turns to an analysis of the competitive effects of the merger. In doing so, it first undertakes a structural analysis of the probable anticompetitive effects of the merger, specifically an examination of market concentration in the relevant VOLUME 144 Initial Decision relevant market. Then, the evidence of contemporaneous and postpost-merger price increases is reviewed. a. Market Concentration Market concentration under the Merger Guidelines is measured by the Herfindahl-Hirschman Index (“HHI”). Merger Guidelines § 1.5. “The HHI is the most prominent method of measuring market concentration, commonly used by the Justice Department, the FTC and the courts in evaluating proposed mergers.” Butterworth Health Corp., 946 F. Supp. at 1294. The HHI is calculated by summing the squares of the market shares of every firm in the relevant market. University Health, 938 F.2d at 1211 n. 12; Merger Guidelines § 1.5. “For example, in a market with six firms with market shares of 25%, 20%, 20%, 15%, 10%, and 10%, the HHI is 1850 (25<2> + 20<2> + 20<2> + 15<2> + 10<2> + 10<2> = 1850).” University Health, 938 F.2d at 1211 n. 12; Merger Guidelines § 1.51 n.17. Under the Merger Guidelines, a market in which the post-merger HHI is above 1800 is considered “highly concentrated,” and a merger in a highly concentrated market that increases the market’s HHI by over 100 is presumed to be “likely to create or enhance market power or facilitate its exercise.” University Health, 938 F.2d at 1211 n. 12; Butterworth Health Corp., 946 F. Supp. at 1294; Merger Guidelines § 1.51.

The geographic market, as proposed by Complaint Counsel’s expert, Haas-Wilson, included only the ENH hospitals (Evanston, Glenbrook, and Highland Park), giving ENH a monopoly in the provision of inpatient services sold to managed care organizations. organizations. CCB at 55. Under Complaint Counsel’s proposed market, the HHI would be 10,000, the highest possible HHI number. number. Merger Guidelines § 1.51 n.17. Complaint Counsel asserts asserts that even using Respondent’s proposed geographic market, market, the post-merger HHI level corresponds to a market that is “highly concentrated,” and the merger is “presumed” likely to EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 253 Initial Decision “create or enhance market power.” CCB at 55-56. Complaint Counsel further argues that ENH cannot demonstrate that the market market share and market concentration figures give an “inaccurate account” of the merger’s effects, where the large post-merger price price increases show that the anticompetitive effects predicted by the the market structure analysis are accurate. CCB at 56. Alternatively, Complaint Counsel argues that Respondent’s market share can be determined based on Evanston’s contemporaneous estimation of its combined core service area (“CCSA”) to compute an HHI of 3426, with a corresponding increase of over 1000. CCB at 9. In 1999, ENH identified the market share in its CCSA as: Evanston, 44%; Highland Park, 11%; Lake Forest, 3%; Advocate Lutheran General, 7%; Rush North Shore, 14%; St. Francis, 7%; downtown teaching hospitals, 7%; and other, 7%. F. 325. Respondent contends that ENH’s “core service area” is not the same as an appropriately defined geographic market and that the information contained in these documents is an unscientific, unverified, and much less accurate form of patient flow data. RRB at 21 n.16. The Court agrees with Respondent that service areas are not the same as geographic market, in part because they are based upon patient flow data which, as previously noted, is more relevant to stage two competition for patients. See Tenet Health Care, 186 F.3d at 1052. Thus, use of ENH’s estimate of a 55% market share in its CCSA is not an appropriate method for determining HHI concentration levels.

Respondent argues that Complaint Counsel’s proposed market of of only the merging parties and Complaint Counsel’s use of ENH’s ENH’s estimation of its CCSA to determine HHI statistics are incorrect. RB at 19; RRB at 55. Respondent’s expert, Noether, computed a post-merger HHI of 1919, an increase of 222 from premerger levels, based on Respondent’s proposed geographic market. F. 314, 323. In addition, Respondent argues that the HHI statistics give an inaccurate account of the merger’s probable effects effects on competition because the evidence shows that: the quality VOLUME 144 Initial Decision quality of care at Highland Park has improved, and is continuing to to improve, dramatically; there are currently several hospitals both both within and outside of the relevant geographic market that are viable alternatives to ENH and which exercise a constraint on ENH’s pricing; and existing hospitals have been repositioning to expand their existing services and add new ones. RRB at 55-56; see see also RB at 20-28, 56-59, 67-107.

As described in section III.B.2 supra, the geographic market is larger than Complaint Counsel’s proposed three hospital market, yet smaller than Respondent’s proposed nine hospital market. The Court’s determination of relevant geographic (and product) market yields an HHI calculation which lies between the parties’ estimates. Adopting and utilizing Respondent’s net inpatient revenue determinations, but excluding Condell and Resurrection hospitals from the calculation, leads to a post-merger HHI of 2739, with an increase of 384. F. 316-19.

The HHI figure of over 2700 is calculated using Respondent’s expert’s market share figures. Noether acknowledged that she was not able to calculate exact market shares given the available data. F. F. 309. Noether did, however, calculate proxy shares using the best best available information, contained in the Medicare Cost Reports, Reports, without substantive critique by Complaint Counsel. F. 309; 309; CCRFF ¶¶ 508-14. The Medicare Cost Reports provide information on total net revenues, both inpatient and outpatient, across all managed care organizations for each hospital. F. 309. Noether provided revenues for inpatient services combined with outpatient services and for inpatient services alone. F. 309. Only the the inpatient revenues are used, to conform with the appropriate product market previously established. See supra Section III.B.1. Noether properly treated St. Francis and Resurrection as separate hospitals, although the hospitals had merged in the late 1990’s. F. 311. Indeed, both Advocate Lutheran General and Rush North Shore Shore are part of larger systems, but it would be improper to include include revenue from other hospitals in those systems in the EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 255 Initial Decision determination of market shares, as such other hospitals are not in the the relevant geographic market. The post-merger HHI of over 2700 2700 is substantially above the Merger Guidelines’ threshold of 1800 to consider a market “highly concentrated,” and the increase of of over 350 far exceeds the Merger Guidelines’ threshold of 100 to to presume that the merger is “likely to create or enhance market power or facilitate its exercise.” Merger Guidelines § 1.51. In 1999, within the relevant geographic and product market, Evanston and Highland Park had a combined market share of approximately thirty-five percent. F. 317, 322, 324. Lake Forest had a market share of { }, Advocate Lutheran General had a market share of { }, Rush North Shore had a market share of { }, and St. Francis had a market share of { } F. 322. Respondent’s post-merger market share increased to approximately forty percent by 2002, with the other four hospitals in the geographic market all losing some market share in the three year period from 1999 to 2002. F. 322. These statistics demonstrate not only that this was a concentrated market in 1999, but that, over time, while ENH’s concentration level has been steadily increasing, ENH’s competitors have lost market share.

Courts have traditionally considered the market share of the combined firm to determine whether the merger is likely to cause anticompetitive effects. Under Philadelphia National Bank, a postmerger market share of thirty percent or higher presents the threat of undue concentration. Philadelphia Natl Bank, 374 U.S. at 364; see also Oracle, 331 F. Supp. 2d at 1110. Here, ENH’s post-merger market share of thirty-five percent in 1999, which increases to forty percent in 2002, is well above the thirty percent threshold established in Philadelphia National Bank. F. 317, 324. Thus, all of the available methods for determining market concentration lead to the same conclusion – that this is a highly concentrated market and that the merger is likely to create or enhance ENH’s market power or facilitate its exercise. This presumption is further supported by the post-merger evidence of ENH’s price increases. VOLUME 144 Initial Decision Complaint Counsel has demonstrated sufficient market concentration to predict probable anticompetitive effects. Because this is a consummated merger case, however, Complaint Counsel was also able to provide contemporaneous and post-acquisition evidence regarding the merger’s impact on ENH’s prices to managed care.

b. Contemporaneous and Post-Acquisition Evidence (1) Introduction Section 7 of the Clayton Act was intended to arrest the anticompetitive effects of market power in their incipiency. Brown Shoe, 370 U.S. at 317. As previously noted, the test of a violation of of § 7 is whether, at the time of suit, there is a “reasonable probability” that the acquisition is likely to result in the condemned condemned restraints. E.I. du Pont, 353 U.S. at 607. Section 7 “requires not merely an appraisal of the immediate impact of the merger upon competition, but a prediction of its impact upon competitive conditions in the future.” Philadelphia Natl Bank, 374 374 U.S. at 362. There “is no requirement that the anticompetitive power manifest itself in anticompetitive action before § 7 can be called into play. If the enforcement of § 7 turned on the existence of of actual anticompetitive practices, the Congressional policy of thwarting such practices in their incipiency would be frustrated.” FTC v. Procter & Gamble, Co., 386 U.S. 568, 577 (1967). Indeed, the Supreme Court in Procter & Gamble stated that the appellate court “misapprehended . . . the standards applicable in a § 7 proceeding” where the appellate court found that the postpost-acquisition evidence did “‘not prove anti-competitive effects of of the merger.’” Procter & Gamble, 386 U.S. at 576. See also Hospital Corp. of Am., 807 F.2d at 1389 (“Section 7 does not require require proof that a merger or other acquisition has caused higher prices in the affected market. All that is necessary is that the merger merger create an appreciable danger of such consequences in the EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 257 Initial Decision future.”). Accordingly, Complaint Counsel is not required to provide provide evidence of actual anticompetitive post-merger effects, only only evidence that anticompetitive effects are probable. It is well settled that contemporaneous and post-acquisition evidence may properly be considered in determining whether the probable effect of a merger will be a substantial lessening of competition. E.g., Purex Corp. v. Procter & Gamble Co., 664 F.2d 1105, 1108 (9th Cir. 1981); United States v. Falstaff Brewing Corp., 383 F. Supp. 1020, 1025 (D.R I. 1974); see also FTC v. Consolidated Foods Corp., 380 U.S. 592, 598 (1965). The Supreme Court, in E.I. du Pont, relied upon “the plain implications of the contemporaneous documents” to determine the motives of the acquisition. E.I. du Pont, 353 U.S. at 602; see also University Health, 938 F.2d at 1220 n.27 (evidence from defendants’ premerger documents evincing an intent to eliminate competition through the proposed acquisition can help establish the government’s prima facie case.). Similarly, post-acquisition evidence is appropriately considered where it “tends to confirm, rather than cast doubt upon, the probable anticompetitive effect” of a merger. Consolidated Foods, 380 U.S. at 598. However, post-acquisition evidence that can be manipulated by the party seeking to use it is entitled to little weight, in part because the actions may have been taken to “improve [the defendant’s] litigating position.” Hospital Corp. of Am., 807 F.2d at 1384; see also General Dynamics, 415 U.S. at 504-05. With respect to the post-acquisition evidence, Respondent argues argues that its expert’s analysis shows a smaller price increase relative to other hospitals than Complaint Counsel’s expert’s analysis; that not all viable competitively benign explanations have have been ruled out; and, that Respondent’s price increases are a result of its learning about demand for its services and that its premerger prices at Evanston were, on average, below market. RB at at 34. In addition, Respondent argues that Evanston and Highland Park were not close substitutes and therefore, ENH, the combined entity, could not have had greater bargaining power than the VOLUME 144 Initial Decision hospitals did before the merger. RB at 35. As discussed more fully below, the Court finds these arguments without merit. Complaint Counsel has presented contemporaneous and postacquisition evidence which establishes that ENH exercised its enhanced post-merger marker power and obtained post-merger price increases substantially above its premerger prices and significantly larger than price increases obtained by other comparison hospitals. F. 326-755. This evidence confirms the predictive assessments made by the structural market analysis. F. 309-25. Complaint Counsel presented contemporaneous documents, testimony of managed care organizations, and empirical analysis to establish the post-merger price increases.

In the hospital services market, determination of relative prices must take into account a variety of factors. First, approximately half of ENH patients are covered by government insurance through Medicare or Medicaid. F. 135. For these patients, hospitals are reimbursed at a rate set by the government. F. 128. Second, managed care organizations negotiate contracts that include fixed rates (per case or per day) and discount off charges rates. F. 173-80. Thus, contract rates cannot be directly compared with each other because they arise through different payment methodologies. Third, relative prices vary depending on patient mix because not all inpatient hospital stays require the same resources for treatment. F. 735-37. Some patients, even those with the same condition, may be sicker and may require more treatment resources than the patient who is less sick. F. 735. Fourth, data on hospital prices is not maintained in a consistent or complete fashion. Indeed, only four managed care organizations provided usable data for analysis in these proceedings and even that data had limitations. F. 491-94, 500. Respondent’s expert, Noether, relied only on data provided by managed care organizations. As Noether indicated, “there were a number of problems with the data that made the measure of price EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 259 Initial Decision certainly less than fully accurate.” F. 470. Noether concluded that the claims data provided by managed care organizations could be used in “forming [her] opinion and reaching [her] conclusions,” but but cautioned that her findings should be considered “in the context context of all the other evidence in the case.” F. 471. Recognizing the limitations of all of the data, Complaint Counsel’s expert, Haas- Haas-Wilson, provided an analysis which utilized four different data data sources. F. 469-692. Reviewing the evidence, the Court concludes that Haas-Wilson’s conclusions are more reliable, in part part because they present more detailed and consistent findings which were validated throughout each of the different data sources. sources. F. 469-692. In addition, contemporaneous documents and testimony of managed care organizations affirm the conclusions of of Haas-Wilson and provide evidentiary support for her empirical analysis. F. 328-468. Given the breadth and variety of this evidence, evidence, Complaint Counsel’s expert’s conclusions on relative price price increases are found credible and persuasive. The merger violates the Clayton Act because the merger reduced reduced competition in the relevant market and enhanced ENH’s market power, regardless of whether ENH’s prices have yet risen to to a supra competitive level. Since the enactment of the Hart-Scott- Hart-Scott-Rodino Antitrust Improvements Act of 1976 (15 U.S.C. § U.S.C. § 18a), most enforcement actions are initiated prior to the proposed merger. Therefore, there are very few recent cases which have examined post-merger evidence and there is relatively little case law regarding the proper analysis of price changes in a consummated merger under Clayton 7. Courts have indicated that, consistent with the Merger Guidelines’ SSNIP test, a 5% price increase is an appropriate value against which to judge a merger. Sutter Health, 130 F. Supp. 2d at 1129; Mercy Health Serv., 902 F. Supp. at 980-81; see also CF Industries, Inc. v. Surface Transp. Bd., Bd., 255 F.3d 816, 823-24 (D.C. Cir. 2001). Analysis of relative price increases in the consummated merger context is fact intensive intensive and depends upon the economic realities of each market. Thus, the focus of this analysis rests, by necessity, on the quality of of the factual evidence presented by the parties. As discussed below, VOLUME 144 Initial Decision below, the evidence in this case is more than sufficient for the Court Court to reach its conclusions.

Respondent contends that “in order to utilize evidence of price increases to prove that a firm possesses market power, that evidence must be accompanied by proof that the price increased above a competitive level and can be sustained at that level over a period of time, or is associated with a reduction of output.” RB at 36. In support of this contention, Respondent cites no Section 7, Clayton Act cases. RB at 36 n.23. Complaint Counsel responds that it is not required to demonstrate a decrease in output, but even if it were, output decreased as a result of ENH’s higher prices including the temporary loss of its contract with One Health and the patients who lost coverage due to the increased cost of health care. CCRB at 19- 20.

The evidence indicates, but does not conclusively establish, that Respondent’s prices were supra competitive. Indeed, Complaint Counsel did not attempt to compare ENH’s prices to a competitive level, instead focusing on ENH’s price increases relative to other hospitals’ price increases. CCB at 44-45; F. 469-97. ENH’s expert, Noether, compared ENH’s inpatient and outpatient prices to inpatient and outpatient prices charged by other hospitals. F. 798, 831. { } F. 831-35. { } F. 831, 833. Thus, Respondent’s own expert’s analysis indicates that ENH’s prices exceed the prices charged by each of the other four hospitals in the geographic market. The evidence, therefore, strongly suggests that prices did rise to a supra competitive level without a reduction of output, although the evidence on that issue is not conclusive. However, as noted earlier, Complaint Counsel need not make such a definitive showing in order to find Respondent in violation of Section 7. A review of the evidence demonstrates that: (1) ENH achieved substantial price increases as a result of the merger; (2) empirical analysis establishes that ENH’s prices rose relative to other EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 261 Initial Decision comparison hospitals; and (3) explanations of price increases other other than market power are ruled out. F. 326-755. The evidence therefore demonstrates that the relative price increases were the result of ENH’s enhanced market power, achieved through elimination of a competitor as a consequence of the merger. Complaint Counsel’s post-acquisition evidence of relative price increases, which confirms the structural evidence of concentration, concentration, clearly establishes the probable anticompetitive effects of the merger necessary to find a violation of Section 7 of the the Clayton Act.

(2) Respondent Achieved Substantial Price Increases as a Result of the Merger Contemporaneous evidence demonstrates that ENH sought and achieved substantial price increases as a result of the merger. It is clear that the primary motivation for the merger was economic, although the parties to the merger were well aware of the importance of quality and brand image, especially for stage two competition for patients. E.g., F. 45, 343, 368. As noted, such evidence is entitled to significant weight. Managed care testimony in this case is confirmed by the contemporaneous actions of the managed care organizations and therefore such testimony is considered credible, despite the fact that the managed care organizations have an interest in the outcome of this litigation.

(a) Evanston and Highland Park Sought Market Power from the Merger As early as 1994, the CEOs of the merging parties shared the view that hospitals should “stand united” in order to get “better pricing” and “leverage” from the managed care organizations. F. 29. 29. In 1998, as merger discussions began, the CEOs wrote: “[p]ricing pressures will escalate on healthcare providers from both both government and managed care.” F. 331. Their recommendations included: “[s]trengthen negotiating positions with with managed care through merged entities and one voice” and VOLUME 144 Initial Decision “[m]maintain and enhance local community ties for long-term success success – make indispensable to marketplace.” F. 331. Evanston’s CEO told managers and the Evanston board that the merger would “[i]ncrease our leverage, limited as it might be, with the managed care players, and help our negotiating posture.” F. 353. Evanston’s CEO candidly admitted at trial that one of the goals of the merger was to obtain better prices and better terms from managed care. F. 330.

The evidence further establishes that Evanston wanted to merge with Highland Park in no small part to eliminate a competitor within the geographic market. Evanston’s management reminded its board of the risk of “not undertaking [the] merger.” F. 334. Skokie Valley Community Hospital, located three miles to Evanston’s south, had been a “sleeping dog” competitor until it affiliated with the Rush system of hospitals, at which point Rush renamed it Rush North Shore, invested heavily in the hospital, and the former “sleeping dog” awoke to become a stronger, more competitive hospital. F. 334. The point of the story was clear: if Evanston did not act first, the same problem could occur to Evanston’s north, and another hospital system would come in to further strengthen the competitive position of Highland Park. F. 334. Thus, one of Evanston’s goals was to stop Highland Park from competing with it. The merger was seen by Evanston as an “opportunity to join forces and grow together rather than compete with each other.” F. 333.

Highland Park similarly sought to eliminate a competitor within within the geographic market. Highland Park’s board chairman recognized that the merger would allow the two health care providers to “[s]top competing with each other.” F. 341. Highland Park management hoped that a merger with Evanston would build “negotiating strength with payers.” F. 340. Evanston, Glenbrook, and Highland Park would form a triangle and “together would have have a significant market penetration in these very affluent, attractive communities.” F. 339 (emphasis added). Highland Park EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 263 Initial Decision saw Evanston, Lake Forest, Northwest Community, and Condell as as merger candidates, the attractiveness of each turning on “how concentrated could this market be for us.” F. 340. Highland Park believed that merging with Evanston would build the greatest pricing strength with managed care organizations. F. 340. In 1999, Highland Park’s CEO and board convened to frankly discuss the merger. F. 342. The CEO described the problem: the reality in my view is that we are not looking at a rosie future economically on this site. Neither are they. We are not looking at the opportunity to control this market individually. The largest . . . payors in this arena have consolidated and are big enough, strong enough, and probably bent on assuring that the physicians who practice here and at Evanston and the institutions don’t make a hell of a lot of money. That is the reality and I am not even laying that on the insurers I am laying that on the employers. The same speech I have made over and over.

F. 342.

The solution was the merger with Evanston: I think the ultimate benefit to these communities is pretty positive. There are cost economies, there are quality issues, there are ways to at least I think to push back on the managed care phenomenon and get get the rates back where they ought to be if you are a a big enough concerted enough entity which is important enough to the employers in this community. I think it would be real tough for any of of the Fortune 40 companies in this area whose CEOs either use this place or that place to walk from VOLUME 144 Initial Decision from Evanston, Highland Park, Glenbrook and 1700 1700 of their doctors.

F. 343. At that same meeting, there was a comment on “the economic benefit of not being out there doing battle with one another in what will be a common battle ground if you want to call it that.” F. 345. The above evidence clearly shows the primary motivation for the merger was to attain enhanced market power which could be utilized by the merged entity in negotiations with the managed care organizations. Such market power, however, could only be obtained through the elimination of a competitor in the geographic market.

The antitrust laws afford neither solace nor escape from the rigors of competition induced by managed care. In Hospital Corporation of America, the Seventh Circuit upheld an FTC challenge to mergers that would have reduced the number of owners/managers of Chattanooga hospitals. The Court recognized that hospitals were under “great pressure” from managed care organizations (and the federal government) to “cut costs.” 807 F.2d at 1389. However, efforts by hospitals to resist this pressure through mergers that confer market power may violate the Clayton Act. The “fewer the independent competitors in a hospital market, the easier they will find it ... to frustrate efforts to control hospital costs.” Id. The Court opined that the Commission was entitled to make such efforts by hospitals “less effective by preserving a substantial number of competitors.” Id. As noted by the Seventh Circuit, hospitals thus risk violating the Clayton Act by acquiring market power to shield them from the pricing pressures of managed care. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 265 Initial Decision (b) ENH Sought to Increase Prices Through Contract Negotiations and Chargemaster Chargemaster Increases Even before the merger was fully consummated, Respondent made extensive efforts to exercise its enhanced market power by increasing its charges to managed care organizations. In December 1999, ENH negotiators sent consent to assignment agreements to managed care organizations to assign the higher of the Evanston or Highland Park rates. F. 349. In January 2000, while the status of many contracts was still in limbo, Chan, who was responsible for managed care contracting for Highland Park, instructed ENH’s billing department to “continue to use the current Highland Park Hospital rates” – in some instances in which Highland Park had higher rates – until all of the hospital contracts had been renegotiated. F. 350.

ENH decided that all three hospitals would operate under one contract, with one price, and one chargemaster, even though other multi-hospital systems in the Chicago area charged different rates for different hospitals. F. 355-66. ENH demanded the same rate regardless of the level or complexity of services provided at each hospital. F. 359. ENH successfully moved all three hospitals to the same contract and equalized the charges for all three facilities postmerger. F. 364. Indeed, under ENH’s billing system, managed care organizations can not “distinguish between services at the three hospitals” to determine which services are rendered at a particular hospital in the system. F. 362. Though Evanston had previously included Glenbrook in its contracts and chargemaster prior to the merger, Glenbrook was developed and built by Evanston (F. 2) and had never been an independent competitor like Highland Park. This consolidation into one contract enabled ENH to charge higher prices at all three hospitals.

The record reveals further strategies by the newly-merged ENH ENH to maximize its pricing. One such method utilized by ENH in in negotiations with managed care organizations was to seek the VOLUME 144 Initial Decision higher of Evanston’s or Highland Park’s existing contract rates and and add a “premium” on top of that. F. 367. The “premium” represented one of ENH’s self described “benefits” of the merger and was depicted by Highland Park’s vice president of business development as resulting from the “additional negotiating power and and leverage with the payors.” F. 367. Bain & Company (“Bain”), an an economic consulting firm, advised ENH that it could “sell” these these higher rates to managed care by emphasizing “the value ENH ENH brings to a payor’s network” such as brand, patient access, cost cost management, and quality, in order to “[j]ustify premium pricing pricing (i.e., above the competitive average).” F. 368. According to ENH, one of the “accomplishments” of the merger was the renegotiation of managed care contracts, which collectively resulted in an increased annualized economic value of at least $18 million for ENH. F. 370. Evanston “had never achieved” a price increase as high as $18 million prior to the merger. F. 371. Although ENH argues that pricing “above the competitive average” does not mean supra competitive pricing, it is clear from the context of all of the contemporaneous documents that one of ENH’s primary motives for the merger was to obtain supra competitive prices. The record further demonstrates that, as a result of its enhanced enhanced market power, ENH succeeded with numerous managed care organizations in negotiating discount off charges arrangements, arrangements, which were “more favorable” for ENH. F. 373. Fixed Fixed rates tend to result in greater discounts – “up to 50%” – than than discount off charges. F. 373. As the Unicare representative explained, in discount off charges arrangements, the “hospital sets their own prices,” and managed care organizations “have no control control over . . . what the services are going to cost in any given admission or service.” F. 374. Moving managed care organizations organizations to discount off charges contracts permitted ENH to institute additional price increases by allowing it to unilaterally increase its chargemaster. F. 384-91. These subsequent price EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 267 Initial Decision increases did not necessitate additional negotiation, and in many cases did not even require notification to managed care organizations. F. 386-87. Respondent notes that some managed care care organizations negotiated some relief from subsequent chargemaster increases, but, as Haas-Wilson’s empirical analysis shows, those limits, where they existed, were not effective. F. 469- 469-692.

As part of the merger integration process, ENH consolidated the Highland Park and Evanston chargemasters in 2000. F. 378. In a “fairly simplistic analysis,” ENH examined the chargemasters at the two hospitals and adopted the higher of the Highland Park or Evanston chargemaster rates for each line item. F. 380. As of September 30, 2000, only nine months after the merger, Neaman, ENH’s CEO, reported to ENH’s board of directors that ENH’s “Unified Pricing Structure” for the chargemaster had already resulted in $5 million of annualized economic value. F. 383. This increase is larger than the estimated increase in net revenues from the renegotiated contracts with any single managed care organization. F. 370, 383. Without the merger, chargemaster increases would most likely have been restrained by the possibility of losing managed care customers through selective contracting, steering, and competition. F. 158-69. As a result of the merger, and its newly-enhanced market power, ENH was able to impose anticompetitive chargemaster increases.

In addition to the price increases obtained in the 2000 renegotiations and through the 2000 chargemaster consolidation, ENH subsequently increased its chargemaster rates four times between 2002 and 2003. F. 384. Together, ENH’s four chargemaster chargemaster increases in 2002 and 2003 represented a { } price price increase. F. 391. ENH instituted a price increase of { } on April 15, 2002; { } on October 1, 2002; { } on June 1, 2003; and { 2003; and { } on October 1, 2003. F. 385, 388-90. The April 15, 15, 2002 increase, alone, was projected to have an annual net impact impact { } F. 385. The evidence does not provide a comparable estimate of the net impact on annual net revenue of the last three VOLUME 144 Initial Decision increases, but it clearly would be substantial. The evidence does not not compare these increases to increases at other hospitals, and they they are included to demonstrate that ENH possessed the market power to impose substantial, unilateral, and repeated price increases. increases.

The fact that ENH realized these substantial increased revenues was not widely advertised. In March 2002, Hillebrand advised that for chargemaster increases, “the only notification we make is to Blue Cross” and that “[w]e should not notify anyone beyond those we have a contractual obligation to do so.” F. 387. After ENH raised its chargemaster prices in April 2002, ENH’s executive vice-president for finance wrote to ENH managers that “[f]or a number of reasons we want to be as quiet as possible and there are relatively few people who have seen the scope of the changes.” F. 386. It is clear that these chargemaster changes added significant increased revenue to the merged ENH. The evidence thus establishes that as a result of the merger, ENH was able to use its enhanced market power to implement a continuous and ongoing mechanism to impose significant price increases through a discount off charges fee arrangement. These increases negatively impact self insured patients, as well. Contrary to Respondent’s assertion, these chargemaster revisions were certainly more than a one time, catch up occurrence and appear to be aimed almost exclusively at revenue enhancement.

(c) Managed Care Representatives’ Testimony Confirms Price Increases As the following evidence demonstrates, managed care representatives’ testimony confirms that ENH significantly increased increased its prices post-merger by negotiating contracts with increased discount off charges terms. As previously noted, by increasing the number of discount off charges terms in managed care care contracts, ENH was able to obtain significant additional EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 269 Initial Decision revenue from managed care organizations through subsequent unilateral chargemaster increases.

(i) United Before the merger in 2000, Highland Park and Evanston hospital representatives formulated a strategy for the renegotiation of a contract with United. F. 394. Bain identified the United contract as a “1st Priority” contract with “upside revenue potential” for which the merged entity had “enough leverage to improve terms.” F. 395. Bain advised ENH that United had reimbursed Evanston 45 to 50% less than it paid Highland Park. F. 395. Moreover, Bain informed Evanston that its outdated contract with United had cost the hospital $30 million over the preceding five years. F. 395. The negotiations resulted in { } F. 396. In 2002, United stated that the merger had enabled ENH to “dominat[e] Chicago’s north shore, providing the only hospital locations . . . ranging between Evanston and Highland Park, as well as a significant stretch of territory moving inland” and noting “the strategic importance of ENH’s geographic exclusivity.” F. 398. In August 2002, United requested a renegotiation of United’s contract with ENH because, since the 2000 contract, ENH had been an “outlier” hospital with “much higher than the average reimbursement.” F. 399. United was concerned in part because the 2000 contract relied primarily on a discount off charges payment methodology, resulting in higher and higher reimbursements from United, which witnessed “alarmin[g] escalating costs in [ENH’s] billed charges” that were “outside of the norms for the market.” F. 400. United was also concerned that in 2002, “from quarter to quarter, the [chargemaster] increases were still occurring. It was not a one-time event.” F. 402. { } F. 403. Having had no success in lowering ENH’s prices, United pursued the more modest goal of asking ENH to stop increasing prices so much. F. 404. { } F. 404. The new contract between ENH VOLUME 144 Initial Decision ENH and United was signed on April 14, 2004, with an effective date of June 1, 2004. F. 405. { } F. 406. { } F. 407. Even today, today, with Lake Forest, Rush North Shore, St. Francis, and other neighboring hospitals in its network, United believes it cannot satisfy its customers without ENH. F. 408. (ii) PHCS Prior to the merger, PHCS obtained competitive pricing from Evanston and Highland Park because PHCS “could choose between the two and work them against each other.” F. 409. On December 1, 1999, ENH notified PHCS of the impending merger and sought to assign Highland Park’s rates. F. 410. In response to that letter, PHCS sought to renegotiate the rates. F. 410. Bain advised ENH that it had “significant leverage in negotiations with PHCS as they have strong North Shore presence and need us in their network.” F. 411. Bain indicated that Highland Park’s pre-merger contract terms with PHCS were significantly more favorable than Evanston’s contract terms. F. 411. ENH justified the request for an increase by indicating that it was one system which “controlled the marketplace,” according to one managed care representative. F. 412. The “best scenario” for PHCS customers, strictly looking at dollars, was to eliminate ENH and redirect enrollees to the surrounding hospitals, such as Lake Forest, Advocate Lutheran General, and St. Francis. F. 413. PHCS believed, however, that customers did not want to “buy the [PHCS] network if they did not have [ENH in] it.” F. 414. Thus, PHCS agreed to the { } F. 418.

(iii) One Health (Great West) In December 1999, ENH contacted One Health (formerly Great Great West) to request the renegotiation of its hospital contract. F. 420-21. Bain noted the “substantial difference” between One Health’s Highland Park and Evanston contract terms. F. 422. Bain EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 271 Initial Decision advised ENH to “[a]chieve [Highland Park] terms or better” in its negotiations with One Health. F. 422.

Having last renegotiated the Highland Park and Evanston contracts in 1996 and in 1995, respectively, One Health “agreed that it had been several years since the contracts had been renegotiated and that it was appropriate to [] increase some of the rates.” F. 423. One Health was willing to give a price increase { } F. 423. In the first half of 2000, ENH and One Health did not reach an agreement on the renegotiation of the PPO and HMO contracts. F. 424. One Health accepted ENH’s notice of termination. F. 424. One Health’s contract with ENH subsequently terminated on August 31, 2000. F. 425. One Health made provisions for women who were in the third trimester of pregnancy at the time of the contract termination. F. 426. While One Health was able to negotiate a continuation of benefits for those expecting mothers, ENH charged One Health rates that were higher than contract rates that had been in place under the 1996 premerger One Health contract. F. 426. One Health customers complained about not having access to ENH, although One Health pointed to Lake Forest, Northwest Community, Advocate Lutheran General, Rush North Shore, and St. Francis as substitutes. F. 427. In the months following the termination of the ENH contract, One Health’s monthly membership reports began to reflect a “loss of membership within [the] network.” F. 428. In addition, before discussions between ENH and One Health resumed in early October 2000, One Health received a written notice of termination, effective December 31, 2000, from Lake Forest and its medical group. F. 429. Since Lake Forest was the primary alternative to Highland Park, it would have been “very problematic” for One Health to have lost Lake Forest Hospital from the network at the same time that One Health had no contract with ENH. F. 429. VOLUME 144 Initial Decision One Health returned to ENH prepared to accede “essentially regardless of what the ultimate price was.” F. 430. One Health accepted a new agreement with an effective date of January 1, 2001, four months after the prior contract lapsed. F. 431. { } F. 432. { } F. 433.

(iv) Aetna Aetna “would have walked away” from Evanston if faced with a significant price increase before the merger. F. 434. “[T]here probably would have been a walk-away point with the two independently. But with the two together, that was a different conversation.” F. 434. With the merger of “three extremely important hospitals negotiating together in a very important geography,” Aetna was “extremely concerned.” F. 435. Bain identified Highland Park’s rates for Aetna’s PPO and POS products as higher than Evanston’s rates for those products. F. 436. Evanston’s contract with Aetna was nearly four years old in November 1999, so Bain recommended renegotiation of the Aetna contract as a priority. F. 436.

Aetna had not renegotiated its contract with Evanston since 1996 and expected ENH to make a proposal to renegotiate. F. 437. Based on the 3% increase per year in medical CPI between 1996 and 1999, Aetna calculated an appropriate increase compounded over three years to be { } F. 437. During the 2000 negotiations, ENH originally sought a discount off charges arrangement for PPO and POS plans. F. 438. Aetna, however, did not agree to that payment methodology. F. 438. ENH and Aetna agreed { } F. 439. { } F. 441. { } F. 443. { } F. 442. { } F. 444. { } F. 445. Aetna believed it “couldn’t walk away” from post-merger ENH because it would have “devastated us,” and “shut down” Aetna’s marketing to local employers. F. 446.

(v) Unicare EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 273 Initial Decision In 2000, Unicare acquired Rush Prudential, another managed care organization. F. 447. Prior to the merger, Rush Prudential had contracted with both Evanston and Highland Park, and Unicare had contracted with just Evanston. F. 447. { } F. 448. With the merger, ENH proposed an unusual “all-or-nothing deal” in which there would be one rate for all three hospitals, regardless of the level of services at each facility.” F. 449. { } F. 450. { } F. 450. Even if Unicare representatives had expected an increase in ENH contract rates after the merger, the rates proposed by ENH in 2000 were above what Unicare considered to be a “reasonable” increase, { } F. 451. { } F. 451.

The result for Unicare { } F. 452. { } F. 453. { } F. 453. According to Unicare, ENH had indicated that it could obtain higher prices because it had “a lot more leverage now that they have three hospitals in their service area” and that ENH had a “stronger presence” in the area, meaning ENH had “basically sewn up the North Shore geography.” F. 455. Unicare would be in a bind without ENH, now a “key provider” in the North Shore. F. 456. ENH’s “contiguous service area” made it “hard, painful, for customers to see [ENH] leave the network.” F. 456.

(vi) Summary The evidence of ENH’s negotiations with managed care organizations clearly demonstrates that the combined ENH had enhanced its market power from the premerger period when Evanston and Highland Park had been negotiating as independent competitors. This increase in market power occurred immediately after and solely due to the merger and not to any other changes in market forces. Moreover, at the time, the price increases were never never ascribed by the parties as being related to improvements in quality of care or any changes in the level of services provided by the ENH hospitals. Rather, ENH’s ability to increase prices stemmed stemmed from its geographic exclusivity in an important region. VOLUME 144 Initial Decision ENH was fully aware of its enhanced market power as a result of the the merger and utilized its newly-formed competitive position to obtain much more favorable contracts with managed care organizations than either Evanston or Highland Park could have negotiated as independent hospitals.

(d) Respondent Highlighted the Managed Care Price Increases as a Merger Accomplishment Internal memoranda indicate that ENH highlighted, even celebrated, the managed care price increases as an achievement directly related to the merger. The contemporaneous documents demonstrate that ENH’s primary merger accomplishment was increased revenues, the majority of which came from managed care organizations. On March 14, 2000, ENH’s COO drafted ENH’s 2001-2003 Strategic Plan. In the draft of the Strategic Plan, ENH’s COO stated:

Through our growth initiatives, we will expand our presence in our marketplace in order to provide leverage to our market position as we negotiate relationships with the purchasers of care. Our goal will be to receive superior pricing for our services and to become indispensable to the purchaser of care as they sell their product in our marketplace. F. 459. This aptly summarizes ENH’s accomplishments. Additional contemporaneous documents highlight the significant significant price increases achieved as a result of the merger. In June June 2000, it was reported that Neaman, ENH’s CEO, “reviewed the the list of merger accomplishments. Important successes have been been accomplished in managed care contracting. There has been a $12 million improvement on the Hospital side and $8 million to physicians’ practices to date.” F. 460. By October 2, 2000, Neaman EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 275 Initial Decision Neaman reported: “[s]ome $24 million of revenue enhancements have been achieved – mostly via managed care renegotiations. (This (This figure does not include some $13 million of additional managed care revenues to participating physicians.).” F. 464. In addition, “[s]ome $12 million of cost improvements have been achieved – mostly from corporate overhead areas.” F. 464. The hospitals’ revenue enhancements from the managed care renegotiations were thus double the revenue enhancements from cost cost improvements. None of these savings were passed on to managed care organizations, or therefore consumers, in the form of of lower prices. See F. 326-755; see also Closing argument, Tr. 6582-83. Nor were any of the initial post-merger price increases obtained by ENH from managed care organizations reduced in subsequent years, with the exception of a { } F. 466. Evanston’s CEO acknowledged that the price increases to managed care organizations were the direct result of the merger. Neaman’s July 3, 2000 “Interdependence” memorandum stated: our success in the merger integration effort is not a product of our “independence,” but of our “interdependence.” Neither Evanston nor Highland Park alone could achieve these results. Our three Hospitals, together with our 1500 physicians as a “fighting unit” appear to have helped provide at least a small advantage for an interim period. F. 462. At a September 27, 2000 meeting, Neaman stated that “the larger market share created by adding Highland Park Hospital has translated to better managed care contracts.” F. 463. Neaman’s October 2, 2000 report reiterated: “[a]s stated previously, none of this could have been achieved by either Evanston or Highland Park alone. The 'fighting unit’ of our three hospitals and 1600 physicians was instrumental in achieving these ends.” F. 465. Respondent’s argument that these statements should not be taken at face value or are taken out of context is unpersuasive. VOLUME 144 Initial Decision ENH thus achieved its goal of “superior pricing” due to its enhanced post-merger market power and competitive position. F. 326-755. ENH, who was in the best position to evaluate the effect of the merger, repeatedly attributed the increased prices to post-merger renegotiations with the managed care organizations. F. 457-68. In addition to the ENH documents, Highland Park representatives testified that all the rates Highland Park Hospital had in place in July 1, 1999, were the best that Highland Park could accomplish at that time without threatening termination. F. 467. Highland Park’s CEO testified that, at the time of the merger, Highland Park would not have been successful in raising its rates because the hospital could not sustain a strategy where it kept losing contracts. F. 468. He did not see an opportunity to raise the rates before the merger. F. 468. The fact that Highland Park executives were concerned about contract terminations premerger is illustrative of the competitive environment that existed before 2000 and stands in contrast to the actions of ENH officials who, given their competitive situation, were not constrained by such prospects in their renegotiations with managed care representatives post-merger. Thus, ENH continued to tout the principal accomplishment of the merger as revenue enhancement, which the evidence indicates resulted from its post-merger market power in managed care negotiations. This market power allowed ENH to maintain significant price increases over a number of years and was achieved as a direct result of the merger. The totality of the evidence thus demonstrates that Evanston and Highland Park merged to eliminate competition from each other, enhance their competitive position in the market, and obtain substantial price increases from managed care organizations. The evidence further demonstrates that as soon as the merger was consummated, Respondent began using its enhanced market power to impose significant price increases on managed care organizations, and ultimately consumers. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 277 Initial Decision (3) Empirical Analysis Establishes That Respondent’s Prices Rose Relative to Other Hospitals In addition to the contemporaneous evidence and managed care testimony, the economic evidence establishes that ENH’s postmerger price increases were attributable to market power. Complaint Counsel’s expert, Haas-Wilson, utilized data from four different sources – managed care organizations; the State of Illinois Department of Public Health (“IDPH”); a Civil Investigative Demand (“CID”) to ENH; and National Economic Research Associates (“NERA”), ENH’s consultant. F. 469. Data from all four sources shows that “for most [managed care plans], there were large post-merger price increases at ENH.” F. 498. The data from the managed care organizations and the State of Illinois contained pricing data for hospitals other than ENH, so only those two sources provide specific data for a comparative analysis of relative price increases. See F. 573-74. The CID and NERA data is compared to the Chicago medical CPI. F. 614, 644. Respondent objects to the use of the Chicago CPI as opposed to a national hospital CPI and objects to the use of this data in a comparative fashion. RRFF ¶ 404. Although not as precise as the relative comparison obtained by Haas-Wilson for the managed care and DPH databases, the CID and NERA data, in combination with the other data, confirms the conclusion that ENH significantly increased prices relative to other hospitals’ price increases. The NERA and CID data is particularly useful because it encompasses many more payors than the managed care and IDPH data. F. 612, 642.

Complaint Counsel acknowledges that “large price increases alone do not mean that the merger gave ENH market power.” CCB CCB at 45; see also Blue Cross & Blue Shield United of Wisconsin v. Wisconsin v. Marshfield Clinic, 65 F.3d 1406, 1411-12 (7th Cir. 1995). Therefore, Haas-Wilson examined whether ENH’s price increases were attributable to changes in the marketplace that would would affect all hospitals equally. F. 477-80. This required VOLUME 144 Initial Decision comparing ENH’s price increases against three control groups of hospitals. F. 481.

The role of the hospital control groups is to control for marketwide factors that might provide alternative (completely benign) explanations for the observed relative price increases, such as changes in cost, regulation, or demand that might be impacting comparison hospitals and the merging hospitals the same way. F. 694-96, 702. Haas-Wilson’s three control groups were: (1) all general acute care hospitals in the Chicago Primary Metropolitan Statistical Area (“PMSA”) (the “Chicago PMSA Hospitals” control group); (2) all general acute care hospitals in the Chicago PMSA, that were not involved with a merger with another hospital between 1996 and 2002 (the “Non-Merging Chicago PMSA Hospitals” control group); and (3) all general acute care hospitals in the Chicago PMSA that were involved in some teaching activity during the study period (the “Chicago PMSA Teaching Hospitals” control group). F. 481. Applying a “difference in differences” technique, Haas-Wilson first calculated the difference in premerger and postmerger prices for ENH and for the control groups, expressed as a percentage, and then compared ENH’s numbers to the control groups’ numbers. F. 477-80.

Respondent argues that Complaint Counsel’s control groups are are overbroad and do not control for idiosyncratic but competitively competitively benign changes to ENH’s prices. RB at 39. Indeed, the the Chicago PMSA Hospitals control group includes one hundred hospitals and the Chicago PMSA Teaching Hospitals control group group includes fifty hospitals. F. 486. However, Haas-Wilson rejected the concept of picking only hospitals that “looked like” Evanston to use as her control group because this would have required making arbitrary decisions on which neither theory nor previous empirical work provided guidance. F. 487. Any attempt to to match hospitals with ENH to form a control group that “looked like” ENH would have to account for the fact that Evanston and EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 279 Initial Decision Highland Park had different characteristics pre-merger. F. 488. Upon Upon review, Haas-Wilson’s methodology in selecting her control groups is considered more reliable and appropriate than that of ENH’s expert, Noether. See also infra Section III.C.2.a.3. Haas-Wilson found that, with the exception of Blue Cross Blue Shield, ENH’s price increases across all managed care organizations were higher than the price increases at the control group hospitals. F. 473. This means that changes in costs, regulations, or demand – market conditions that would be expected to cause similar price increases across all hospitals – could not explain the higher prices at ENH. F. 698-713.

ENH’s argument that its Blue Cross Blue Shield rates are inconsistent with market power (RB at 52-53) is unpersuasive. Blue Cross Blue Shield is the largest managed care organization in Chicago, and accounts for approximately twenty percent of ENH’s business. F. 561. Thus, Blue Cross Blue Shield has the power to limit ENH’s price increases. That ENH has not, to date, imposed price increases on Blue Cross Blue Shield does not undermine the conclusion that ENH gained market power through the merger. As Bain acknowledged, ENH’s bargaining position with each managed care organization was different and ENH’s “leverage” in contract negotiations with Blue Cross Blue Shield was “less than with most payors.” F. 562. There is no dispute that Blue Cross Blue Shield had a very strong bargaining position against ENH. Thus, Blue Cross Blue Shield has the power to limit ENH’s price increases. That ENH has not, to date, imposed price increases on Blue Cross Blue Shield does not undermine the conclusion that ENH gained market power through the merger.

Haas-Wilson observed that changes in ENH’s patient mix, customer mix, and teaching intensity varied from the control group group hospitals. In order to assess the impact of these changes, Haas-Wilson conducted a multiple regression analysis that compared compared ENH’s percentage price changes against the control groups’ price increases while at the same time accounting for the VOLUME 144 Initial Decision three variables. F. 727. The regression analysis showed that ENH’s ENH’s percentage price increases were higher than the control groups’ price increases even after accounting for changes in patient patient mix, customer mix, and teaching intensity. F. 583. The only only exception to Haas-Wilson’s pricing analysis results was Blue Cross Blue Shield – ENH increased its prices, but the percentage increase was similar or the same as the increases at the control group group hospitals. F. 571-72. This means that changes in customer mix, patient mix, and teaching intensity also do not explain ENH’s ENH’s price increases. F. 583.

The pricing analyses conducted by both Complaint Counsel’s expert and Respondent’s expert show significantly higher percentage price increases by ENH than by other hospitals. Haas-Wilson found that ENH’s price increases to the following managed care organizations exceeded the price increases of the control groups by the amounts shown: { } F. 520-22, 535-37, 558-60. Haas-Wilson’s results are presented as ranges because the specific price increase results depend on the measurement and control group against which prices are compared. F. 481. Haas-Wilson’s results are statistically significant at the 1% level, the “highest level of significance.” F. 489, 502, 524, 540, 584, 591-93, 599-601, 608-10. The IDPH data includes all managed care plans in Illinois, thereby allowing Haas-Wilson to compute ENH’s price increases across all managed care organizations. F. 573. Across all managed care plans, ENH’s price increases exceeded the control groups by 11 to 18%, i.e., if other hospitals raised prices by 10%, ENH raised prices by 21 to 28%. F. 591-93, 599-601, 607-10. ENH’s price increase would be even higher if Blue Cross Blue Shield was excluded because Blue Cross Blue Shield was the only managed care organization that did not incur a price increase from ENH that was higher than the control group hospitals’ price increases. F. 571- 72.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 281 Initial Decision Respondent’s economic expert, Dr. Jonathan B. Baker, agreed that ENH’s post-merger price increases were higher than other hospitals. F. 688-89. Even Baker calculated ENH’s post-merger price increase as 9 to 10% higher than his control group hospitals. F. 689-90. Moreover, Baker’s figure represents data from only four managed care organizations: United, Aetna, Humana, and Blue Cross Blue Shield. F. 675-79. Including Blue Cross Blue Shield, the largest managed care organization and with whom ENH has little leverage, weighs down ENH’s number. F. 561-62. Not included in Baker’s calculations are data from One Health or any of the other health plans included in the Illinois Department of Public Health data. See F. 675-79, 685. Only Haas-Wilson presented aggregated pricing analysis results that covered all managed care plans. (4) Explanations of Price Increases Other than Market Power Are Ruled Out Haas-Wilson examined ten possible explanations for ENH’s higher prices, including the two principal explanations advanced by Respondent, learning about demand and improved quality of care. See infra III.C.2.a and III.C.2.b. Haas-Wilson did not test every conceivable reason for the price increase, just those that were reasonable and supported by sound economic theory. F. 693-95, 702. Utilizing multiple regression analyses, Haas-Wilson ruled out six alternative explanations by the pricing analysis: increases in cost, changes in regulation, increases in demand, changes in patient mix, changes in customer mix, and changes in teaching intensity. F. 698- 755. Also excluded was the possibility that ENH offset the higher inpatient prices with lower outpatient prices because the data showed that ENH’s outpatient prices did not decrease relative to the control groups. F. 703, 717-26.

Two other possible explanations, learning about demand and quality of care improvements, are also ruled out. F. 714-16, 756-837, 756-837, 853-868. As discussed in Section III.C.2.a, the learning about demand theory is flawed; is inconsistent with Respondent’s VOLUME 144 Initial Decision contemporaneous actions; and Respondent’s empirical analysis supporting the theory is unreliable. The evidence also does not demonstrate that overall quality of care at Highland Park improved improved relative to other hospitals, as discussed at length in Section III.C.2.b. ENH’s expert conceded that there is no need to adjust the higher prices to account for quality of care if the quality at at ENH did not increase relative to control group hospitals. F. 838. 838. Thus, the evidence demonstrates that learning about demand and quality of care improvements do not justify ENH’s price increases to managed care organizations. An analysis of the empirical data establishes that enhanced market power is the only plausible, economically sound, and factually well-founded explanation for ENH’s post-merger relative price increases. F. 469-755. This conclusion is corroborated by the business documents and testimony of managed care organizations and ENH employees. F. 327-468. There is also no dispute that ENH’s price increases were higher than other comparison hospitals’ price increases. F. 473-74, 690. Respondent’s expert, Noether, acknowledged that a hospital merger could lead to market power at the same time the hospital learns more about demand for its services. F. 757. Respondent’s expert, Baker, similarly conceded that the pattern of price increases at United, Aetna, and Humana was consistent with ENH obtaining market power through the merger. F. 684. Thus, through the elimination of Highland Park as a competitor, which enhanced ENH’s market power, the merger is likely to result in the restraints condemned under Section 7 and poses an appreciable danger of anticompetitive consequences. 2. Procompetitive Justifications The analysis of market concentration establishes a “highly concentrated” market and constitutes presumptive evidence of the probable anticompetitive effects of the merger. In addition, Complaint Counsel established, through direct evidence, that ENH EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 283 Initial Decision ENH exercised its enhanced market power to raise prices significantly to managed care organizations. As such, Complaint Counsel has established a prima facie case of Clayton 7 liability. The burden thus shifts to Respondent to rebut the presumption arising from the market concentration statistics and evidence of direct anticompetitive effects. See Baker Hughes, 908 F.2d at 982. “The more compelling the prima facie case, the more evidence the the defendant must present to rebut it successfully.” Baker Hughes, Hughes, 908 F.2d at 991.

A respondent may present evidence of a number of factors that are relevant in determining whether a transaction is likely to substantially lessen competition. In this case, Respondent offers two main arguments to rebut Complaint Counsel’s prima facie showing. First, it contends that the post-merger price increases are not due to market power, but rather were the result of ENH, coincident with the merger, “learning about demand” for its services. RB at 40-54. Second, ENH argues that the price increases can be accounted for by post-merger “quality of care improvements” to Highland Park. RB at 67-99. In addition, Respondent offers further arguments regarding the merging hospitals’ nonprofit status, the lack of barriers to entry, and the weakness of the acquired hospital. RB at 58-67. As set forth below, Respondent’s arguments are unpersuasive. Respondent fails, therefore, to rebut Complaint Counsel’s prima facie case. a. Learning About Demand Respondent asserts that as a result of its premerger due diligence diligence and review of information about Highland Park’s contract contract rates with managed care organizations, Evanston learned that some of its contracts were outdated and that its rates were below below market. RB at 40. Respondent further contends that it used this new information to negotiate post-merger price increases that brought its prices “in-line with those charged by other comparison hospitals.” RB at 40. Complaint Counsel contends that Evanston did did not underprice itself before the merger; that Evanston had higher VOLUME 144 Initial Decision higher ultimate prices; and that the price level comparison conducted by Respondent’s expert, including the choice of control groups, is flawed. CCB at 60-65.

A review of the record refutes Respondent’s assertions and demonstrates that the price increases ENH was able to command after the merger were not a consequence of obtaining new information, but instead were the result of newly created market conditions which affected the demand for ENH’s services -- the elimination of Highland Park as a price constraining competitor. See supra Section III.C.1.b. As discussed below, the evidence demonstrates that there are flaws in the learning about demand theory as applied in this case; that Respondent’s contemporaneous actions are not consistent with the learning about demand theory; and that the empirical analysis conducted by Respondent’s expert in support of the theory is unreliable.

(1) Unsupported Foundations for the Theory Experts from both sides agree that Respondent’s prices rose after the merger. See F. 473-74, 690. Respondent contends, however, that prior to the merger, Evanston was priced below a competitive level and that, during due diligence work connected with the merger, Evanston learned that, for some contracts, it had the same or lower contract rates than Highland Park. RB at 40. From this new information regarding Highland Park’s rates, Evanston asserts that it learned that it was underpricing itself. RB at 40. Therefore, Respondent argues, the increase in post-merger prices merely reflects ENH’s attempt to “catch up” with competitive pricing levels and obtain fair market value for its services. RB at 40. As the evidence demonstrates, however, there are a significant number of problems with this theory.

First, Respondent does not contend that it merely raised Evanston’s prices so that they were comparable to Highland Park’s Park’s rates. Rather, Respondent asserts that, as a teaching or EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 285 Initial Decision “academic” hospital, Evanston was entitled to even higher rates than than Highland Park. RB at 48. In this regard, Haas-Wilson testified testified that the “empirical literature . . . suggests that costs and therefore prices 'might’ be different at hospitals that are engaged in in 'teaching activity’ versus those that are not.” F. 758. In fact, Noether’s empirical analysis shows that her control group of “academic hospitals” are priced higher than her control group of “community hospitals.” F. 818-19.

Though the evidence indicates that managed care organizations pay more for “advanced teaching hospitals” or “academic teaching hospitals” (presumably, those that offer inter alia, quaternary care), the evidence does not show that Evanston qualified for such treatment. Representatives from One Health, PHCS, and United testified that they do not view any of the ENH hospitals as “advanced teaching hospitals” or as “academic teaching hospitals.” F. 772-83. Evanston, for example, does not offer quaternary services such as major organ transplants or a severe burn unit. F. 203. Although Evanston is a “teaching hospital” (Evanston Hospital/ENH has been named by one publication as a top 15 teaching hospital and a top 100 hospital in the country, F. 786), it is not considered a toptier, major academic center like the University of Chicago or Rush- Presbyterian-St. Luke’s, against whom its rates were compared by ENH’s expert. F. 775, 779, 782.

Therefore, the empirical evidence does not support Respondent’s Respondent’s assumption that Evanston’s fair market value at the time of the merger was either higher than Highland Park’s, or comparable to those hospitals in Noether’s academic control group. group. Learning about Highland Park’s non-teaching hospital rates rates at the time of the merger told Evanston nothing about other hospitals’ rates or prices, and most certainly did not provide any information about rates or charges at teaching hospitals or advanced advanced teaching hospitals. Respondent’s argument implies that certain teaching hospitals, due to their enhanced level of services, form their own product market because the demand for their services VOLUME 144 Initial Decision services is higher, an argument that was rejected in Long Island Jewish Medical Center. 983 F. Supp. at 138-40 (finding government’s characterization of an anchor hospital as a relevant product market unnecessarily restrictive). Next, even if Evanston deserved higher prices based on its teaching status, Highland Park would not. After the merger, only one department at Highland Park had residents, and that department only had 6 residents at the time of trial, below the Medicare Payment Advisory Commission (“MedPAC”) definition of .25 residents per bed. F. 809. Being owned by a teaching hospital did not transform Highland Park into a teaching hospital. F. 992. However, managed care organizations who wanted any of the three ENH hospitals in their hospital networks had to contract with all three for the same higher rates. F. 355-66. Therefore, even if the evidence demonstrates that Evanston deserved higher prices because of its teaching status, this does not provide any justification for charging the same higher rates for Highland Park, a non-teaching community hospital. Thus, the learning about demand theory does not explain or justify price increases that ENH instituted at Highland Park. Finally, in an effort to explain its post-merger price increases, Respondent merges its learning about demand argument with its contention that some of its contracts were outdated. RB at 43-44. Indeed, a number of managed care representatives testified that their contracts with Evanston were, in fact, outdated and that Evanston was due for an increase consistent with medical CPI. F. 437. However, those managed care organizations also testified that the price increases obtained by ENH well exceeded their expectations of a reasonable increase. F. 392-456. Evanston presumably was, or should have been, fully aware that some of its contracts were outdated and did not need the Highland Park merger to learn of this fact. Thus, any argument regarding ENH’s outdated contracts does not support Respondent’s learning about demand theory and is irrelevant to the analysis of the issue. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 287 Initial Decision (2) Contemporaneous Actions In addition to the practical problems attendant with the learning about demand justification, the theory is inconsistent with Respondent’s contemporaneous actions. Respondent appears to lay much of the blame for its allegedly under-market contract prices on its lead negotiator, Jack Sirabian, who claimed at trial that his objective in negotiating managed care contracts was to be in every managed care network and that he sought to nurture relationships with managed care organizations, rather than to get the best possible deal for Evanston. RB at 41. However, after learning about Highland Park’s allegedly higher rates with the merger, ENH nevertheless retained and rewarded Sirabian and his supervisor Hillebrand, who had general oversight for managed care contracting, with substantial post-merger bonuses. F. 761-70. It seems counter-intuitive that a firm would retain, let alone reward, an individual who was thought to be principally responsible for below market contracts, one of which Bain described as having cost ENH approximately $30 million over the past five years. F. 395; RB at 42. Such conduct is particularly peculiar in light of ENH’s decision, post-merger, not to retain Theresa Chan, who had negotiated what Respondent now claims were superior contracts with managed care organizations on behalf of Highland Park. F. 771. It also contradicts the trial testimony of ENH’s COO, who testified that ENH’s negotiating stance was equally “aggressive” before and after the merger. F. 767. Although Bain advised ENH that it “should recognize its position and not be afraid to ask to be paid fair market value” for its services, F. 764, Respondent was not able to point to any contemporaneous documents which reflect that ENH’s learning about Highland Park’s rates taught ENH about other hospitals’ pricing or that its “fair market value” would be comparable to advanced teaching hospitals rather than community hospitals. VOLUME 144 Initial Decision (3) Empirical Analysis (a) Highland Park’s Prices Compared to Evanston’s Prices Respondent has not demonstrated that it did, in fact, learn that it was underpricing itself as compared to Highland Park. F. 784-97. Sirabian testified that in approximately one third of the thirty-five or forty managed care contracts, Highland Park had higher contract rates than Evanston. F. 787. However, rates are just one factor that goes into determining ultimate prices. There are multiple factors in hospital contracts that determine the actual price or the reimbursement per case. F. 789. In addition to per diem rates, contracts also include stop loss provisions, which specify at what point the per diem no longer applies and instead the hospital gets reimbursed on a different basis specified in the contract. F. 790. The contract itself also shows nothing about the hospital’s chargemaster. F. 791. Thus, if two hospitals have contracts that specify a ten percent discount off charges, without knowing the respective chargemasters, knowing the discount off charges rates does not show which hospital had higher ultimate prices. F. 791. As Chan identified at the time, the evidence demonstrates that Evanston’s chargemaster was higher than Highland Park’s chargemaster, premerger. F. 793. Based on Noether’s calculations of of actual price levels in the premerger period, the prices at Evanston Evanston were higher than the prices at Highland Park. F. 794. An analysis by Baker also showed that Evanston’s premerger prices were higher than Highland Park’s prices for three out of the four managed care organizations examined. F. 797. Therefore, although although Highland Park had higher rates on some contracts, factoring in the different chargemasters and services offered, Highland Park’s premerger prices to the four managed care organizations examined by Noether were actually below Evanston’s Evanston’s prices. F. 787-97. Thus, because Evanston’s ultimate EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 289 Initial Decision prices were actually higher than Highland Park’s ultimate prices, ENH could not have learned about demand from this comparison. (b) Noether’s Control Groups Were Flawed Finally, the empirical studies performed by Noether are not economically sound and do not confirm Respondent’s proposition that ENH’s price increases reflect its learning about demand. To evaluate Respondent’s learning about demand theory, Noether compared ENH’s premerger and post-merger prices to those of two control groups of hospitals. Noether testified that she developed her list of eighteen hospitals for her control groups after she “reviewed the evidence from a variety of sources in the record and developed a list based on [her] analysis of the information,” including hospitals which Noether testified were “in some way competitors to Evanston and/or Highland Park.” F. 802.

Noether then divided these eighteen hospitals into two control groups – “academic hospitals” and “community hospitals” – based on breadth of services, teaching intensity, and size. F. 808. Noether decided that ENH should be compared to the “academic hospitals” group, which she defined as including: Northwestern Memorial, Rush-Presbyterian-St. Luke’s, Advocate Lutheran General, Advocate Northside, University of Chicago and Loyola. F. 805. The remaining twelve hospitals became Noether’s community hospital control group. F. 806.

Noether’s “academic” control group, however, is not reliably defined as it primarily utilizes subjective rating factors. Specifically, there is no official government designation defining what criteria are used to establish hospitals as “community hospitals” or “academic hospitals.” F. 807. Without sufficient explanation, Noether established her academic control group as only hospitals with 370 or more Diagnosis Related Groups (“DRGs”), more than .25 residents per bed, and more than 300 staffed beds. F. 808. VOLUME 144 Initial Decision Noether’s teaching intensity classification is consistent with the MedPAC definition which defines a “major teaching hospital” as a hospital with at least .25 residents per bed. F. 809. However, MedPAC does not evaluate diagnosis related groups. For example, the number of DRGs can vary depending on the time period used, and can even vary depending on whether a fiscal or calendar year is used. F. 812. There is no basis in the health care literature to require a hospital to be above a certain number of DRGs in order to be considered an “academic hospital.” F. 814. Similarly, the MedPAC criteria defining a major teaching hospital do not rely on size as an evaluation factor. F. 817. The evidence does not justify the arbitrary cutoff number chosen by Noether for size. F. 817, 829. The record thus casts doubt as to whether Noether utilized objective standards to construct her “academic hospital” control group and whether the standards she utilized are consistent with established industry criteria.

The six “academic” hospitals selected by Noether for her “academic” comparison group are larger than ENH, some of them with significantly more beds. F. 817, 829. In addition, the four quaternary hospitals in her academic control group – Loyola, Northwestern Memorial, University of Chicago, and Rush- Presbyterian-St. Luke’s – handle significantly more complex cases than ENH and perform sophisticated quaternary services, such as severe burn cases or liver and kidney transplants, which are not treated at Evanston. F. 824-25. Notably, four of the six hospitals included in Noether’s “academic” control group are among the most expensive hospitals in Chicago. F. 818. As previously noted, the evidence does not support Respondent’s contention that the ENH should be priced at the level of these top-tier major teaching hospitals. Noether’s academic control group excluded less expensive hospitals even though many of those excluded hospitals can handle most of the patients Evanston treated and treat more complex cases than ENH. F. 819.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 291 Initial Decision Given the above contradictions, it is difficult to evaluate Noether’s conclusions against either objective research standards or the facts of the case. This is especially true when one considers that, of the six hospitals placed in the academic control group to which Noether compared ENH’s prices, only one such hospital was included in her proposed geographic market. F. 805; RB at 23. Moreover, as discussed earlier, Noether only analyzed one data source which included usable data from only four managed care organizations. See supra Section III.C.1.b. Even if Respondent’s learning about demand theory was valid and countered the direct evidence of anticompetitive effects (price increases), the theory is not relevant to the structural evidence of market concentration. Accordingly, the flaws noted in Noether’s methodology and data, along with managed care organizations’ testimony and contemporaneous evidence, demonstrate that Respondent’s learning about demand theory cannot explain the post-merger price increases at ENH.

b. Quality of Care Respondent’s second main argument in rebuttal to Complaint Counsel’s prima facie case is that the quality improvements at Highland Park justify ENH’s increased prices and outweigh any anticompetitive effects of the merger. RB at 69-71. This argument raises the issue of whether quality of care is relevant to the competitive effects analysis, and if so, whether it should be considered a procompetitive justification. Respondent contends that quality of care improved at Highland Highland Park as a result of the merger; that Respondent’s expert as as well as independent assessments affirm improvements in quality quality of care at both Evanston and Highland Park post-merger; and, that no fact witness called by Complaint Counsel countered any any showing of quality improvement at Highland Park. RB at 67- 67-107. Respondent’s argument is not cast as an “efficiency” defense, but rather as an assertion that quality of care improvements improvements are procompetitive justifications that should be VOLUME 144 Initial Decision considered in conjunction with the competitive effects analysis. RB RB at 68; Closing argument, Tr. 6478-79. Complaint Counsel contends that Respondent failed to demonstrate that: quality of care improved patient outcomes and satisfaction; that the quality changes were merger specific; and that any such benefits outweigh the anticompetitive harm. CCB at 11-17. Given the evidence of market power, Complaint Counsel asserts that any doubts must be resolved against the validity of the quality of care defense. CCB at 17-18. Complaint Counsel further states that any merger specific efficiencies that have been verified should be given due weight, but asserts that Respondent’s claimed improvements cannot be sufficiently proved or quantified. CCB at 12.

(1) Legal Framework The precise role of quality of care in the antitrust context has yet to be determined. “[B]ecause contemporary antitrust law does not create many obvious placeholders for nonprice concerns, quality may be litigated under alternative guises.” Peter Hammer & William Sage, Antitrust, Health Care Quality, and the Courts, 102 Colum. L. Rev. 545, 563 (April 2002). The economic testimony in this case appears to view quality as part of the cost/price continuum. F. 838- 39. The Eighth-Circuit has suggested that quality of care may be relevant to the competitive effects analysis. Tenet Health Care, 186 F.3d at 1054.

The district court in Rockford Memorial rejected a quality of care argument as irrelevant to the competitive effects analysis, stating:

Undoubtedly, the improvement in services would have a positive effect for consumers of healthcare in in the relevant market and economic benefits for the the area as a whole. Unfortunately, the creation of a EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 293 Initial Decision a tertiary referral center, while a laudatory goal, is not relevant for our purposes today. The court’s exclusive role is to evaluate the merger’s effect on competition for the relevant market and no more. United States v. Rockford Memorial Corp., 717 F. Supp. 1251, 1288 (N.D. Ill. 1989), aff’d, 898 F.2d 1278 (7th Cir. 1990). In Rockford Memorial, the district court found “the defendants’ intention to create a state-of-the-art tertiary referral center and all its corresponding benefits in quality and community development as irrelevant for the present § 7 inquiry.” Id. at 1289. On appeal, the Seventh Circuit was unpersuaded by the merging parties’ defenses, stating: “[t]he government showed large market shares in a plausibly defined market in an industry more prone than many to collusion. The defendants responded with conjectures about the motives of nonprofits, and other will o’ the wisps, that the district judge was free to reject, and did.” Rockford Memorial, 898 F.2d at 1286. Respondent, sub judice, argues that the district court’s holding in in Rockford Memorial is inapposite because it was limited to the “‘present § 7 inquiry’” and because the Seventh Circuit did not rely rely on the district court’s remarks on quality of care. RB at 71 n.49 n.49 (quoting Rockford Memorial, 898 F.2d at 1289). Respondent contends that enforcement officials at the FTC and DOJ have publicly agreed that quality, innovation, and similar factors are an important part of analyzing the competitive effects of a transaction; transaction; that in bringing recent enforcement actions, governmental antitrust agencies have asserted that quality and innovation are relevant in merger analysis; and that in more recent joint venture and non-merger cases, the Commission and courts have have found that improvements in quality and innovation are also relevant. RB at 68-71. Moreover, as Respondent correctly observes, observes, economists on both sides agree that quality improvements improvements should be taken into account in evaluating whether the merger, on balance, had a positive or negative impact on VOLUME 144 Initial Decision competition. RB at 71, 838-39. Complaint Counsel acknowledges quality as a legitimate defense, citing the Merger Guidelines. CCB at CCB at 12; CCRB at 38.

The Merger Guidelines recognize that “mergers have the potential to generate significant efficiencies by permitting a better utilization of existing assets, enabling the combined firm to achieve lower costs in producing a given quantity and quality than either firm could have achieved without the proposed transaction” and that efficiencies “can enhance the merged firm’s ability and incentive to compete, which may result in lower prices, improved quality, enhanced service, or new products.” Merger Guidelines § 4; H.J. Heinz, 246 F.3d at 720. The Merger Guidelines indicate that the “[a]gency considers whether cognizable efficiencies likely would be sufficient to reverse the merger’s potential to harm consumers in the relevant market.” Merger Guidelines § 4. Thus, the Merger Guidelines recognize quality, at least in the guise of an efficiency, as a relevant antitrust consideration.

The D.C. Circuit has acknowledged that “although the Supreme Court has not sanctioned the use of the efficiencies defense in a section 7 case, the trend among lower courts is to recognize the defense.” H.J. Heinz, 246 F.3d at 720. As noted, Respondent does not argue that economic efficiencies in the form of cost savings were passed on to consumers. Closing argument, Tr. 6584-85. In fact, the record is clear that any cost savings realized by the merger were not passed on to consumers in the form of lower prices. F. 326-755. As with many components of this case, the law with respect to quality of care is not well-settled. Given the difficulty of proof inherent in the analysis of quality of care arguments and the confusion which can result from the attempt to quantify quality of care improvements, the courts in non-merger contexts treat the issue issue with skepticism. See, e.g., FTC v. Indiana Federation of Dentists, 476 U.S. 447, 464 (1986) (“even if concern for the quality EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 295 Initial Decision quality of patient care could under some circumstances serve as a justification for a restraint” of trade, the evidence did not support a a finding under the facts).

If quality of care is relevant to a hospital merger action under Section 7, it is not clear whether it should be considered a procompetitive justification, an affirmative defense, or an efficiency. Antitrust, to date, has not recognized a single approach to a quality of care defense. Respondent, however, argues that quality of care should be analyzed as a procompetitive justification under the competitive effects analysis, RB 71-72, and the Court will treat it as such. Assuming arguendo, that quality of care is relevant to the analysis of the competitive effects of a merger, the facts nevertheless do not support Respondent’s theory. As discussed supra, the merger increased concentration in the market for healthcare services in the relevant market (F. 309-25); enhanced ENH’s market power (F. 309- 755); and resulted in relative price increases to managed care organizations (F. 392-692) and ultimately consumers (F. 187-90). Considering the substantial evidence of anticompetitive effects, Respondent’s few merger specific improvements to Highland Park do not constitute a sufficiently procompetitive justification that outweighs the harm to competition as a result of the merger. (2) Factual Analysis Respondent compares post-merger Highland Park in 2005 with with premerger Highland Park in 1999 to argue that Highland Park’s Park’s quality of care has substantially improved as a result of the merger. Respondent is correct that significant improvements have been made to Highland Park and that those improvements can be verified. However, there are a number of problems with Respondent’s efforts to demonstrate a procompetitive justification. justification. First, there is no quantifiable evidence that the improvements at Highland Park enhanced competition and thus benefitted consumer welfare. Indeed, the evidence does not demonstrate that the post-merger price increases to managed care VOLUME 144 Initial Decision organizations were related to the improvements at Highland Park. F. F. 838-52. Second, there is insufficient evidence of overall improvement in quality of care relative to other hospitals. That is, improvements were made at Highland Park, but it is not clear that those improvements affected quality, or, if they did, that they improved quality in relation to hospitals generally. Therefore, there there is no way to determine whether the improvements at Highland Highland Park were due specifically to the merger or to nationwide nationwide efforts to improve patient care. The improvements only only occurred, for the most part, at one of the three ENH hospitals, hospitals, although the price increases were obtained for all three hospitals. Third, although there were many improvements in Highland Park’s physical plant and equipment, processes, and hospital organization, only two of these improvements are found to to be merger specific – the EPIC integrated medical electronic record system and the academic affiliation and clinical integration. integration. Although Highland Park, in 2005, has improved since 1999, the evidence does not show that it has improved more than it it would have but for the merger. As explained below, as a factual matter, these merger specific improvements are not sufficient to overcome the significant anticompetitive effects associated with the the merger and did not justify the post-merger price increases to managed care organizations.

(a) Improvements Can Be Verified Respondent cannot rely on “mere speculation and promises,” and and its proof should be subject to “rigorous” analysis, given the high high HHI numbers associated with the merger. H.J. Heinz, 246 F.3d F.3d at 721. ENH must “substantiate” the purported improvements improvements and verify their magnitude. Merger Guidelines § 4; Staples, 970 F. Supp. at 1089 (efficiency claims fail if “unreliable” “unreliable” and “unverified”). However, because this is a consummated merger case, Respondent has provided significant evidence of actual improvements to Highland Park. Respondent’s arguments cannot therefore be dismissed as “mere speculation and EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 297 Initial Decision promises.” Indeed, the evidence demonstrates that ENH has, in fact, fact, invested $120 million into Highland Park and has made many many improvements to Highland Park that can be verified. See F. 876-993. The mere fact of financial investments and physical improvements to one of the merging entities, however, does not, of of itself, provide a legally sufficient procompetitive justification for for the merger.

(b) Price Increases to Managed Care Were Not Related to Improvements at Highland Park The record establishes that at the time it increased its prices, ENH did not justify its price increases to managed care based on improvements being made at Highland Park. F. 840. Managed care representatives testified that during contract negotiations, the topic of quality improvements simply never came up. F. 844-47. ENH’s COO admitted that he did not tell managed care representatives that the higher prices were justified by quality changes to Highland Park. F. 842. Similarly, ENH’s CEO conceded that he never saw any documents correlating the higher prices with the quality changes at Highland Park. F. 843.

Even after implementing these changes, ENH never advertised them to managed care organizations. F. 841-47. If quality improvements justified the price increases to managed care, logic would dictate that ENH would have gone out of its way to advertise, advertise, or at least inform, managed care organizations of such improvements. Respondent argues that a press release which mentioned planned clinical service improvements put managed care, care, and the public, on notice of the improvements. RRB at 77-78. 77-78. However, the solitary general press release does not alter the the Court’s analysis. See F. 848. The lack of contemporaneous documentation or managed care testimony supporting Respondent’s Respondent’s quality of care argument thus undermines its litigation litigation position. Rather, the totality of the evidence strongly suggests that Respondent’s quality of care argument is a post hoc VOLUME 144 Initial Decision attempt to justify its post-merger price increases found to exist even even by its own expert.

A review of the record shows that there is no substantial evidence that managed care’s demand for ENH’s services changed as a result of its quality improvements. That is, the improvements at Highland Park did not translate into an increase in demand. Highland Park was already a highly desirable hospital in terms of quality, and remained so after the merger. Highland Park management and outside observers believed that the quality of care at Highland Park was “very good, if not excellent” at the time of the merger. F. 850. Highland Park was also described as an “excellent community hospital” that “delivers basic services at a very high level.” F. 851. Evanston and Highland Park “were both very good hospitals.” F. 852. Nevertheless, the managed care representatives testified that the value of ENH to their networks was principally due to the hospitals’ geography, not quality. F. 226-42. This is not a case where the merger created a hospital that provided better medical care than the hospitals could have provided separately. See Tenet Health Care, 186 F.3d at 1054.

The record reveals that it would have been hard for ENH to justify the price increases to managed care because of quality improvements due, simply, to the timing of the improvements. ENH negotiated its price increases before any quality improvements were ever implemented. Indeed, many of the price increases were instituted in 2000, long before many of the improvements occurred. F. 909, 916, 981. For example, only six days after the merger was finalized, ENH reported that it had renegotiated a managed care contract, which was effective January 1, 2000. F. 457. Few quality improvements had occurred that quickly, and several, such as the ambulatory care center, did not become operational until as late as 2005. F. 911.

The evidence discussed below demonstrates that the post-merger post-merger price increases to managed care and, ultimately, EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 299 Initial Decision consumers, were not justified by ENH’s improvements at Highland Highland Park. These improvements, therefore, cannot overcome Complaint Counsel’s strong showing of anticompetitive effects. (c) No Evidence of Improvement in Overall Quality of Care Relative to Other Hospitals Quality of medical care is not easily defined or measured. In fact, Respondent did not present an explanation of how to value the “improvements” or how to compare them to the price increases to managed care organizations. There was significant debate in this case regarding whether several changes made by ENH to Highland Park were, in fact, improvements, and, if improvements, whether they affected quality of care. Quality of care is continually evolving and changing with additional medical developments. In addition, there is no definitive measurement of quality, with one exception, discussed below. Accordingly, the ultimate determination that quality of care improvements do not outweigh the anticompetitive effects of the merger does not rest on the extent to which quality improved, but rather on the fact that most of the improvements to Highland Park were not merger specific and cannot, therefore, be considered in the competitive effects analysis. Just as price increases must be compared to other hospitals’ price price increases to rule out industry-wide changes, the same can be said for quality of care improvements. Complaint Counsel’s healthcare quality expert, Dr. Patrick S. Romano, testified that since since the late 1990's, there has been a nationwide trend of improved improved quality, with one major study finding an average per state state inpatient improvement rate of 12% through 2001. F. 859. Respondent did not provide sufficient evidence to determine whether Highland Park improved its overall quality relative to hospitals generally. As a result, Respondent has not demonstrated whether the improvements are unique to Highland Park and the merger, or simply part of an overall trend unrelated to Highland VOLUME 144 Initial Decision Park’s merger with Evanston. Assertions of quality of care improvements to Highland Park without reference to relative improvements at other hospitals cannot overcome Complaint Counsel’s prima facie case.

Respondent argues that the improvements at Highland Park outweigh any purported anticompetitive effects of the merger. RB at 69-71. Complaint Counsel argues that such improvements did not inure to the benefit of patients who did not use Highland Park, but who were affected by the price increase, because the combined ENH was priced as a single unit. CCB at 14-15. Respondent replies that if quality improved at one part of the integrated ENH system, without a decrease in quality at any other part of the system, then the quality for the whole system would have improved. RRB at 8. Evanston is a larger hospital than Highland Park. F. 5, 22. Significantly more managed care dollars go to treat patients at Evanston than for patients treated at Highland Park. In H.J. Heinz, the D.C. Circuit found that cost reductions must be measured across the new entities’ combined production, not just the premerger output of one of the merging parties. H.J. Heinz, 246 F.3d at 721. Here, ENH did not present evidence establishing that quality improved as a whole over the combined ENH system relative to other hospitals. F. 853-68. As in H.J. Heinz, Respondent failed to present evidence from which the improvements could be measured across the combined entity and therefore any evidence of improvements cannot overcome the showing of anticompetitive effects. The parties argue extensively about whether quality improved in in sixteen areas identified by Respondent. The Court has carefully considered the parties’ arguments and evidence on quality of care, including the extensive data on outcomes, structure, process measures, and patient satisfaction. This quality of care evidence, however, is inconclusive in many instances. For example, Complaint EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 301 Initial Decision Complaint Counsel’s expert, Romano, using U.S. Agency for Healthcare Research and Quality (“AHRQ”) measures found { } at } at Highland Park relative to a control group. F. 861. However, using the Joint Commission on Accreditation of Healthcare Organizations measure (“JCAHO”), Romano found { } at Highland Highland Park, although that evidence was not statistically significant. F. 862. In obstetrics, using the AHRQ measures, Romano Romano found { } using the JCAHO measures. F. 863. This conflict in the evidence may stem, in part, from the different methodologies utilized by AHRQ and JCAHO to risk adjust the data. data. F. 864. These conflicting findings, however, cannot be reconciled on the record provided. In particular, unlike individuals who may consider quality as it relates to a particular service area, managed care organizations consider actual and perceived overall quality of care. Whether quality increased or decreased in a particular service area is not the critical issue in the antitrust analysis. Rather, the focus should be on whether there was an overall overall improvement in quality relative to other hospitals and whether the public perceived Highland Park as providing high quality medical treatment.

The record does not provide definitive evidence on patient satisfaction. Complaint Counsel’s expert relies, in part, on patient satisfaction data from Press Ganey for certain hospital procedures. The reliability of this data, however, is unclear. F. 865-68. Respondent’s healthcare quality expert, Dr. Mark R. Chassin, made a made a rough estimate that the response rate of this data was only about twenty percent, which Complaint Counsel’s healthcare quality quality expert admitted would be suboptimal. F. 867. In addition, the the experts were not aware of the survey methodology used by Press Press Ganey, so that the survey’s trustworthiness could not be determined. F. 868. Again, however, the proper focus should be on on overall quality improvement relative to other hospitals rather than than limited patient satisfaction data. As discussed above, managed managed care organizations were not aware of a significant increase increase in overall quality at Highland Park and believed that it was was an excellent hospital both before and after the merger. F. 846- VOLUME 144 Initial Decision 846-47, 851. Although other evidence of patient satisfaction was presented, none of it presents scientifically valid, comprehensive, and reliable data. In addition, the only industry-wide and nationally nationally recognized measure of overall quality did not demonstrate demonstrate an improvement at Highland Park, as described below. below.

JCAHO regularly evaluates overall hospital quality nationally, including at Highland Park and Evanston. JCAHO accreditation is necessary to qualify for Medicare, as well as most managed care plans. F. 853, 858. In 1999, in its last year before the merger, Highland Park received a preliminary score of 95 and a final score of 96. F. 853. In 1999, Evanston received a preliminary score of 94 and a final score of 95 in 2000 under the same standard. F. 854. These scores are based on approximately 1200 elements of hospital performance. F. 856. In 2002, Highland Park received a JCAHO score of 94. F. 853. Accordingly, based on the JCAHO standard, there is no evidence that the overall quality of care at post-merger Highland Park improved relative to other hospitals. In fact, Highland Park’s JCAHO score declined slightly. Thus, the JCAHO evidence, at least from 1999 to 2002, does not support Respondent’s argument that overall quality of care improved at Highland Park. Rather, Highland Park’s overall quality of service before the merger was excellent and was not declining, as Respondent depicts. After the merger with Evanston, Highland Park continued to maintain its reputation for quality.

(d) Majority of Improvements Were Not Merger Specific To be relevant to the Section 7 analysis, the asserted quality of care improvements, in addition to being verifiable, must be merger merger specific. H.J. Heinz, 246 F.3d at 721-22; Cardinal Health, 12 12 F. Supp. 2d at 62-63; Mercy Health Serv., 902 F. Supp. at 987; Merger Guidelines § 4. In other words, efficiencies (or in this case, case, procompetitive justifications), will be considered only where EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 303 Initial Decision comparable savings or effects cannot “reasonably be achieved by the the parties through other means.” Merger Guidelines at § 3.5. Efficiencies are not merger specific if they could be produced by practical alternatives less restrictive of competition, i.e., generated independent of the merger. University Health, 938 F.2d at 1222 n.30; n.30; Cardinal Health, 12 F. Supp. 2d at 62-63; Long Island Jewish Jewish Med. Ctr., 983 F. Supp. at 147.

For example, in H.J. Heinz, the D.C. Court of Appeals rejected the claim that Heinz could produce better baby food by acquiring Beech-Nut and its recipes. H.J. Heinz, 246 F.3d at 722. The Court of Appeals reasoned that Heinz, on its own and without the need of a merger, could simply invest more money to make a better tasting product. H.J. Heinz, 246 F.3d at 722. Thus, to be cognizable, the benefits of quality of care improvements must be merger specific because otherwise, the benefits could be achieved without the concomitant loss of competition. See H.J. Heinz, 246 F.3d at 722. As explained below, the evidence conclusively demonstrates that the majority of improvements made by ENH were not merger specific. The record establishes that Highland Park would likely have improved quality even without the merger. In 1999, Highland Park Park outlined a premerger strategic plan which included plans to invest more than $100 million to improve its quality of care. F. 871- 871-72. The long range capital budget identified $43 million for investment in strategic initiatives and master plan items such as cardiology services, ambulatory services, oncology, assisted living, living, and facility expansion and $65 million for hospital construction, routine capital, and information technology. F. 873. The investments were to be directed at, among other things: enhancing its core clinical competencies (cardiac surgery, oncology, oncology, and specialty surgery) by itself or with other hospitals, strengthening its medical staff with new doctors and nurses as well well as enhancing leadership and morale, upgrading technology, equipment, and facilities, and increasing patient satisfaction and outcomes so that they would exceed those of competitors and VOLUME 144 Initial Decision national standards. F. 870. Absent the merger, with the need to keep keep itself attractive relative to Evanston and other competing hospitals for managed care and patients, Highland Park would have have had every incentive to continue improving its quality of care. This proposed expenditure of over $100 million compares favorably favorably to the $120 million spent on Highland Park by ENH. Highland Park’s finance committee concluded that based on growth through new clinical services and existing cash and investments and cash flow, the hospital could “generate sufficient cash” to “restore the profitability” of Highland Park and fund the proposed improvements. F. 874. The evidence thus demonstrates that Highland Park had sufficient funds to make the planned improvements to the hospital. See also infra Section III.C.2.e. The evidence thus supports Complaint Counsel’s arguments that Highland Park intended to make improvements, had a history of making improvements, had the economic ability to make improvements, and would have made the improvements because to do so was in Highland Park’s economic self interest. Certainly, the improvements made by Highland Park, without a merger, may have differed from the improvements actually made by ENH. But, the antitrust inquiry is not whether Highland Park would be identical today, absent the merger, but only whether the improvements made by ENH are merger specific. Except for two quality improvements discussed below, the answer is no. Therefore, the expenditures by ENH for improvements to Highland Park cannot overcome Complaint Counsel’s evidence of anticompetitive effects because Highland Park could have made all but the two improvements without merging with Evanston.

Respondent’s claimed quality improvements generally fall into into three categories: (1) new or improved facilities or equipment; (2) increased staffing, improved training, and culture of teamwork; teamwork; and (3) new or improved procedures. None of these changes are merger specific. With sufficient funds, new or improved EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 305 Initial Decision improved facilities or equipment could have been purchased. With proper funding Highland Park could have increased staff and in many areas; Highland Park had already begun improvements to training and teamwork. Contrary to ENH’s assertion, a change in culture does not emanate only from a merger – it can occur as the result of different management or in response to recommendations from outside organizations. Similarly, it does not take a merger for a for a hospital to implement new procedures. The only two benefits benefits that would not have been achieved absent the merger are the the acquisition of the state of the art EPIC computerized records management system and the academic affiliation and clinical integration. These two merger specific improvements are discussed discussed below in section III.C.2.b.3.e. The other fourteen of Respondent’s improvements were not merger specific, as explained explained immediately below.

(i) Obstetrics and Gynecological Services At the time of the merger, the Obstetrics and Gynecological (“Ob/Gyn”) department was the largest patient care area at Highland Park. F. 876. After the merger, ENH instituted nighttime and weekend coverage by obstetricians; installed a full-time chair of the Ob/Gyn department; improved nurse training models of care; instituted an Ob/Gyn preoperative surgery review program; and initiated physician discipline proceedings against a few of Highland Park’s Ob/Gyn physicians. F. 877-82. Respondent argues that Highland Park had major quality deficiencies, including inadequate coverage, lack of effective leadership, inadequate nursing, inappropriate practice patterns, and a weak quality assurance program in its delivery of obstetrics and gynecological services. RB at 75-77. According to Respondent, these problems were identified in 1998, but corrections were not instituted until ENH intervened after the merger. RB at 75-76.

Prior to the merger, Highland Park had invited the { } review of review of the hospital as part of its ongoing effort to improve quality quality of care. F. 883. { } made a number of recommendations to VOLUME 144 Initial Decision to improve the{ } F. 884. Many changes were made in response to response to the { } report. F. 885. In fact, Highland Park’s efforts to efforts to implement { } recommendations were subsequently recognized by the Chicago Hospital Risk Pooling Program after a site visit and report issued in November of 1999. F. 886. The Chicago Hospital Risk Pooling Program made additional recommendations for improvement. F. 887. The evidence demonstrates that Highland Park was aware of the need to improve and was, in fact, making the necessary improvements. There is no evidence that Highland Park was incapable of changing its Ob/Gyn nursing culture, rather, the evidence shows that Highland Park was aware of and actively taking steps to change the culture, but that such changes take time. F. 885- 86, 903-10. The improvements made by ENH to obstetrics and gynecological services could have been implemented by Highland Park without merging with Evanston.

(ii) Quality Assurance Program ENH changed the structure within the clinical departments of how oversight of physicians was conducted by replacing part-time and private practice chairs with full-time ENH clinician chairs; took disciplinary action against a number of Highland Park physicians; and reviewed physician practices during periodic recredentialing. F. 888-90. Respondent criticizes Highland Park’s premerger quality assurance program as being ineffective. RB at 77-82. Highland Park, premerger, regularly had initiated disciplinary actions against its physicians, including the suspension, reduction, or or removal of staff privileges. F. 892. There are a number of examples of Highland Park’s review of adverse events prior to the merger and it is not clear whether the culture actually improved under ENH. F. 893. Indeed, { } was requested by Highland Park, Park, premerger, because of a disciplinary action in the { } F. 894. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 307 Initial Decision 894. The quality assurance changes made by ENH at Highland Park Park after the merger merely reflect the emerging consensus in the field of quality assurance. F. 895. Highland Park had an active quality assurance program and the Court is persuaded that it, like many hospitals, likely would have kept up with the emerging consensus in the field of quality assurance. In addition, Highland Park could have utilized clinical department heads, if it had chosen chosen to organize its departments in that manner, without merging. merging. Thus, improvements to the quality assurance program could have been implemented by Highland Park through means other than the merger with ENH.

(iii) Quality Improvement Program Critical pathways and care maps are protocols identifying best practices for treatment of patients. F. 897. After the merger, the critical pathways at ENH were aligned with the care maps being used at Highland Park, improving both. F. 896. Respondent criticizes Highland Park’s premerger quality improvement program as being inadequate. RB at 77-82.

Highland Park’s strategic plan for 1999-2003, included among its goals to: provide documented and measurable outcomes of quality which exceed those of the competition and establish national national standards and provide a continuum of care for the patient across the delivery system including providing the highest quality clinical and non-clinical services. F. 870. Prior to the merger, Highland Park conducted an internal review of quality programs which highlighted areas for improvement. F. 898. Nothing in the record suggests that ENH’s critical pathways were better than the care maps used by Highland Park before the merger or that Highland Highland Park would not have continued to develop other care maps maps after 1999 on its own. F. 899. Indeed, the evidence does not clearly show whether the critical pathways are always being followed. F. 900. The evidence demonstrates that critical pathways pathways are constantly being revised and improved and Highland Highland Park likely would have continued to make similar VOLUME 144 Initial Decision improvements to its care maps. F. 901. The quality improvement changes made by ENH at Highland Park after the merger merely reflect the emerging consensus in the field of quality improvements. improvements. F. 902. Thus, improvements to the quality improvement program could have been implemented by Highland Park without merging with Evanston.

(iv) Nursing Staff ENH improved communication and teamwork between nurses and physicians; improved nurse training; and eventually improved recruiting, vacancy, and turnover rates. F. 903-05, Respondent claims that Highland Park lacked several key elements of an effective nursing program and that without the cultural change that ENH brought to Highland Park, nursing services would not have improved. RB at 83-84.

Highland Park had a “high quality nursing staff” in the 1990's. F. 907. Nonetheless, in 1999, Highland Park adopted a comprehensive initiative to train, retain, and reward its nurses. F. 908. The nursing culture at Highland Park underwent a transition from a punitive and dysfunctional culture to a much more effective culture over a period of years beginning before the merger and continuing until 2004. F. 909. The change in the nursing culture was an evolutionary process that took many years. F. 910. Indeed, it seems highly unlikely that Highland Park is unusual in having nurse staffing problems. The evidence is clear, however, that Highland Park was aware of and committed to improving these problems. Improvements in the nursing staff could thus have been implemented by Highland Park without merging with Evanston.

(v) Physical Plant ENH built a new ambulatory care center which opened in February 2005, which houses radiation medicine, nuclear medicine, EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 309 Initial Decision medicine, the Kellogg Cancer Care Center, and a new breast imaging center. F. 911. ENH built a new cardiac cath lab to support support the interventional cardiology program; renovated and expanded the emergency department and psychiatry units; and added added modern equipment in a variety of areas. F. 912. ENH replaced replaced the Highland Park patient care building’s electrical distribution and ventilation systems, plumbing, and waste pipes and and built a new central plant at Highland Park, including a new power plant that houses utilities such as electric generators, backup backup generators, boilers, and air ventilation equipment. F. 913-14. 913-14. ENH added an additional boiler, new air handlers for the ventilation system, replaced the electrical generator, and added a second emergency electrical generator. F. 915. ENH began remodeling all of its patient rooms in December 2003. F. 916. The process of remodeling patient rooms is continuing and scheduled at at least through 2006. F. 916. ENH added a new parking garage and and made improvements to the lobby corridor and entrance to Highland Park. F. 917. Respondent asserts that it invested millions of dollars into expansions and renovations of Highland Park’s facilities. RB at 85-86.

On April 15, 1999, the Illinois Department of Public Health and Healthcare Financing Administration performed a facility survey of Highland Park which identified 144 physical plant deficiencies that needed to be corrected to continue to participate in Medicare. F. 918. On August 26, 1999, 26 items were removed from the list and 3 were added, for a total of 121 deficiencies. F. 919. On December 9, 1999, a reinspection was conducted and 88 additional items were removed from the list, leaving a total of 33 items. F. 920. The plan for correction of these remaining items was submitted by Highland Park on December 28, 1999 and these remaining items were corrected by ENH by August 1, 2000. F. 920. Highland Park was aware of and had addressed or planned to address all of the issues identified during these inspections.

The evidence does not demonstrate that ENH’s expenditures were merger specific because, as previously noted, premerger, VOLUME 144 Initial Decision Highland Park had budgeted a total of $108 million in capital expenditures through 2003, for, among other things, upgrading technology, equipment, and facilities. F. 872-73. The financial condition of Highland Park would have allowed it to make these improvements to its physical plant. F. 1028-69. Thus, improvements improvements to the physical plant could have been implemented by by Highland Park without merging with Evanston. (vi) Oncology Services Through the Kellogg Cancer Center at Highland Park, ENH implemented a multidisciplinary approach that brought together an oncology team consisting of the physician oncologist, nurse, pharmacist, psychologist, social workers, and nutritionists who were available to patients in one location. F. 920. ENH brought subspecialty oncologists to Highland Park so that patients would not have to travel for their consultations. F. 922. The Kellogg Cancer Center moved into a new section of the ambulatory care center in March 2005. F. 923. Respondent points to the benefits of improvements in the delivery of oncology services at Highland Park through the expansion of the Kellogg Cancer Center as a merger specific improvement. RB at 86-88.

Before the merger, Highland Park had already undertaken numerous initiatives in oncology services and had a variety of options other than the merger to achieve these same ends. F. 924. Highland Park also had detailed plans to expand multi-disciplinary multi-disciplinary oncology services alone or with other hospitals. F. F. 925. Highland Park had considered joint comprehensive oncology oncology programs with organizations other than ENH. F. 925. In the 1990's, Highland Park had created centers of excellence for oncology and breast cancer that it was continually improving until the time of the merger. F. 926. These centers of excellence already had access to the necessary technology, physicians, and research protocols in place to develop a comprehensive oncology program, and Highland Park merely needed to develop the community EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 311 Initial Decision perception of excellence in these areas. F. 927. To this end, Highland Highland Park could have expanded its oncology services and research activities through an affiliation agreement with a hospital other than ENH and, in fact, it had been exploring this option at the the time of the merger. F. 928. Thus, improvements to oncology services could have been implemented by Highland Park without merging with Evanston.

(vii)Radiology, Radiation Medicine, and Nuclear Medicine ENH purchased a linear accelerator for Highland Park; added two new CT scanners in Highland Park’s radiology department; upgraded radiation therapy equipment; and purchased additional equipment. F. 929-30. ENH purchased a CT pet, a diagnostic tool, for the nuclear medicine department. F. 931. ENH extended RADNET, its radiology imaging system and PACS, its filmless radiology imaging system, to Highland Park. F. 932. ENH added additional radiology staff to improve turnaround times for reading radiology reports. F. 933. Respondent claims that these changes, including the significant investment in new equipment, improved the radiology services at Highland Park. RB at 91. Highland Park had a premerger budget of $9.5 million to improve radiology services. F. 934. Highland Park had the resources and the commitment to improve radiology, radiation medicine, and nuclear medicine. Thus, improvements to radiology services could have been implemented by Highland Park without merging with Evanston.

(viii) Emergency Care ENH improved both the physical layout and service components components of Highland Park’s emergency department. F. 935. ENH ENH expanded physician coverage; renovated and expanded facilities; improved physician and nurse staffing; and improved the the fast track procedure in the emergency department. F. 936. VOLUME 144 Initial Decision Respondent claims that it has significantly improved the emergency emergency care rendered at Highland Park. RB at 89-90. Prior to the merger, the emergency department at Highland Park was “very good,” and was “on par, if not better” than Highland Park’s peers. F. 937. Throughout the 1990's, Highland Park had continually made improvements to its emergency care: it had implemented a fast-track program to improve turnaround times; it had added physician assistants to the emergency room; it had streamlined the radiology process; and it had reduced the time that it took for a patient to receive an EKG. F. 938. Further, Highland Park planned to “expand the Emergency Department from a facilities standpoint.” F. 939. In fact, Highland Park could have made the changes to the emergency department absent the merger. For example, most emergency departments at hospitals like Highland Park are staffed through contracts with physician groups, and Highland Park simply could have “demanded” higher staffing of the emergency room as a condition of its contract. F. 940. Thus, improvements in emergency care could have been implemented by Highland Park without merging with Evanston. (ix) Laboratory Medicine Prior to the merger, Highland Park operated Consolidated Medical Labs (“CML”), a joint venture with Lake Forest that consisted of a main laboratory located between the two hospitals with satellite laboratories at Highland Park and Lake Forest. F. 941. After the merger, ENH decided to close CML and expand the on-site laboratory at Highland Park. F. 942. Certain tests are sent to the lab at Evanston. F. 942. ENH constructed new histology and cytology laboratories at Highland Park, installed over $1 million in state of the art lab equipment, and introduced more stringent quality controls. F. 943. Respondent asserts that it made significant changes in the laboratory services that were furnished at Highland Park. RB at 90-91.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 313 Initial Decision Prior to the merger, Highland Park’s joint venture for laboratory services with Lake Forest operated “actually exceptionally well.” F. 944. CML afforded Highland Park’s lab “greater volume,” “access to greater human pathology,” and the “opportunity to provide a greater benchmark in terms of [the lab’s] performance.” F. 944. Highland Park could have implemented further changes in its laboratory in the absence of the merger. F. 945. Many of the changes that ENH made after the merger were simply consistent with updates that all hospital laboratories made during that period in order to meet licensing and accreditation standards. F. 946. Thus, improvements in the laboratory services could have been implemented by Highland Park without merging with Evanston.

(x) Pharmacy Services ENH installed twenty Pyxis automated drug distribution machines at Highland Park in 2000. F. 947. In the summer of 2003, ENH added an additional pharmacist to dispense medications at night. F. 949. ENH decentralized the pharmacists. F. 948. Respondent highlights changes to pharmacy services at Highland Park, including the installation of Pyxis, as a quality of care improvement. RB at 91-92.

Highland Park’s strategy prior to the merger was to implement “the latest technology to support patient care across the continuum.” F. 870. The Pyxis system did not become available to hospitals until the late 1990's, when there was a “trend” in which pharmaceuticals and medications were decentralized in order to be located within the individual units. F. 950. Pyxis costs about $20,000 per machine, and Highland Park could have installed the machines on its own. F. 951. Thus, improvements in the pharmacy services, including the installation of Pyxis or a similar system, could have been implemented by Highland Park without merging with Evanston. VOLUME 144 Initial Decision (xi) Cardiac Surgery ENH opened a cardiac surgery program at Highland Park in June 2000. F. 952. Cardiac surgery is a necessary component of a fullservice cardiology program. F. 953. Cardiac surgery procedures include coronary artery bypass grafting, valve procedures, and surgery on the aorta. F. 954. Respondent touts the benefits of introducing cardiac surgery and interventional cardiology programs at Highland Park. RB at 94-96.

Before the merger, Highland Park already had plans to open a cardiac surgery program with Evanston or another hospital. F. 955. Highland Park also considered a joint cardiac surgery program with Northwestern Memorial or Advocate Lutheran General. F. 956. ENH runs successful joint cardiac surgery programs with Swedish Covenant and Louis A. Weiss. F. 957. Highland Park and Evanston executed a contract for a joint cardiac surgery program before the merger. F. 958. The Certificate of Need Application for the Highland Park cardiac surgery program suggests that the collaboration necessary to implement the program did not depend on the merger. F. 959. Thus, improvements in cardiac surgery and interventional cardiology could have been implemented by Highland Park without merging with Evanston.

(xii) Interventional Cardiology Interventional cardiology refers to the treatment of obstructions in coronary arteries (coronary disease) by dilating the plaques obstructing the arteries and inserting little wire tubes called stents to keep the arteries open. F. 960. After the merger, ENH established an interventional cardiology program at Highland Park. F. 961. ENH built a new cardiac catheterization lab at Highland Park that performs both diagnostic and interventional procedures such as angioplasties. F. 962.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 315 Initial Decision Highland Park’s premerger medical staff included physicians with the expertise to perform interventional cardiac procedures. F. 963. Highland Park planned to expand the diagnostic capabilities of its existing cardiac catheterization lab and to provide emergent angioplasty in conjunction with the planned cardiac surgery program or even “without open heart on-site.” F. 964. Thus, improvements to interventional cardiology could have been implemented by Highland Park without merging with Evanston.

(xiii) Psychiatry Before the merger and through the spring of 2001, Highland Park and Evanston each had separate inpatient psychiatric units that treated both adult and adolescent patients. F. 965. In the spring of 2001, ENH consolidated the adolescent inpatient services at Highland Park and the adult inpatient services at Evanston. F. 966. ENH hired several, adolescent psychiatrists to staff the Highland Park adolescent unit. F. 967. ENH remodeled the psychiatric unit in December 2003 to include private patient rooms with a keyless entry system and secure furniture. F. 968.

Highland Park could have chosen to refer its adult patients to Evanston or another hospital and expand its adolescent services without the merger. In addition, Highland Park could have chosen to expand its adolescent services, without the merger and without closing the adult services.

(xiv) Intensivist Program ENH added an intensivist program to Highland Park after the merger. F. 970. An intensivist is a physician who specializes in the care of intensive care patients and who has more experience dealing with the complications of these critically ill people. F. 971. Intensivists also have an administrative role in overseeing and coordinating the medical and nursing staff that provide care to critically ill patients. F. 972. Respondent claims credit for the intensivist program at Highland Park. RB at 96-97. VOLUME 144 Initial Decision Intensivist programs in hospitals like Highland Park became popular only after the merger. F. 973. Pulmonary Physicians of the North Shore, which provides the intensivist coverage at Highland Park, does so through a contractual arrangement. F. 974. Highland Park did not need to merge with Evanston in order to provide the intensivist services currently provided by Pulmonary Physicians of the North Shore. Highland Park could independently contract to have an intensivist program. F. 975. Thus, the intensivist program could have been implemented by Highland Park without merging with Evanston.

(e) Merger Specific Improvements The Court next addresses the two previously mentioned improvements made by ENH which the Court does find to be merger specific. Upon integration with ENH, medical care providers at Highland Park had access to comprehensive medical records through a state of the art computerized information system known as EPIC. In addition, the merger provided academic affiliation and clinical integration. These benefits could only reasonably have been achieved through the merger with Evanston. Especially in the latter case, these were not benefits that a stand alone Highland Park could have obtained.

(i) Electronic Medical Records System In 2001, ENH decided that its current medical records system was not sufficient to meet the needs of its three hospitals and ENH ENH began its search for a better system. F. 976. In June 2001, the the EPIC system was selected from a group of competing technologies. F. 977. EPIC is a nationally recognized software system for managing patient records for both hospitals and physicians and was selected, in part, for its ability to work with physician offices. F. 978. The EPIC system was implemented in order to integrate records from health care providers who practiced practiced at all three ENH hospitals, at the faculty practice medical EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 317 Initial Decision medical group, and at all the affiliated physician practices that were were willing to participate. F. 980.

The use of EPIC allows physicians to review records of other care givers that have seen a patient. EPIC became functional at Highland Park in December 2003. F. 981. Because EPIC currently integrates information from three hospitals and seventy physician offices, care givers who currently use EPIC at Highland Park have access to a wealth of information. F. 980. EPIC includes a computerized physician order entry system and clinical decision support systems. F. 979. Respondent rightfully contends that it improved quality at ENH by installing the EPIC system. RB at 97- 99. EPIC is thus a merger specific improvement in that Highland Park, as an independent hospital, was unlikely to license such a state of the art, comprehensive system.

The evidence, however, does not establish that a stand alone Highland Park would have needed to change its medical records system to EPIC. Meditech, the medical records system used by Highland Park before the merger, was and is an “excellent” system that other hospitals continue to use today. F. 986. The Meditech system, however, was not state of the art. For example, Meditech, as deployed at Highland Park, was not paperless, could not be accessed remotely, and lacked ambulatory capability. F. 985. Even if an independent Highland Park licensed EPIC, the benefits would be limited by the fewer number of health care providers linked into the system.

The federal government has established a national initiative to develop a universally accessible electronic healthcare record for all all citizens and in 2004, the Office of National Healthcare Information Technology was created to achieve this end. F. 987. Therefore, medical records systems and technology are likely to continue to evolve, and EPIC may not remain the state of the art system that it is today. Even if EPIC is maintained by Highland Park, much of the integrated benefit will be lost because the other ENH hospitals and physician providers would not, presumably, be VOLUME 144 Initial Decision connected to the same licensed system. A stand alone Highland Park, Park, however, would not require the same level of integration that that currently exists with ENH.

(ii) Academic Affiliation and Clinical Integration As previously noted, the merger did not transform Highland Park into an academic hospital. Indeed, family medicine is the only department at Highland Park that utilizes residents and at the time of trial the department maintained only 6 residents. F. 988. There is no evidence that Highland Park benefitted simply by being owned by a teaching hospital. F. 993.

However, since the merger, physicians in pathology, radiology, emergency medicine, cardiology, cardiac surgery, and some parts of anesthesiology rotate through all three ENH campuses. F. 989. Following the merger, about sixty Highland Park physicians obtained appointments at Northwestern Medical School. F. 990. This interaction with Northwestern Medical School is clearly a merger specific benefit. The evidence does not establish, however, that the relationship with Northwestern Medical School had a noticeable impact on quality of care of patients, patient satisfaction, or improved structure, process, or outcomes. See F. 853-68. Nonetheless, it has been a benefit to the physicians who were able to obtain faculty appointments and this relationship may have encouraged some top physicians to join the staff at Highland Park. This affiliation with the medical school will be lost upon divestiture. (3) Merger Specific Quality of Care Improvements Do Not Outweigh Probable Anticompetitive Effects As discussed, the vast majority of improvements at Highland Park were not merger specific. The Court is aware of the significant EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 319 Initial Decision significant improvements at Highland Park, including the substantial substantial time and resources taken to fund and make such improvements a reality. The Court is also cognizant that Highland Park, under ENH, continues to be an excellent hospital. The finding finding that the majority of the alleged procompetitive justifications justifications are not merger specific in an antitrust context is in no no way intended to undermine their importance to care givers or patients at ENH. However, their ultimate impact on overall relative relative quality of patient care, patient satisfaction, and outcomes is is limited. F. 853-68. Considering the persuasive evidence of the merger’s anticompetitive effects, Respondent’s two merger specific specific improvements to Highland Park, if legally cognizable and relevant to the analysis, do not sufficiently outweigh the merger’s harm to competition and ultimately to consumers. Even if Respondent’s quality of care theory was valid and countered the direct evidence of anticompetitive effects (price increases), the quality improvements are not relevant to the structural evidence of of market concentration. Nor do Respondent’s remaining defenses, defenses, including nonprofit status, ease of entry, and failing firm, firm, save the merger.

c. Nonprofit Status Respondent has argued that ENH’s nonprofit mission reduces the potential for competitive harm. Specifically, Respondent asserts that ENH has a deep commitment to the community; that the ENH Board consists largely of members of the community; that ENH provides benefits to the community, including charity care and new services; and that ENH created an independent foundation which provides grants to local organizations. RB at 65-67. Respondent further asserts that courts have recognized that the nonprofit status of hospitals may be taken into account in evaluating a merger case. RB at 65-66.

Complaint Counsel asserts that ENH’s nonprofit status did not prevent ENH from exercising market power and that ENH’s management structure, just like for profit entities, created incentives VOLUME 144 Initial Decision incentives for ENH to raise prices, including awarding significant bonuses and salary increases for achieving revenue and income growth. CCRB at 36-37. Complaint Counsel further asserts that courts have explicitly rejected the argument that a hospital’s nonprofit status renders a merger not anticompetitive. CCRB at 36- 36-37.

In both Rockford Memorial and Hospital Corporation of America, the Court of Appeals for the Seventh Circuit rejected hospitals’ arguments that their nonprofit status removed ground for concern that hospitals might seek to maximize profits through avoidance of price or service competition. Rockford Memorial, 898 F.2d at 1285; Hospital Corp. of Am., 807 F.2d at 1390. As explained in Rockford Memorial:

We are aware of no evidence – and the [appellees] present none, only argument – that nonprofit suppliers of goods or services are more likely to compete vigorously than profit-making suppliers. . . . If the managers of nonprofit enterprises are less likely to strain after that last penny of profit, they may be less prone to engage in profit-maximizing collusion but by the same token less prone to engage in profit-maximizing competition.

Rockford Memorial, 898 F.2d at 1285.

Moreover, the Seventh Circuit has stated that “[t]he adoption of the nonprofit form does not change human nature . . . , as the courts have recognized in rejecting an implicit antitrust exemption for nonprofit enterprises.” Hospital Corp. of Am., 807 F.2d at 1390 (citing Natl Collegiate Athletic Assn v. Bd. of Regents, 468 U.S. 85, 100 n.22 (1984)). “‘Nonprofit hospitals, in fact, make rather sizable profits and these profits have been growing over time.’” Hospital Corp. of Am., 807 F.2d at 1390 (citation omitted). EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 321 Initial Decision Respondent points to district court cases that recognized that the nonprofit status of hospitals may be taken into account in evaluating the potential anticompetitive effects. RB at 65-67 (citing Long Island Jewish Med. Ctr., 983 F. Supp. at 146; Butterworth Health Corp., 946 F. Supp. at 1296-97; United States v. Carilion Health Sys., 707 F. Supp. 840, 849 (W.D. Va. 1989) (unpublished opinion)). In these cases, the district courts were asked to speculate about the potential effects of a proposed merger and each held that the nonprofit status might serve as a check on anticompetitive behavior. E.g., Butterworth Health Corp., 946 F. Supp. at 1297 (nonprofit status was material where economist’s findings suggested that the proposed merger was not likely to result in price increases). But in this case, there is substantial evidence of actual price increases postmerger. F. 326-755. Thus, an inquiry into whether the nonprofit status of the hospitals might serve as a check on price increases is not a relevant inquiry. See Hospital Corp. of Am., 807 F.2d at 1390 (While “different ownership structures might reduce the likelihood of collusion, . . . this possibility is conjectural.”). Further, the court in Long Island Jewish Medical Center held only that “nonprofit status may be considered if supported by other evidence that such status would inhibit anticompetitive effects.” 983 F. Supp. at 146. In this case, Respondent has presented evidence that the Healthcare Foundation of Highland Park provides funds to support indigent or uncompensated care at Highland Park, dispenses grants to charities in the Highland Park area, and has improved access to healthcare for underserved populations in southeast Lake County. F. 1012. This evidence, however, does not overcome the convincing evidence presented by Complaint Counsel that ENH’s nonprofit status has not inhibited the anticompetitive effects of the merger. See F. 326-755.

Although ENH’s Board of Directors contains community representatives, the ENH board did not actively monitor the pricing pricing decisions of hospital management. F. 1003. Further, the VOLUME 144 Initial Decision senior executives of ENH received enhanced compensation agreements and substantially higher awards at the end of 2000 compared to the awards in 1998 and 1999. F. 998-1000. Thus, ENH’s compensation contracts did not align management’s interests interests with consumers on the issue of price. F. 1001. And, most importantly, when ENH set prices for the 2000 contract renegotiations with managed care organizations, the fact that it was a was a nonprofit entity did not restrain its efforts to obtain higher prices. See F. 326-755. Thus, the evidence in this case is consistent consistent with cases holding that “if there is the potential for anticompetitive behavior, there is nothing inherent in the structure of of the corporate board or the nonprofit status of the hospitals which which would operate to stop any anticompetitive behavior.” Mercy Mercy Health Serv., 902 F. Supp. at 989. The entirety of the evidence, including ENH’s contemporaneous documents, testimony, and the post-merger pricing data, shows that ENH exercised market power and that its nonprofit status was irrelevant to that end. Accordingly, Respondent’s nonprofit status does not rebut Complaint Counsel’s prima facie case. d. Entry or Expansion Concentration in the relevant market may not inherently lead to to collusive or anticompetitive behavior when existing competitors competitors could easily enter the market and provide enough capacity to defeat an exercise of market power. See Hospital Corp. Corp. of Am., 807 F.2d at 1387; Long Island Jewish Med. Ctr., 983 983 F. Supp. at 149 (“A merger is not likely to cause an antianti-competitive effect if other participants can enter the relevant markets and reduce the likelihood of a price increase above competitive levels.”). If customers could turn to new entries in the market in sufficient numbers to make the exercise of market power power unprofitable for merging hospitals, then any present concentration in the relevant market would be irrelevant. Rockford EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 323 Initial Decision Rockford Memorial, 717 F. Supp. at 1281. Therefore, among factors factors which courts have previously considered to be relevant is ease of entry into the market. Id. “Most hospital cases have stated the inability to build new hospitals as a strong barrier to entry.” Mercy Health Serv., 902 F. Supp. at 986. It is against this standard that the record is reviewed to determine the relative ease or difficulty difficulty of entering the relevant market. Respondent asserts that, in order for a merger to harm competition, repositioning by the non-merging firms must be unlikely. RB at 58. Respondent argues that Complaint Counsel has not demonstrated significant barriers to expansion such that rival hospitals would be unable to reposition themselves to compete with ENH. RB at 58. Respondent further asserts that competitor hospitals are able to and have expanded their capacity and service offerings. RB at 59. Complaint Counsel counters that evidence of hospitals’ actions to expand capacity or enter in the area is not sufficient to constrain and has not constrained ENH’s prices. CCRB at 33 n.34. A new entrant must overcome significant regulatory barriers to enter the relevant market. The Illinois Certificate of Need (“CON”) law presents a barrier to persons wishing to provide new acute hospital inpatient care in the relevant geographic market. See F. 1014. The Illinois Health Facilities Planning Board, when reviewing a CON application for additional beds, considers whether the proposed beds are actually needed at the facility. F. 1016. Other hospitals can intervene to oppose a hospital’s CON application. F. 1020. Based on the Planning Board’s current addendum to its inventory, there is no need for additional beds in the Evanston, Glenview, and Highland Park areas for services in medical/surgical, pediatrics, or intensive care units. F. 1018. Moreover, there have been no CON applications for the construction of new hospitals in the area around Highland Park, Evanston, or Glenbrook over the past five years. F. 1021. No new entry by a hospital has occurred in the North Shore area since the VOLUME 144 Initial Decision merger. F. 1027. And, while the regulatory environment for entry and expansion may ease if the Illinois CON law is repealed, as scheduled for July 1, 2006 (F. 1023), any effect this may have on entry or repositioning by incumbent providers is speculative. Further, irrespective of the CON law, it takes about two and a half to to three years to build a new hospital. F. 1024. The critical question is whether expansion from existing hospitals or entry by new hospitals is sufficient to constrain ENH’s prices. Cardinal Health, 12 F. Supp. 2d at 55-58 (entry or expansion “must be proven to 'be timely, likely, and sufficient in its magnitude, character and scope to deter or counteract the competitive effects of concern’”) (quoting Merger Guidelines, § 3.0); Chicago Bridge & Iron, Dkt. No. 9300, at 31. See also Staples, 970 F. Supp. at 1088 (finding that expansion by Wal-Mart would not constrain the merging parties’ prices). The evidence in this case clearly shows that other hospitals do not significantly constrain ENH’s prices. See F. 326-755.

The substantial evidence in this case is that expansion from existing hospitals has not counteracted the ENH price increases implemented subsequent to the merger, There is insufficient evidence that new entry or repositioning by rival hospitals will be timely, likely, and sufficient in its magnitude, character, and scope to constrain ENH. Therefore, the evidence does not demonstrate that entry or expansion is likely to replace the competition lost through the acquisition or to sufficiently constrain ENH from future anticompetitive actions.

e. Failing Firm Respondent also asserts that, at the time of the merger, Highland Highland Park was in a deteriorating financial condition, which, it argues, is an additional factor contributing to a finding that the merger did not substantially lessen competition. RB at 61-65. Complaint Counsel asserts that Highland Park’s premerger financial EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 325 Initial Decision financial condition was sound and that Highland Park could have pursued an arrangement – a sale, merger, or alliance – with another another entity that would have resolved any financial issues without without the attendant antitrust problems of this merger. CCRB at 33- 33-36.

The acquired firm’s weakness is a factor that a defendant may introduce to rebut the government’s prima facie case. University Health, 938 F.2d at 1221; Kaiser Aluminum, 652 F.2d at 1339; United States v. Intl Harvester Co., 564 F.2d 769, 774 (7th Cir. 1977). “A weak financial condition may mean that a company will be a far less significant competitor than current market share, or production statistics, appear to indicate.” FTC v. Arch Coal, Inc., 329 F. Supp. 2d 109, 153 (D.D.C. 2004). However, such a defense is credited “only in rare cases, when the defendant makes a substantial showing that the acquired firm’s weakness, which cannot be resolved by any competitive means, would cause that firm’s market share to reduce to a level that would undermine the government’s prima facie case.” University Health, 938 F.2d at 1221. “Since weak firms are not in grave danger of failure . . . it is not certain that their weakness ‘will cause a loss in market share beyond what has been suffered in the past, or that such weakness cannot be resolved through new financing or acquisition by other than a leading competitor.’” Id. (citation omitted).

A merger is not likely to create or enhance market power or facilitate its exercise if the following circumstances are met: (1) the allegedly failing firm would be unable to meet its financial obligations in the near future; (2) it would not be able to reorganize successfully under Chapter 11 of the Bankruptcy Act; (3) it has made unsuccessful good-faith efforts to elicit reasonable alternative offers of acquisition of the assets of the failing firm that would both keep its tangible and intangible assets in the relevant market and pose a less severe danger to competition than does the proposed merger; and (4) absent the acquisition, the assets of the failing firm would exit the relevant market. Merger Guidelines § 5.1; Arch Coal, 329 F. Supp. 2d at 154.

VOLUME 144 Initial Decision In this case, Respondent failed to make such a showing. The evidence demonstrates that Highland Park’s premerger financial condition was essentially sound. It had more than sufficient cash and assets to cover debts ($235 million in cash and assets, compared to $120 million in long-term debt), continue operations, expand services, and invest in new facilities and equipment. F. 1044-45. In developing Highland Park’s 1999-2003 financial plan, the Lakeland finance and planning committee noted, “[c]ash and investments are forecasted to grow from $238 million in 1998 to $323 million in 2003,” forecasted that its investments would generate a return of $28 million in incremental net revenues in 2003, emphasized that “[e]existing cash and investments are available to fund strategic initiatives and generate new programs,” and concluded that Highland Park “can remain financially strong over the foreseeable future.” F. 1029, 1031-33. Highland Park’s 1999-2004 Financial Plan projected that cash and investments would increase by $48 million from 1999-2004, and that long-term debt would be reduced by $24.3 million, excluding amortization, and projected that it had sufficient cash flow for both planned capital expenditures ($79 million) and planned strategic initiatives ($24 million). F. 1037-38. The Highland Park board and management was advised that “the financial condition of Highland Park was such that it did not require a financial reason to go forward with the merger.” F. 1040. In the fall of 1998, Highland Park contemplated both a merger strategy, as well as an independent, stand alone growth strategy. F. 1056. Highland Park was prepared to proceed with the status quo, unaffiliated option if the ENH merger talks failed. F. 1057. If the merger with ENH had not closed, Highland Park had “the financial financial wherewithal to sustain [itself].” F. 1059. Highland Park management and board believed that “[t]here was no urgency to have an alternative immediately available.” F. 1059. Stearns, Highland Park’s Chairman of the Board, testified that he believed that Highland Park was not in danger of exiting the market for at least ten years. F. 1059. Highland Park never considered filing for EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 327 Initial Decision bankruptcy. F. 1064. This stands in marked contrast to the facts in FTC v. Freeman Hospital, where the hospital to be acquired had a limited future of only two to three years. 911 F. Supp. 1213, 1225 (W.D. Mo. 1995), aff’d, 69 F.3d 260 (8th Cir. 1995). In the fall of 1998, Highland Park contemplated a number of potential merger partners besides Evanston, including Northwest Community, Lake Forest, and Condell. F. 1065. If the ENH merger had not closed, Highland Park was prepared to continue to explore other partnership options. F. 1067. Highland Park had “an attractive service area,” and therefore, Highland Park’s chairman of the board believed it “would be attractive to other partnership candidates.” F. 1069.

The evidence in this case thus demonstrates that Highland Park was able to meet its financial obligations in the near future; was not in danger of bankruptcy; was exploring other options, including remaining a stand alone entity; and was not in danger of exiting the market in the foreseeable future. Therefore, Respondent has failed to show that, because of Highland Park’s financial prospects, Complaint Counsel’s prima facie case does not accurately reflect the acquisition’s likely effect on future competition. D. Affirmative Defense 1. Evanston and Highland Park Are Separate Persons Subject to Section 7 of the Clayton Act Respondent asserts as an affirmative defense that prior to the merger, Evanston and Highland Park were not separate persons as required for the application of Section 7 of the Clayton Act, and that that the merger is exempt from antitrust liability under the Copperweld doctrine (Copperweld Corp. v. Independence Tube Co., Co., 467 U.S. 752 (1984)). Answer, p. 20. In Copperweld, the Supreme Court held that a parent company and its wholly-owned subsidiary, as a single entity, were not capable of conspiring in VOLUME 144 Initial Decision violation of the Sherman Act. 467 U.S. at 771. Section 7 of the Clayton Act provides in pertinent part: “[n]o person . . . shall acquire, directly or indirectly, the whole or any part of the stock or or other share capital . . . [or] the whole or any part of the assets of of another person” when “the effect of such acquisition may be substantially to lessen competition, or to tend to create a monopoly.” monopoly.” 15 U.S.C. § 18 (2005).

Respondent asserts that the merger of Evanston and Highland Park did not involve two “persons” because at the time of the merger they were sister corporations owned by the same parent. RB at 110- 13. Complaint Counsel asserts that Evanston and Highland Park were “separate economic actors pursuing separate economic interests” and thus do not qualify for the Copperweld defense. CCB at 84.

In the early 1990's, Evanston and Lakeland (Highland Park’s parent), along with Children’s Memorial Hospital Center and Northwestern Memorial Hospital, formed the Northwestern Healthcare Network (“Network”). F. 35-39. Among the goals of the Network was to allow hospitals to come together to respond to anticipated marketplace behavior with respect to managed care contracting and exclusive contracting with certain managed care organizations. F. 40. The four hospital members entered into a Network Affiliation Agreement that provided for the creation of a council of governors that had control over the Network, including, inter alia, the authority to appoint and to remove members of the board of directors of the Network. F. 46. The Network negotiated contracts for the provision of hospital services by its member hospitals with the International Brotherhood of Teamsters, Health Network, Great West, and MultiPlan. F. 42. Respondent asserts that, because, under the Network Affiliation Affiliation Agreement, the Network became the “sole member” of the member hospitals, in accordance with the Illinois General Not For Profit Corporation Act of 1986, as amended, Evanston and EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 329 Initial Decision Highland Park were no longer two separate “persons,” as that term is term is defined by the Clayton Act. RB at 111. A review of the evidence, however, demonstrates that under the Network Affiliation Affiliation Agreement, Evanston and Highland Park remained separate economic entities.

Under the Network Affiliation Agreement, the governing boards of each of the hospitals retained “local autonomy and control,” of their own hospitals. F. 48. Each institution developed its own budget and operated independently. F. 49. Members of the Network only shared the cost of running the Network. F. 68. There was no combined profit and loss or profit-sharing. F. 68. Each hospital retained autonomy and control over the decisions related to the delivery of health care services at its hospital. F. 60. Each hospital maintained its own medical staff and retained the exclusive authority over granting medical staff privileges at its hospital. F. 55, 57. The Network could not terminate the employment of the administrators of the individual member hospitals, except for limited, specifically defined reasons. F. 52. Each hospital developed its own hospital program expansion plans. F. 61.

Each hospital also retained the authority to enter into a contract or to refuse to enter into a contract with each individual managed care organization. F. 65. The Network did not have the authority to enter into a contract binding on the individual member hospitals. F. 65. The hospitals that were members of the Network continued to compete with each other, unilaterally negotiating contracts with managed care companies, “‘slicing’ each other up in the market,” and “undercutting each other.” F. 66.

The evidence in this case, thus, demonstrates that Evanston and and Highland Park remained “separate economic actors pursuing separate economic interests,” and that their merger “suddenly [brought] together economic power that was previously pursuing divergent goals.” Copperweld, 467 U.S. at 769. Factors the Supreme VOLUME 144 Initial Decision Supreme Court considered in Copperweld in making its determination were whether a parent and its wholly owned subsidiary had “a complete unity of interests”; and whether “their general corporate actions [were] guided or determined not by two separate corporate consciousnesses, but one.” Id. The key to determining if two separate organizations actually constitute a “single entity” for assessing whether they are incapable of conspiring with each other in violation of Section 1 of the Sherman Act is assessment of “economic unity.” Freeman v. San Diego Assoc., 322 F.2d 1133, 1147-48 (9th Cir. 2002). “Where there is substantial common ownership, a fiduciary obligation to act for another’s economic benefit or an agreement to divide profits and losses, individual firms function as an economic unit and are generally treated as a single entity.” Id. at 1148. “[I]n the absence of economic unity, the fact that joint venturers pursue the common interests of the whole is generally not enough, by itself, to render them a single entity.” Id.

As summarized above, the hospitals in the Network did not have a fiduciary obligation to act for each other’s economic benefit or to divide profits and losses; they did not function as an economic unit, but rather, retained autonomy with respect to hospital administration, staff, delivery of health care services, budget, and expansion plans. F. 46-71. Further, unlike the Copperweld parent company, the NH Network could not “keep a tight rein” over the individual member hospitals because the NH Network could not “assert full control at any moment if the [member hospitals] fail[ed] to act in the [NH Network’s] best interests.” Copperweld, 467 U.S. at 771-72. In fact, managed care organizations testified that premerger, the competition between Highland Park and Evanston had allowed them to negotiate lower rates. F. 229-32.

Respondent also asserts that Evanston and Highland Park were were not separate persons, as required by Section 7 because the EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 331 Initial Decision parties were not required to file a Report and Notification Form (“HSR Form”) pursuant to the Hart-Scott-Rodino Antitrust Improvement Act of 1976, as amended (“HSR Act”). RB at 111-12. 111-12. Respondent asserts that, prior to the merger, the parties to the merger asked the staff of the FTC’s Premerger Notification Office whether they would be required to file an HSR Form, given given the fact that the Network served as the sole corporate member member of a number of hospitals and hospital holding companies, and that the parties to the proposed merger were nonprofit, tax exempt corporations. RB at 111-12 (citing FTC Premerger Notification Office Informal Staff Opinion No. 9908002 (August 10, 10, 1999), available at http://www.ftc.gov/bc/hsr/informal/opinions/9908002.htm). That the parties to the merger may not have been required to file a Report and Notification Form pursuant to the HSR Act does not change the conclusion that Evanston and Highland Park were separate “persons.” The Clayton Act makes clear that the administration of the HSR Act has no bearing on an FTC action brought under Section 7: “[a]ny action taken by the [FTC] . . . or any failure of the [FTC] . . . to take any action under [the HSR Act] shall not bar any proceeding or any action with respect to such acquisition at any time under any other section of this Act.” 15 U.S.C. § 18a(i). Further, Section 7 permits a merger challenge at “any time the acquisition threatens to ripen into a prohibited effect.” E.I. du Pont, 353 U.S. at 597. Thus, Complaint Counsel’s action is not barred. The mechanics of this merger and the dissolution of the NH Network further confirm that Evanston and Highland Park were not a single entity controlled by the NH Network. The NH Network did not direct the hospitals to merge; instead, Evanston and Highland Park independently agreed to merge and notified the NH Network afterward of their plans. NH Network members confirmed their independence when, in 1999, the member hospitals voted to dissolve the NH Network rather than submit themselves to the “full control” of the NH Network. F. 76.

VOLUME 144 Initial Decision The evidence conclusively shows that, under the Copperweld doctrine, Evanston and Highland Park were not already “one person” at the time of their merger. Therefore, Evanston’s merger with Highland Park is subject to Section 7 of the Clayton Act. E. Summary of Liability 1. Count I is Sustained Count I of the Complaint alleges that the merger of ENH and Highland Park has substantially lessened competition in the relevant market, in violation of Section 7 of the Clayton Act, as amended, 15 U.S.C. § 18. Complaint ¶ 27. Specifically, the Complaint alleges that as a result of the merger, ENH has been able “to exercise market power in the relevant market.” Complaint ¶ 18. The Complaint asserts that “ENH negotiated uniform prices for the three hospitals as a single system and raised prices at all three locations” and that the “price increases that resulted from the merger are large and far beyond those achieved by comparable hospitals during this time period.” Complaint ¶¶ 1, 24. Count I further alleges that the market created by the merger is “highly concentrated” as measured by the Herfindahl-Hirschman Index. Complaint ¶ 18. As explained above, the evidence demonstrates that the relevant relevant product market is general acute care inpatient services sold sold to managed care organizations, including primary, secondary, and tertiary inpatient services. The evidence further demonstrates that the following seven hospitals are properly included in the relevant geographic market: Evanston, Glenbrook, Highland Park, Lake Forest, Advocate Lutheran General, Rush North Shore, and St. St. Francis. The concentration level under the post-merger Herfindahl-Hirschman Index for the relevant market is 2739 with an an increase of 384, which corresponds to a “highly concentrated” market and the presumption that the merger is likely to “create or EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 333 Initial Decision enhance market power.” Merger Guidelines § 1.51. This prediction prediction is confirmed by direct evidence that ENH exercised its enhanced post-merger market power through elimination of a competitor and obtained post-merger price increases significantly above its premerger prices and substantially larger than price increases obtained by other comparison hospitals. Neither Respondent’s learning about demand theory nor quality of care improvements justify the substantial price increases to managed care care organizations and consumers. Respondent’s other defenses are are similarly unpersuasive. The only viable explanation for Respondent’s higher prices is that the merger gave ENH enhanced market power.

Complaint Counsel has thus demonstrated a reasonable probability that the structure of the merger will create an appreciable danger of anticompetitive consequences and will substantially lessen competition and harm consumer welfare in the future. Accordingly, as Complaint Counsel has established a violation of Section 7 of the Clayton Act, Count I is SUSTAINED.

2. Count II is Dismissed as Moot Count II also charges that the merger of ENH and Highland Park has substantially lessened competition, in violation of Section 7 of the Clayton Act, but does not allege a relevant product or geographic market. See Complaint ¶¶ 28-32 (the paragraphs alleging the relevant product and geographic markets in Count I, paragraphs 16- 18, are not incorporated by reference into Count II). Complaint Counsel argues that Counts I and II are alternative approaches to establishing a violation of Section 7 of the Clayton Act. CCB at 51; Closing argument, Tr. 6546-47.

In light of the Court’s finding of Respondent’s liability under Count I, it is unnecessary to reach the government’s Count II claim. claim. See Brown v. McCormick, 87 F. Supp. 2d 467, 481 (D. Md. 2000); Mitchell v. Penton/Industrial Publishing Co., Inc., 486 F. VOLUME 144 Initial Decision Supp. 22, 26 (N.D. Oh. 1979). As Count II is not dispositive of the the issues presented, it is moot.

Assuming arguendo, that the merits of Count II were still in issue, Complaint Counsel’s direct effects theory of liability does not, in any event, allow it to forgo its burden of proving the relevant market under a Clayton 7 claim. As the Seventh Circuit noted in Republic Tobacco, neither Toys “R” Us, Inc. v. FTC, 221 F.3d 928 (7th Cir. 2000) nor Indiana Federation of Dentists, 476 U.S. at 447 (cited by Complaint Counsel), allows an antitrust plaintiff to dispense entirely with market definition. 381 F.3d at 737. The antitrust plaintiff must show at least the rough contours of a relevant market. Id. Only upon such a showing and additional proof that the defendant commands a substantial share of the market can “direct evidence of anticompetitive effects . . . establish the defendant’s market power – in lieu of the usual showing of a precisely defined relevant market and a monopoly market share.” Id. Complaint Counsel’s reading of Rockford Memorial, 898 F.2d at 1282-83, regarding the “convergence” of the Sherman and Clayton Act enforcement schemes is unpersuasive and does not overcome the Seventh Circuit’s subsequent holding in Republic Tobacco. Thus, while Indiana Federation of Dentists, 476 U.S. at 460-61, concluded that if there is direct evidence of anticompetitive effects, “elaborate market analysis” is not required, it does not stand for the proposition urged by the government that “it is unnecessary to define a product or geographic market for the purposes of a claim under section 7 of the Clayton Act.” Complaint Counsel Interrog. Answers at 33. In Count II, by not alleging a relevant product or geographic market, Complaint Counsel asks the Court to adopt a novel theory of of Clayton 7 liability. To do so would undermine decades of established merger jurisprudence – a departure that this Court is unwilling to undertake. The Court’s previous Order denying EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 335 Initial Decision Respondent’s Motion to Dismiss Count II is entirely consistent with with the language of Section 7, the case law discussed herein, and the Merger Guidelines – all of which require Complaint Counsel to to carry its burden of defining the relevant market. Complaint Counsel’s interpretation of Section 7 thus fails as a matter of law. Accordingly, for the above-stated reasons, Count II is DISMISSED.

F. Remedy 1. Applicable Standards The effect of the acquisition of Highland Park by ENH has been to substantially lessen competition in violation of Section 7 of the Clayton Act, as amended. Once a violation is found, the Commission has an obligation to order effective relief to protect the public from further violations. The antitrust laws traditionally have favored divestiture to remedy an illegal merger’s competitive concerns. Much of the case law has followed this rationale and found divestiture “the most appropriate means for restoring competition lost as a consequence of a merger or acquisition.” Chicago Bridge & Iron, Dkt No. 9300, at 94.

Section 11(b) of the Clayton Act states that the Commission “shall” order a divestiture of “the stock, or other share capital, or assets held” in violation of Section 7. 15 U.S.C. § 21(b). Through Section 11 of the Clayton Act, Congress expressly directed the Commission to issue orders requiring the violator of Section 7 to divest itself of the assets held in violation of the Clayton Act. California v. American Stores, Co., 495 U.S. 271, 284-85 and n.11 (1990); FTC v. Western Meat Co., 272 U.S. 554, 559 (1926). Supreme Court precedent holds that divestiture is the usual and and proper remedy where a violation of Section 7 has been found. United States v. E.I. du Pont de Nemours & Co., 366 U.S. 316, 329, 329, 331 (1961) (ruling that an undoing of the acquisition is “a VOLUME 144 Initial Decision natural remedy,” and “should always be in the forefront of a court’s court’s mind when a violation of § 7 has been found.”). It is “well settled that once the Government has successfully borne the considerable burden of establishing a violation of law, all doubts as as to the remedy are to be resolved in its favor.” Id. at 334. In E.I. du du Pont, the Supreme Court acknowledged the drastic nature of the the divestiture remedy, but held that it is the “most effective” means means to restore premerger levels of competition. Id. at 326; In re RSR Corp., 88 F.T.C. 800, 894 (Dec. 2, 1976), aff’d, 602 F.2d 1317 1317 (9th Cir. 1979).

In Ford Motor Co. v. United States, the Supreme Court held that Section 7 relief must be directed to that which is “necessary and appropriate in the public interest to eliminate the effects of the acquisition offensive to the statute’ . . . or which will 'cure the ill effects of the illegal conduct, and assure the public freedom from its continuance.’” 405 U.S. 562, 573 n.8 (1972) (citations omitted); see also American Stores, 495 U.S. at 285 n.11 (A person who is allowed to continue holding ownership over stock or assets that created a Section 7 violation would be engaging in a perpetual violation, thus divestiture is the only effective remedy.). As such, the relief must not be punitive but must be designed to “redress the violations” and “to restore competition.” Id. at 573. Cases cite the well-established standard that the Commission’s remedy is proper as long as there is a “reasonable relationship between the remedy and the unlawful conduct at issue.” Atlantic Refining Co. v. FTC, 381 U.S. 357, 377 (1965); FTC v. Ruberoid Co., 343 U.S. 470, 473 (1952); FTC v. National Lead Co., 352 U.S. 419, 428 (1957). Respondent asserts that any consideration of the proposed divestiture order must begin with the premise that “[d]ivestiture is itself an equitable remedy designed to protect the public interest.” E.I. du Pont, 366 U.S. at 326. As such, “in the case of a judicial determination that an acquisition was in violation of Section 7, a claim of hardship attendant upon complete divestiture can be EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 337 Initial Decision considered in determining the appropriate remedy for the redress of of antitrust violations where something short of divestiture will effectively redress the violation.” United States v. Intl Tel. & Tel. Corp., 349 F. Supp 22, 31 (D. Conn. 1972). Thus, “while divestiture is normally the appropriate remedy in a Section 7 proceeding, on occasion, it may possibly be impracticable or inadequate . . . which underscores the importance of the Commission’s having a range of alternatives in its arsenal of remedies.” In re Ekco Prod. Co., 65 F.T.C. 1163, 1217 (June 30, 1964), aff’d, 347 F.2d 745 (7th Cir. 1965). As noted in In re Retail Credit Co., 92 F.T.C. 1, 123 (July 7, 1978) (“[t]his is not to say that divestiture is an automatic sanction, mechanically invoked in merger cases.”). Similarly, in In re National Tea Co., 69 F.T.C. 226 (Mar. 4, 1966), the Commission stated, “[a]t least we think it appropriate, in the circumstances of this case, to give those natural forces of competition a chance to correct the imbalances in those markets before turning to the more stringent remedy of divestiture.” Id. at 278.

2. Divestiture is the Appropriate Remedy In addressing the issue of appropriate relief in this case, the Court is guided by the basic principle set forth by the Commission in In re Fruehauf Corp., 90 F.T.C. 891, 892 n.1 (Dec. 21, 1977), that “the burden rests with respondent to demonstrate that a remedy other than full divestiture would adequately redress any violation which is found.” Such an exception to the general rule favoring divestiture can be invoked, however, “only when the proof of their probable efficacy is clear and convincing. In the absence of proof to the contrary the assumption of this Commission must be that ‘only divestiture can reasonably be expected to restore competition and make the affected markets whole again.’” Diamond Alkali, 72 F.T.C. at 742 (quoting National Tea Co. 69 F.T.C. at 277). VOLUME 144 Initial Decision Upon review of the record, Respondent has failed to meet its burden by identifying any hardship which would entitle it to an exception to the divestiture rule. Nor has Respondent persuaded the Court that any alternative remedies to divestiture would effectively “redress the violation” found herein. The Commission has noted that the purpose of Section 7 relief is to “undo the probable anticompetitive effects of the unlawful merger, to restore competition to the state in which it existed at the time of the merger, or to the state in which it would be existing at the time the relief is ordered.” Retail Credit Co., 92 F.T.C. at 161. It is against this standard that Respondent’s proposed alternative remedies must be considered and assessed.

First, Respondent proposes imposition of a “prior notice” order which would obligate ENH to notify the Commission over a period of five years, before it made any future acquisitions of providers of general acute care inpatient hospital services in the relevant geographic market. See Respondent’s Proposed Order A. Such a remedy, Respondent argues, would be reasonably related to the transaction by insuring that any non-reportable acquisition of inpatient services in the relevant market that ENH may pursue in the future would be reviewed by Commission staff prior to consummation. Such a remedy, Respondent asserts, while acknowledging a past violation of Section 7, would not, given what Respondent argues to be an absence of evidence of any present or future anticompetitive effects, interfere with “present competitive market conditions,” nor require any action that would destroy the quality improvements that are currently benefitting consumers. RB at 125.

Such relief, however, fails to speak to the present competitive market conditions that have given rise to the Section 7 violation in in this case. Respondent cannot demonstrate how such behavioral relief will “undo the . . . [present] anti-competitive effects of the unlawful merger to restore competition” to the levels prior to the acquisition. Retail Credit, 92 F.T.C. at 161. The proposed relief EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 339 Initial Decision further ignores the significant post-merger price increases and the evidence that any post-merger quality improvements are outweighed outweighed by the anticompetitive effects generated by the illegal acquisition of Highland Park. Respondent’s alternative remedy therefore fails to redress the violation found and fails to “make the affected markets whole again.” See Diamond Alkali, 72 F.T.C. at 742.

Respondent’s second alternative, that the Court enter a “narrowly crafted conduct remedy” requiring Evanston and Highland Park to negotiate and maintain separate managed care contracts, is similarly unpersuasive. RB at 125-26. In the absence of structural relief, given the geographic dynamics of the relevant market, the Court is not persuaded that the “natural forces of competition” will be able to adequately redress the anticompetitive imbalances that currently exist as a result of the ENH merger with Highland Park. Thus, Respondent’s alternative remedy, of allowing the managed care organizations to select specific pricing methodologies in bidding ENH’s inpatient service contracts, would not effectively restore competition to the premerger landscape. Respondent’s proposed remedy fails to demonstrate how such practices would restore competition in the relevant market. The ill effects emanating from the ENH merger are not amenable to short term, transitory cures. The Commission must therefore have leeway to devise effective relief to restore the relevant market’s pretransaction competitive balance.

It has not been shown that non-structural relief could effectively effectively redress the violations at issue in this case. Nevertheless, Nevertheless, Respondent asserts several specific reasons why divestiture would not be the most appropriate remedy to protect the the public interest. Respondent argues that divestiture would threaten a number of quality improvements and services achieved as as a result of the merger. RB at 116-23. The argument that the VOLUME 144 Initial Decision Highland Park community would ultimately be harmed as a result of of divestiture, however, is without merit, both legally and in fact. As a matter of law, the Court’s evaluation of the competitive effects of this merger has determined that, on balance, anticompetitive harm has occurred as a consequence of this transaction, despite procompetitive benefits that resulted. Upon such a finding, divestiture, on balance, could not be deemed to harm consumers as it would eliminate the anticompetitive harm that has been found to exceed any quality benefits. As noted earlier, the evidence demonstrates that most quality of care improvements at Highland Park were not merger specific and will not be lost upon divestiture. F. 869-975. In addition, as discussed below, the evidence does not demonstrate that any quality benefits will be significantly diminished as a result of divestiture.

Respondent asserts that divestiture will harm the community by slowing the rate of improvements in Highland Park’s quality of care in the future and by eliminating: improvements already achieved; the benefits of the academic affiliation and clinical integration ENH brings to Highland Park; the leadership structure and collaborative culture; and several important services such as cardiac surgery, interventional cardiology, and EPIC. RB at 116-20. It is true, as discussed below, that some benefits of the merger will be lost, including the current electronic medical records system, EPIC; academic affiliation and clinical integration; and cardiac surgery. The evidence demonstrates, however, that these benefits are insubstantial in relation to the anticompetitive harm resulting from the merger.

Upon divestiture, Highland Park will need to determine how it wishes to maintain its medical records. Highland Park will need to invest in a records management system, through EPIC or another vendor. Highland Park may pursue a license from EPIC, although even if Highland Park created its own EPIC system, the benefits of of having records from multiple hospitals and some physician EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 341 Initial Decision offices would be lost. F. 976-87. There is no evidence, however, from which to quantify the loss of value that would result from Highland Park’s choice of medical records systems. This is merely merely one of many decisions that will need to be made by Highland Highland Park as it transitions into either a stand alone hospital or joins another hospital system.

To the extent that Highland Park physicians participated in teaching residents and benefitted from the affiliation with Evanston, those benefits will also be lost. F. 988-93. However, Highland Park physicians will continue to be able to improve their abilities through professional development activities at Evanston or other venues. Most Highland Park physicians were excellent before the merger and the Court is confident they will remain so after the merger. In addition, Highland Park would not be able to continue the cardiac surgery program on its own. However, Highland Park could continue cardiac surgery as a joint venture with Evanston, similar to the joint cardiac surgery programs that Evanston has with Swedish Covenant and Louis A. Weiss. F. 957. Or, Highland Park could seek a different partner for its cardiac surgery program. F. 955-56. Even if Highland Park closes the cardiac surgery program, it could still continue to provide interventional cardiac procedures. F. 964. The record thus establishes that Highland Park, upon divestiture, divestiture, has the ability to maintain or establish acceptable levels levels of quality care in most service areas, including the collaborative and multi-disciplinary culture. As to intensivist coverage, Highland Park simply needs to maintain the contract that that it has already in place to provide those services. F. 974-75. There is no non-financial reason not to do that on a stand alone basis. The same is true for changes in the emergency department, heart attack care, cancer care, and critical pathways. F. 888-959. Similarly, most of the changes in obstetrics and gynecology, nursing, nursing, quality assurance, quality improvements, physical plant, VOLUME 144 Initial Decision laboratory medicine and pathology services, pharmacy services, and and radiology and radiation medicine could be maintained in the event of divestiture. F. 876-975. The changes relating to physical plant, lab, and Pyxis, have already been made at Highland Park and and would remain upon divestiture. F. 911-20, 941-947. The quality quality improvement system could also remain in place at Highland Highland Park because physicians and nurses are familiar with it. F. F. 896-902. Adult psychiatric services could be added to Highland Park or referred to another hospital. F. 965-69. Thus, Highland Park Park would likely continue post-merger organizational, clinical, and and cultural changes and implement nearly any quality improvements it deems beneficial.

“In section 7 cases, the principal purpose of relief is to restore competition to the state in which it existed prior to, and would have continued to exist, but for the illegal merger.” In re B.F. Goodrich Co., 110 F.T.C. 207, 345 (March 15, 1988) (emphasis supplied). The evidence demonstrates that only full divestiture of Highland Park can be expected to effectively restore competition in the market. Various managed care organization witnesses affirm this conclusion, having testified that Highland Park, as an independent, stand alone, premerger entity, gave them a valuable alternative with which to restrain Evanston’s prices. F. 229-32. Restoration of the competitive landscape that existed before the merger would thus likely prevent Evanston from predicating any anticompetitive pricing based on its post-merger knowledge of demand and pricing for its services. The record does not therefore indicate that divestiture would significantly harm consumers by eliminating the enumerated postmerger improvements at Highland Park.

The Commission has ordered divestiture of integrated assets in in consummated merger cases numerous times where violations of the Clayton Act have been found. E.g., Chicago Bridge & Iron, Dkt. Dkt. 9300, at 92; In re Olin Corp., 113 F.T.C. 400, 618-19 (June 13, 13, 1990); In re Crown Zellerbach Corp., 54 F.T.C. 769, 808 (Dec. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 343 Initial Decision (Dec. 26, 1957), aff’d, 296 F. 2d 800 (9th Cir. 1961); Ekco Products, Products, 65 F.T.C. at 1228-30. In the instant case, Respondent has has not presented sufficient evidence to depart from the usual and customary remedy of divestiture. As such, upon consideration of the the entire record in this case, divestiture is the most effective and appropriate remedy to restore competition and is hereby ordered. The attached Order is designed to unwind the merger and remedy the anticompetitive effects arising from the unlawful transaction. 3. Relief Courts have given significant deference to the Commission’s expertise fashioning such relief because, as the Supreme Court noted in Ruberoid, “Congress expected the Commission to exercise a special competence in formulating remedies to deal with problems in the general sphere of competitive practices.” 343 U.S. at 473. Similarly, in Hospital Corporation of America, the Seventh Circuit noted the Commission’s “broad discretion, akin to that of a court of equity, in deciding what relief is necessary to cure a violation of law and ensure against its repetition.” 807 F.2d at 1393. The Commission has wide discretion in determining what type of order is necessary to remedy the unfair practices found. Jacob Siegel Co. v. FTC, 327 U.S. 608, 611 (1946); National Lead Co., 352 U.S. at 428. “The relief which can be afforded” from an illegal illegal acquisition “is not limited to the restoration of the status quo quo ante” but may include that “which is necessary and appropriate appropriate in the public interest.” Ford Motor Co., 405 U.S. at 573 573 n.8. Thus, in addition to fashioning an appropriate divestiture remedy, the Commission also has authority to order ancillary relief. relief. Ancillary relief is ordered here, in order to: (1) correct any informational and bargaining imbalance that may exist between Respondent and the prospective acquirer of the divested assets; (2) (2) enhance and expand the competitive viability of the buyer; and and (3) reduce any adverse incentives of Respondent, which may put put the divested business at risk. See Federal Trade Commu, A VOLUME 144 Initial Decision Study of the Commission’s Divestiture Process (1999) (available at at www.ftc.gov/os/1999/08 /divestiture.pdf). A few of the provisions sought by Complaint Counsel are not “necessary and appropriate in the public interest,” as required by Ford Motor Company. 405 U.S. at 573 n.8. Deferring to the Commission’s expertise in fashioning effective relief, the Proposed Order submitted by Complaint Counsel is herein adopted, except as noted below. Provisions found to be beyond the relief necessary to cure the violation or unnecessary are:

- The proposed requirement that Respondent take actions necessary to assist the Acquirer in ensuring the provision or continuation of a cardiac surgery program at Highland Park Hospital that is capable of providing an equivalent standard of care in substantially the same manner as the cardiac surgery program established at Highland Park after the merger. Proposed Order, II.E.

- The proposed language relating to the purpose of the divestiture and the factors the Commission will consider. Proposed Order, II.L.

- The proposed requirement that Respondent vest pension benefits and provide any ENH Employee to whom the Acquirer has made a written offer of employment with reasonable financial incentives to accept a position with the Acquirer. Proposed Order, II.H.3.

- The proposed indemnification clauses, for holding harmless both the Monitor and the EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 345 Initial Decision Divestiture Trustee. Proposed Order, V.B.6, VI.C.7.

Accordingly, such provisions are not adopted and shall not be ordered. In addition, slight modifications from the language proposed by Complaint Counsel are made within the following paragraphs of the Proposed Order: I.H, I.K, I.X, I.Z, I.AA, II.A, II.D, and VI.C.5.

IV. SUMMARY OF CONCLUSIONS OF LAW 1. Respondent Evanston Northwestern Healthcare (“ENH”) is a nonprofit corporation organized, existing, and doing business under the laws of the state of Illinois.

2. In the challenged merger, consummated on January 1, 2000, ENH acquired the assets of Highland Park Hospital (“Highland Park”).

3. Section 7 of the Clayton Act applies to asset acquisitions by nonprofit hospitals.

4. The Federal Trade Commission (“FTC”) has jurisdiction over Respondent and the subject matter of this proceeding pursuant to Sections 7 and 11 of the Clayton Act, 15 U.S.C. §§ 18, 21(b). 5. Complaint Counsel bears the burden of proof of establishing each element of the violations alleged in the Complaint by a preponderance of the evidence.

6. Section 7 of the Clayton Act prohibits any acquisition of stock or assets “where in any line of commerce . . . in any section of the country, the effect of such acquisition may be substantially to lessen competition or to tend to create a monopoly.” 15 U.S.C. § 18. VOLUME 144 Initial Decision 7. The appropriate line of commerce (relevant product market) within which to evaluate the probable anticompetitive effects of the merger is general acute care inpatient services sold to managed care organizations, which includes primary, secondary, and tertiary inpatient services.

8. The appropriate section of the country (relevant geographic market) within which to evaluate the probable anticompetitive effects of the merger is the area encompassing the following seven hospitals: Evanston, Glenbrook, Highland Park, Lake Forest, Advocate Lutheran General, Rush North Shore, and St. Francis. 9. Section 7 of the Clayton Act is designed to arrest in its incipiency the substantial lessening of competition from the acquisition by one corporation of the assets of a competing corporation. Complaint Counsel must show a reasonable probability that the transaction would substantially lessen competition in the future.

10. Complaint Counsel must establish a prima facie cage that the acquisition is unlawful. A prima facie case may be made by showing that the transaction will significantly increase market concentration and by introducing other types of evidence relating to market conditions.

11. Market concentration under the Merger Guidelines is measured by the Herfindahl-Hirschman Index (“HHI”). Under the Merger Guidelines, a market in which the post-merger HHI is above 1800 is considered “highly concentrated,” and a merger in a highly concentrated market that increases the market’s HHI by over 100 is presumed to be “likely to create or enhance market power or facilitate its exercise.”

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 347 Initial Decision 12. In the relevant geographic market determined by the Court, the merger results in an HHI of 2739, with an increase of 384 from premerger levels.

13. The post-merger HHI of 2739 is substantially above the Merger Guidelines’ threshold of 1800 to consider a market “highly concentrated,” and the increase of over 384 far exceeds the Merger Guidelines’ threshold of 100 to presume that the merger is “likely to create or enhance market power or facilitate its exercise.” 14. In the relevant geographic market, in 1999, Evanston and Highland Park had a combined market share of approximately 35%. ENH’s post-merger market share increased to approximately 40% by 2002, with the other four hospitals in the geographic market all losing market shares from 1999 to 2002.

15. The post-merger market share presents the threat of undue concentration.

16. Complaint Counsel established a prima facie case by demonstrating sufficient market concentration to predict probable anticompetitive effects.

17. These predictions of probable anticompetitive effects are confirmed by Complaint Counsel’s demonstration that ENH exercised its enhanced post-merger market power and obtained postmerger price increases substantially above its premerger prices and significantly larger than price increases obtained by other comparison hospitals.

18. Complaint Counsel established that the price increases were achieved as a result of market power and not because of learning about demand or improvements in quality of care. 19. Respondent’s learning about demand theory cannot explain the post-merger price increases at ENH.

VOLUME 144 Initial Decision 20. The vast majority of the quality of care improvements made by ENH to Highland Park were not merger specific. Two quality of care improvements to Highland Park were merger specific, but they do not justify ENH’s increased post-merger prices or outweigh the probable anticompetitive effects of the merger. 21. The evidence demonstrates that entry or expansion from existing hospitals is not likely to replace competition lost through the acquisition or to sufficiently constrain ENH from future anticompetitive actions.

22. The evidence demonstrates that the nonprofit status of ENH has not operated to constrain ENH’s exercise of market power. 23. The evidence demonstrates that Highland Park was able to meet its financial obligations in the near future; was not in danger of bankruptcy; had other options besides merging with ENH; and was not in danger of exiting the market in the foreseeable future. 24. Respondent did not rebut the presumption of a violation of Section 7 of the Clayton Act.

25. The merger is subject to Section 7 of the Clayton Act because Evanston and Highland Park were not already “one person” at the time of the merger.

26. The merger violates Section 7 of the Clayton Act because “the effect of such acquisition may be substantially to lessen competition or to tend to create a monopoly.” 15 U.S.C. § 18. 27. Complaint Counsel met its burden of proof in support of Count I of the Complaint because, in a line of commerce, in an activity affecting commerce in a section of the country, the effect of EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 349 Initial Decision of the merger may be substantially to lessen competition. Count I is is therefore SUSTAINED.

28. In light of the Court’s finding of liability under Count I, it is unnecessary to reach Count II, as it is not dispositive of the issues presented and is thus moot. Count II is therefore DISMISSED. 29. Divestiture is the most effective and appropriate remedy to address the violation in this case.

30. Complete divestiture of all Highland Park assets acquired in the merger is required to restore competition as it existed prior to the merger.

31. Relief designed to restore competition as it existed prior to the merger is appropriate.

32. The Order entered hereinafter is necessary and appropriate to remedy the violation of law found to exist. ORDER I.

IT IS HEREBY ORDERED that, as used in this Order, the following definitions shall apply:

A. “Acquirer” means any Person approved by the Commission to acquire the Highland Park Hospital Assets pursuant to this Order.

B. “Acquirer Hospital Business” means all activities relating to general acute care inpatient hospital services and other related health care services to be conducted by the Acquirer in connection with the Highland Park Hospital Assets. VOLUME 144 Initial Decision C. “Acute Care Hospital” means a health care facility licensed as a hospital, other than a federally-owned facility, having a duly organized governing body with overall administrative and professional responsibility, and an organized professional staff, that provides 24-hour inpatient care, that may also provide outpatient services, and having as a primary function the provision of General Acute Care Inpatient Hospital Services.

D. “Commission” means the Federal Trade Commission. E. “Direct Cost” means the cost of direct material and direct labor used to provide the relevant assistance or service. F. “Divestiture Agreement” means any agreement between Respondent (or between a Divestiture Trustee appointed pursuant to Paragraph VI of this Order) and an Acquirer approved by the Commission, and all amendments, exhibits, attachments, agreements, and schedules thereto that have been approved by the Commission, to accomplish the purpose and requirements of this Order.

G. “Divestiture Trustee” means the Person appointed pursuant to Paragraph VI of this Order.

H. “ENH” means Evanston Northwestern Healthcare Corporation, its directors, officers, employees, agents, attorneys, representatives, successors, and assigns; its subsidiaries, divisions, joint ventures, groups, and affiliates affiliates controlled by ENH (including, but not limited to, ENH Faculty Practice Associates and ENH Medical Group, Group, Inc.), and the respective directors, officers, employees, agents, attorneys, representatives, successors, and assigns of each. ENH Faculty Practice Associates is an an Illinois non-profit corporation that, inter alia, employs physicians who primarily serve the patients of ENH, and is EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 351 Initial Decision is the sole shareholder of ENH Medical Group, Inc., an Illinois for-profit corporation.

I. “ENH Contractor” means any Person that provides physician or other health care services pursuant to a contract with ENH (including, but not limited to, the provision of emergency room, anesthesiology, pathology, or radiology services) in connection with the operation of the Post-Merger Hospital Business at Highland Park Hospital.

J. “ENH Employee” means any Person employed by ENH in the operation of the Post- Merger Hospital Business, including, but not limited to, any physician employed by ENH Faculty Practice Associates.

K. “ENH License” means: (i) a worldwide, royalty-free, paidup, perpetual, irrevocable, transferable, sublicensable, nonexclusive license to all Intellectual Property owned by or licensed to ENH relating to operation of the Post-Merger Hospital Business other than the HPH Name and Marks, which shall be divested, assigned and conveyed to the Acquirer on a permanent and exclusive basis, to the extent allowable under the existing ENH License, and (ii) such tangible embodiments of the licensed rights (including but not limited to physical and electronic copies) as may be necessary or appropriate to enable the Acquirer to utilize the rights.

L. “ENH Medical Staff Member” means any physician or other health care professional who: (1) is not an ENH Employee, and (2) is a member of the ENH medical staff, including, but not limited to, any ENH Contractor.

M. “General Acute Care Inpatient Hospital Services” means a broad cluster of basic medical and surgical diagnostic and treatment services for the medical diagnosis, treatment, and VOLUME 144 Initial Decision and care of physically injured or sick persons with short term term or episodic health problems or infirmities, that include include an overnight stay in the hospital by the patient. General Acute Care Inpatient Hospital Services include what what are commonly classified in the industry as primary, secondary, and tertiary services, but exclude: (i) services at at hospitals that serve solely military and veterans, (ii) services at outpatient facilities that provide same-day service service only, (iii) those specialized services known in the industry as quaternary services, and (iv) psychiatric, substance abuse, and rehabilitation services. N. “Highland Park Hospital” means the Acute Care Hospital located at 718 Glenview Avenue, Highland Park, Illinois 60035.

O. “Highland Park Hospital Assets” means all of ENH’s right, title, and interest in and to Highland Park Hospital and all related healthcare and other assets, tangible or intangible, business, and properties, including any improvements or additions thereto made subsequent to the Merger, relating to the operation of the Post-Merger Hospital Business in Highland Park, Illinois, including, but not limited to: 1. All real property interests (including fee simple interests and real property leasehold interests), whether or not located on the Highland Park Hospital campus; 2. All personal property, including equipment and machinery;

3. All inventories, stores, and supplies; EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 353 Initial Decision 4. All rights under any contracts and agreements (e.g., leases, service agreements such as dietary and housekeeping services, supply agreements, procurement contracts), including, but not limited to, all rights to contributions, funds and other provisions for the benefit of Highland Park Hospital pursuant to the Foundation Agreement dated December 16, 1999, between ENH and Highland Park Hospital Foundation (“Foundation Agreement”);

5. All rights and title in and to use of the HPH Name and Marks on a permanent and exclusive basis (even as to ENH), and an ENH License to all other Intellectual Property (“Licensed Intellectual Property”); provided, however, that ENH may retain a worldwide, royalty-free, paid-up, perpetual, irrevocable, transferable, sublicensable, non-exclusive license to the Licensed Intellectual Property; provided further, however, that ENH shall retain no rights to use the HPH Name and Marks;

6. All governmental approvals, consents, licenses, permits, waivers, or other authorizations;

7. All rights under warranties and guarantees, express or implied;

8. All items of prepaid expense; and 9. All books, records, and files (electronic and hard copy). Provided, however, that the Highland Park Hospital Assets shall not include assets not located exclusively in Highland Park, Illinois, whose use is shared with or among other ENH Acute Care Hospitals.

VOLUME 144 Initial Decision P. “Hospital Provider Contract” means a contract between a Payor and any hospital to provide General Acute Care Inpatient Hospital Services and related healthcare services to enrollees of health plans.

Q. “HPH Name and Marks” means the name “Highland Park Hospital” and “HPH,” and any variation of these names, in connection with the Highland Park Hospital Assets, and all other associated trade names, business names, proprietary names, registered and unregistered trademarks, service marks and applications, domain names, trade dress, copyrights, copyright registrations and applications, in both published works and unpublished works, relating to the Highland Park Hospital Assets.

R. “Intellectual Property” means, without limitation: (i) the HPH Name and Marks; (ii) all copyrights, copyright registrations and applications, in both published works and unpublished works, other than those associated with the HPH Name and Marks; (iii) all patents, patent applications, applications, and inventions and discoveries that may be patentable; (iv) all know-how, trade secrets, software, technical information, data, registrations, applications for governmental approvals, inventions, processes, best practices (including clinical pathways), formulae, protocols, protocols, standards, methods, techniques, designs, quality control practices and information, research and test procedures and information, and safety, environmental and health practices and information; (v) all confidential or proprietary information, commercial information, management systems, business processes and practices, customer lists, customer information, customer records and and files, customer communications, procurement practices practices and information, supplier qualification and approval practices and information, training materials, sales sales and marketing materials, customer support materials, EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 355 Initial Decision advertising and promotional materials; and (vi) all rights in in any jurisdiction to limit the use or disclosure of any of the the foregoing, and rights to sue and recover damages or obtain injunctive relief for infringement, dilution, misappropriation, violation or breach of any of the foregoing.

S. “Merger” means the merger of Highland Park Hospital into ENH pursuant to the Agreement and Plan of Merger among Evanston Northwestern Healthcare Corporation, Lakeland Health Services, Inc., and Highland Park Hospital, dated as of October 29, 1999, which was consummated on or about January 1, 2000.

T. “Monitor” means the Person appointed pursuant to Paragraph V of this Order.

U. “Payor” means any Person that pays, or arranges for payment, for all or part of any General Acute Care Inpatient Hospital Services for itself or for any other Person. Payor includes any Person that develops, leases, or sells access to networks of Acute Care Hospitals.

V. “Person” means any individual, partnership, firm, corporation, association, trust, unincorporated organization or other entity or governmental body.

W. “Post-Merger Hospital Business” means all activities relating to the provision of General Acute Care Inpatient Hospital Services and other related health care services conducted by ENH after the Merger, including, but not limited to, all health care services, including outpatient services, offered at Highland Park Hospital. X. “Pre-Merger Highland Park Hospital Business” means all activities relating to the provision of General Acute Care Inpatient Hospital Services and other related healthcare VOLUME 144 Initial Decision services that Highland Park Hospital was offering prior to the Merger.

Y. “Respondent” means ENH.

Z. “Transitional Administrative Services” means administrative assistance with respect to the operation of an Acute Care Hospital and related health care services, including but not limited to assistance relating to billing, accounting, governmental regulation, human resources management, information systems, managed care contracting, and purchasing.

AA. “Transitional Clinical Services” means clinical assistance and support services with respect to operation of an Acute Care Hospital and related health care services, including but not limited to cardiac surgery, oncology services, and laboratory and pathology services.

BB. “Transitional Services” means Transitional Administrative Services and Transitional Clinical Services. II.

IT IS FURTHER ORDERED that:

A. No later than one hundred eighty (180) days from the date the divestiture requirements of this Order become final, Respondent shall divest and convey the Highland Park Hospital Assets at no minimum price, absolutely and in good faith, to an Acquirer that receives the prior approval of the Commission and in a manner (including an executed divestiture agreement) that receives the prior approval of the Commission. To the extent that:

1. The Highland Park Hospital Assets as of the date the divestiture requirements of this Order become final do EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 357 Initial Decision not include (i) assets that Respondent acquired on the date of the Merger, or (ii) assets that replaced those acquired on the date of the Merger, (iii) any other assets assets that Respondent acquired and has used in or that are related to the Post-Merger Hospital Business in Highland Park, Illinois, then Respondent shall add to the the Highland Park Hospital Assets additional assets (of a a quality that meets generally acceptable standards of performance) to replace the assets that no longer exist, are no longer controlled by Respondent, or are no longer longer located in Highland Park, Illinois; 2. After the Merger and prior to the date the divestiture requirements of this Order become final, Respondent terminated any clinical service, clinical program, support function, or management function (i) performed by the Pre-Merger Highland Park Hospital Business, or (ii) performed by the Post-Merger Hospital Business in Highland Park, Illinois, then Respondent shall restore such service, program, or function (of a quality that meets generally acceptable standards of care or performance), no later than the date the Highland Park Hospital Assets are divested, or any other date that receives the prior approval of the Commission. Provided, however, that Respondent shall not be required to to replace any asset or to restore any service, program or function contemplated by Paragraphs II.A.1 or II.A.2 of this this Order if, in each instance, Respondent can demonstrate demonstrate to the Commission that termination of such asset, service, program or function was for good cause or that the Acquirer does not need such asset, service, program program or function to effectively operate the Acquirer Hospital Business in a manner consistent with the purpose of of this Order, and the Commission approves the divestiture VOLUME 144 Initial Decision without the replacement or restoration of such asset, service, service, program or function.

B. Respondent shall comply with all terms of the Divestiture Agreement approved by the Commission pursuant to this Order, and any breach by Respondent of any term of the Divestiture Agreement shall constitute a violation of this Order.

C. Respondent shall cooperate with the Acquirer to ensure that the Highland Park Hospital Assets are transferred to the Acquirer as a financially and competitively viable Acute Care Hospital operating as an ongoing business, including but not limited to providing assistance necessary to transfer to the Acquirer all governmental approvals needed to operate the Highland Park Hospital Assets as an Acute Care Hospital.

D. No later than the date the Highland Park Hospital Assets are divested, to the extent allowable under the existing ENH Licenses, ENH shall grant to the Acquirer an ENH License to all Licensed Intellectual Property for any use in the Acquirer Hospital Business, and shall take all actions necessary to facilitate the unrestricted use of the Licensed Intellectual Property by the Acquirer.

E. Respondent shall take all actions necessary and shall effect all arrangements in connection with the divestiture of the Highland Park Hospital Assets as will ensure that the Acquirer can conduct the Acquirer Hospital Business in substantially the same manner as Respondent has conducted conducted the Post-Merger Hospital Business at Highland Park Hospital, with an independent full-service medical staff staff capable of providing General Acute Care Inpatient Hospital Services, and an independent full-service hospital staff and management, including, but not limited to, EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 359 Initial Decision providing Transitional Services, the opportunity to recruit and employ ENH Employees, and the opportunity to recruit, recruit, contract with, and extend medical staff privileges to to any ENH Medical Staff Member, including as provided in in Paragraphs II.F, II.G, and II.H of this Order. F. At the request of the Acquirer, for a period not to exceed twelve (12) months from the date Respondent divests the Highland Park Hospital Assets, except as otherwise approved by the Commission, and in a manner (including pursuant to an agreement) that receives the prior approval of the Commission:

1. Respondent shall provide Transitional Services to the Acquirer sufficient to enable the Acquirer to conduct the Acquirer Hospital Business in substantially the same manner that Respondent has conducted the Post-Merger Hospital Business at Highland Park Hospital; and 2. Respondent shall provide the Transitional Services required by this Paragraph II.F at substantially the same level and quality as such services are provided by Respondent in connection with its operation of the Post- Merger Hospital Business.

Provided, however, that Respondent shall not (i) require the the Acquirer to pay compensation for Transitional Services that exceeds the Direct Cost of providing such goods and services, (ii) terminate its obligation to provide Transitional Transitional Services because of a material breach by the Acquirer of any agreement to provide such assistance, in the the absence of a final order of a court of competent jurisdiction, or (iii) include a term in any agreement to provide Transitional Services that limits the type of damages damages (such as indirect, special, and consequential VOLUME 144 Initial Decision damages) that the Acquirer would be entitled to seek in the event of Respondent’s breach of such agreement. G. Respondent shall allow the Acquirer an opportunity to recruit and employ any ENH Employee in connection with the divestiture of the Highland Park Hospital Assets so as to enable the Acquirer to establish an independent, full-service medical staff, hospital staff and management, including as follows:

1. No later than six (6) weeks before execution of a divestiture agreement, Respondent shall (i) identify each ENH Employee, (ii) allow the Acquirer an opportunity to interview any ENH Employee, and (iii) allow the Acquirer to inspect the personnel files and other documentation relating to any ENH Employee, to the extent permissible under applicable laws. 2. Respondent shall (i) not offer any incentive to any ENH Employee to decline employment with the Acquirer, (ii) remove any contractual impediments with Respondent that may deter any ENH Employee from accepting employment with the Acquirer, including, but not limited to, any non-compete or confidentiality provisions of employment or other contracts with Respondent that would affect the ability of the ENH Employee to be employed by the Acquirer, and (iii) not otherwise interfere with the recruitment of any ENH Employee by the Acquirer, including, but not limited to, by refusing or threatening to refuse to extend medical staff privileges at any Respondent Acute Care Hospital.

3. For a period of two (2) years from the date the divestiture of the Highland Park Hospital Assets is completed, Respondent shall not, directly or indirectly, EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 361 Initial Decision hire or enter into any arrangement for the services of any any ENH Employee employed by the Acquirer, unless such ENH Employee’s employment has been terminated terminated by the Acquirer; provided, however, this Paragraph II.G.3 shall not prohibit Respondent from: (i) (i) advertising for employees in newspapers, trade publications, or other media not targeted specifically at the employees, or (ii) hiring employees who apply for employment with Respondent, as long as such employees were not solicited by Respondent in violation violation of this Paragraph II.G.3.

H. Respondent shall allow the Acquirer an unimpeded opportunity to recruit, contract with, and otherwise extend medical staff privileges to any ENH Medical Staff Member in connection with the divestiture of the Highland Park Hospital Assets so as to enable the Acquirer to establish an independent, complete, full-service medical staff, including as follows:

1. No later than the date of execution of a divestiture agreement, Respondent shall (i) identify each ENH Medical Staff Member, (ii) allow the Acquirer an opportunity to interview any ENH Medical Staff Member, and (iii) allow the Acquirer to inspect the files and other documentation relating to any ENH Medical Staff Member, to the extent permissible under applicable laws.

2. Respondent shall (i) not offer any incentive to any ENH ENH Medical Staff Member to decline to join the Acquirer’s medical staff, (ii) remove any contractual impediments with Respondent that may deter any ENH ENH Medical Staff Member from joining the Acquirer’s Acquirer’s medical staff, including, but not limited to, any non- compete or confidentiality provisions of employment or other contracts with Respondent that VOLUME 144 Initial Decision would affect the ability of the ENH Medical Staff Members to be recruited by the Acquirer, and (iii) not otherwise interfere with the recruitment of any ENH Medical Staff Member by the Acquirer, including, but not limited to, by refusing or threatening to refuse to extend medical staff privileges at any Respondent Acute Acute Care Hospital.

I. Except in the course of performing its obligations under this Order, Respondent shall:

1. not provide, disclose, or otherwise make available any trade secrets or any sensitive or proprietary commercial or financial information relating to the Acquirer or the Acquirer Hospital Business to any Person other than the Acquirer, and shall not use such information for any reason or purpose;

2. disclose trade secrets or any sensitive or proprietary commercial or financial information relating to the Acquirer or the Acquirer Hospital Business to any Person other than the Acquirer (i) only in the manner and to the extent necessary to satisfy its obligations under this Order and (ii) only to Persons who agree in writing to maintain the confidentiality of such information; 3. enforce the terms of this Paragraph II.I as to any Person and take such action as is necessary, including training, to cause each such Person to comply with the terms of this Paragraph II.I, including any actions that Respondent would take to protect its own trade secrets or sensitive or proprietary commercial or financial information.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 363 Initial Decision J. No later than ninety (90) days from the date the Highland Park Hospital Assets are divested, Respondent shall terminate any Hospital Provider Contract negotiated or amended after the Merger that is in effect as of the date the divestiture provisions of this Order become final; provided, however, that nothing in this Paragraph II.J. shall preclude Respondent (i) from completing any post-termination obligations relating to any Hospital Provider Contract or (ii) from entering into a new Hospital Provider Contract with any Payor after the current contract has been terminated. III.

IT IS FURTHER ORDERED that:

A. From the date this Order becomes final (without regard to the finality of the divestiture requirements herein) until the date the Highland Park Hospital Assets are divested pursuant to this Order, Respondent shall take such actions as are necessary to maintain the viability, marketability, and competitiveness of the Highland Park Hospital Assets and the Post-Merger Hospital Business relating to the Highland Park Hospital Assets. Among other things that may be necessary, Respondent shall:

1. Maintain the operations of the Post-Merger Hospital Business relating to the Highland Park Hospital Assets in the ordinary course of business and in accordance with past practice (including regular repair and maintenance of the Highland Park Hospital Assets).

2. Use best efforts to maintain and increase sales of the Post-Merger Hospital Business relating to the Highland Highland Park Hospital Assets, and to maintain at budgeted levels for the year 2005 or the current year, whichever are higher, for all administrative, technical, VOLUME 144 Initial Decision and marketing support for the Post-Merger Hospital Business relating to the Highland Park Hospital Assets. Assets.

3. Use best efforts to maintain the current workforce and to retain the services of employees and agents in connection with the Post-Merger Hospital Business relating to the Highland Park Hospital Assets, including payment of bonuses as necessary, and maintain the relations and good will with customers, suppliers, vendors, employees, landlords, creditors, agents, and others having business relationships with the Post- Merger Hospital Business relating to the Highland Park Hospital Assets.

4. Assure that Respondent’s employees with primary responsibility for managing and operating the Post- Merger Hospital Business relating to the Highland Park Hospital Assets are not transferred or reassigned to other areas within Respondent’s organization except for transfer bids initiated by employees pursuant to Respondent’s regular, established job posting policy. 5. Provide sufficient working capital to maintain the Post- Merger Hospital Business relating to the Highland Park Hospital Assets as an economically viable and competitive ongoing business and shall not, except as part of a divestiture approved by the Commission pursuant to this Order, remove, sell, lease, assign, transfer, license, pledge for collateral, or otherwise dispose of the Highland Park Hospital Assets. B. No later than forty five (45) days from the date this Order becomes final, Respondent shall file a verified written report EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 365 Initial Decision report to the Commission that identifies (i) all assets included in the Highland Park Hospital Assets, (ii) all assets assets originally acquired or that replace assets originally acquired by Respondent as a result of the Merger, (iii) all assets relating to the Post-Merger Hospital Business in Highland Park, Illinois, that are not included in the Highland Highland Park Hospital Assets, and (iv) all clinical services, services, support functions, and management functions that that ENH discontinued at Highland Park Hospital after the Merger (hereinafter “Accounting”).

IV.

IT IS FURTHER ORDERED that no later than ten (10) days from the date this Order becomes final (without regard to the finality of the divestiture requirements herein), Respondent shall provide a copy of this Order and Complaint to each of Respondent’s officers, employees, or agents having managerial responsibility for any of Respondent’s obligations under Paragraphs II and III of this Order. V.

IT IS FURTHER ORDERED that:

A. At any time after this Order becomes final (without regard to the finality of the divestiture requirements herein), the Commission may appoint a Person (“Monitor”) to monitor Respondent’s compliance with its obligations under this Order, consult with Commission staff, and report to the Commission regarding Respondent’s compliance with its obligations under this Order.

B. If a Monitor is appointed pursuant to Paragraph V.A of this Order, Respondent shall consent to the following terms and conditions regarding the powers, duties, authorities, and responsibilities of the Monitor:

VOLUME 144 Initial Decision 1. The Monitor shall have the power and authority to monitor Respondent’s compliance with the terms of this Order, and shall exercise such power and authority and carry out the duties and responsibilities of the Monitor pursuant to the terms of this Order and in a manner consistent with the purposes of this Order and in consultation with the Commission or its staff. 2. Within ten (10) days after appointment of the Monitor, Respondent shall execute an agreement that, subject to the approval of the Commission, confers on the Monitor all the rights and powers necessary to permit the Monitor to monitor Respondent’s compliance with the terms of this Order in a manner consistent with the purposes of this Order. If requested by Respondent, the Monitor shall sign a confidentiality agreement prohibiting the use, or disclosure to anyone other than the Commission (or any Person retained by the Monitor pursuant to Paragraph V.B.5 of this Order), of any competitively sensitive or proprietary information gained as a result of his or her role as Monitor, for any purpose other than performance of the Monitor’s duties under this Order. 3. The Monitor’s power and duties under this Paragraph V shall terminate three business days after the Monitor has completed his or her final report pursuant to Paragraph V.B.7(ii), or at such other time as directed by the Commission.

4. Respondent shall cooperate with any Monitor appointed appointed by the Commission in the performance of his his or her duties, and shall provide the Monitor with full full and complete access to Respondent’s books, records, records, documents, personnel, facilities and technical EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 367 Initial Decision information relating to compliance with this Order, or to to any other relevant information, as the Monitor may reasonably request. Respondent shall cooperate with any any reasonable request of the Monitor. Respondent shall shall take no action to interfere with or impede the Monitor’s ability to monitor Respondent’s compliance with this Order.

5. The Monitor shall serve, without bond or other security, at the expense of Respondent, on such reasonable and customary terms and conditions as the Commission may set. The Monitor shall have authority to employ, at the expense of Respondent, such consultants, accountants, attorneys and other representatives and assistants as are reasonably necessary to carry out the Monitor’s duties and responsibilities. The Monitor shall account for all expenses incurred, including fees for his or her services, subject to the approval of the Commission. 6. If at any time the Commission determines that the Monitor has ceased to act or failed to act diligently, or is unwilling or unable to continue to serve, the Commission may appoint a substitute to serve as Monitor in the same manner as provided by this Order.

7. The Monitor shall report in writing to the Commission (i) every sixty (60) days from the date this Order becomes final, (ii) no later than thirty (30) days from the date Respondent completes its obligations under this Order, and (iii) at any other time as requested by the staff of the Commission, concerning Respondent’s compliance with this Order.

C. Respondent shall submit the following reports to the Monitor: (i) no later than twenty (20) days after the date the the Monitor is appointed by the Commission pursuant to VOLUME 144 Initial Decision Paragraph V.A, a copy of the Accounting required by Paragraph III.B of this Order; and (ii) copies of all compliance reports filed with the Commission. D. The Commission may on its own initiative, or at the request of the Monitor, issue such additional orders or directions as may be necessary or appropriate to assure compliance with the requirements of this Order.

VI.

IT IS FURTHER ORDERED that:

A. If Respondent has not divested, absolutely and in good faith the Highland Park Hospital Assets within the time and manner required by Paragraph II.A of this Order, the Commission may at any time appoint one or more Persons as Divestiture Trustee to divest the Highland Park Hospital Assets, at no minimum price, in a manner that satisfies the requirements of this Order.

B. In the event that the Commission or the Attorney General brings an action pursuant to § 5(1) of the Federal Trade Commission Act, 15 U.S.C. § 45(1), or any other statute enforced by the Commission, Respondent shall consent to the appointment of a Divestiture Trustee in such action. Neither the appointment of a Divestiture Trustee nor a decision not to appoint a Divestiture Trustee under this Paragraph VI shall preclude the Commission or the Attorney General from seeking civil penalties or any other relief available to it, including appointment of a court-appointed Divestiture Trustee, pursuant to § 5(1) of the Federal Trade Commission Act, or any other statute enforced by the Commission, for any failure by the Respondent to comply with this Order.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 369 Initial Decision C. If a Divestiture Trustee is appointed by the Commission or a court pursuant to this Paragraph VI, Respondent shall consent to the following terms and conditions regarding the Divestiture Trustee’s powers, duties, authority, and responsibilities:

1. Subject to the prior approval of the Commission, the Divestiture Trustee shall have the exclusive power and authority to effect the divestiture for which he or she has been appointed pursuant to the terms of this Order and in a manner consistent with the purposes of this Order. 2. Within ten (10) days after appointment of the Divestiture Trustee, Respondent shall execute an agreement that, subject to the prior approval of the Commission and, in the case of a court-appointed Divestiture Trustee, of the court, transfers to the Divestiture Trustee all rights and powers necessary to permit the Divestiture Trustee to effect the divestiture for which he or she has been appointed.

3. The Divestiture Trustee shall have twelve (12) months from the date the Commission approves the agreement described in Paragraph VI.C.2 of this Order to accomplish the divestiture, which shall be subject to the prior approval of the Commission. If, however, at the end of the twelve-month period the Divestiture Trustee has submitted a plan of divestiture or believes that divestiture can be achieved within a reasonable time, the divestiture period may be extended by the Commission, or, in the case of a court appointed Divestiture Trustee, by the court.

4. Respondent shall provide the Divestiture Trustee with full and complete access to the personnel, books, records records and facilities related to the assets to be divested, VOLUME 144 Initial Decision divested, or to any other relevant information, as the Divestiture Trustee may request. Respondent shall develop such financial or other information as such Divestiture Trustee may reasonably request and shall cooperate with the Divestiture Trustee. Respondent shall shall take no action to interfere with or impede the Divestiture Trustee’s accomplishment of the divestiture. divestiture. Any delays in divestiture caused by Respondent shall extend the time for divestiture under this Paragraph in an amount equal to the delay, as determined by the Commission or, for a court-appointed court-appointed Divestiture Trustee, by the court. 5. The Divestiture Trustee shall use his or her best efforts to negotiate the most favorable price and terms available in each contract that is submitted to the Commission, but shall divest expeditiously at no minimum price. The divestiture shall be made only to an Acquirer that receives the prior approval of the Commission, and the divestiture shall be accomplished only in a manner that receives the prior approval of the Commission; provided, however, if the Divestiture Trustee receives bona fide offers from more than one acquiring entity, and if the Commission determines to approve more than one such acquiring entity, the Divestiture Trustee shall divest to the acquiring entity or entities selected by Respondent from among those approved by the Commission; provided, further, that Respondent shall select such entity within ten (10) business days of receiving written notification of the Commission’s approval. 6. The Divestiture Trustee shall serve, without bond or other security, at the cost and expense of Respondent, on on such reasonable and customary terms and conditions conditions as the Commission or a court may set. The EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 371 Initial Decision Divestiture Trustee shall have the authority to employ, at at the cost and expense of Respondent, such consultants, consultants, accountants, attorneys, investment bankers, bankers, business brokers, appraisers, and other representatives and assistants as are necessary to carry out the Divestiture Trustee’s duties and responsibilities. responsibilities. The Divestiture Trustee shall account for for all monies derived from the divestiture and all expenses incurred. After approval by the Commission or, or, in the case of a court-appointed Divestiture Trustee, by the court, of the account of the Divestiture Trustee, including fees for his or her services, all remaining monies shall be paid at the direction of the Respondent, Respondent, and the Divestiture Trustee’s power shall be be terminated. The Divestiture Trustee’s compensation shall be based at least in significant part on a commission arrangement contingent on the Divestiture Trustee’s divesting the assets.

7. If the Divestiture Trustee ceases to act or fails to act diligently, the Commission may appoint a substitute Divestiture Trustee in the same manner as provided in this Paragraph VI for appointment of the initial Divestiture Trustee.

8. The Divestiture Trustee shall have no obligation or authority to operate or maintain the assets to be divested. 9. The Divestiture Trustee shall report in writing to the Commission every sixty (60) days concerning the Divestiture Trustee’s efforts to accomplish the divestiture.

D. The Commission or, in the case of a court-appointed Divestiture Trustee, the court, may on its own initiative, or at at the request of the Divestiture Trustee, issue such VOLUME 144 Initial Decision additional orders or directions as may be necessary or appropriate to accomplish the divestiture required by this Order.

VII.

IT IS FURTHER ORDERED that:

A. Respondent shall file a verified written report with the Commission setting forth in detail the manner and form in which it intends to comply, is complying, and has complied with this Order (i) no later than thirty (30) days from the date this Order becomes final (without regard to the finality of the divestiture requirements herein), and every thirty (30) days thereafter (measured from the date the first report is filed) until the divestiture of the Highland Park Hospital Assets is accomplished, and (ii) thereafter, every sixty (60) days (measured from the date of divestiture) until the date Respondent completes its obligations under this Order; provided, however, that Respondent shall also file the report required by this Paragraph VII at any other time as the Commission may require.

B. Respondent shall include in its compliance reports, among other things required by the Commission, a full description of the efforts being made to comply with the relevant Paragraphs of this Order, a description (when applicable) of all substantive contacts or negotiations relating to the divestiture required by Paragraph II of this Order, the identity of all parties contacted, copies of all written communications to and from such parties, internal documents and communications, and all reports and recommendations concerning the divestiture, the date of divestiture, and a statement that the divestiture has been accomplished in the manner approved by the Commission. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 373 Initial Decision VIII.

IT IS FURTHER ORDERED that Respondent shall notify the Commission at least thirty (30) days prior to (1) any proposed dissolution of Respondent, (2) any proposed acquisition, merger or consolidation of Respondent, or (3) any other change in Respondent that may affect compliance obligations arising out of this Order, including but not limited to assignment, the creation or dissolution of subsidiaries, or any other change in Respondent. IX.

IT IS FURTHER ORDERED that, for the purpose of determining or securing compliance with this Order, and subject to any legally recognized privilege, and upon written request with reasonable notice, Respondent shall permit any duly authorized representative of the Commission:

A. Access, during office hours and in the presence of counsel, to all facilities and access to inspect and copy all nonprivileged books, ledgers, accounts, correspondence, memoranda and other records and documents in the possession or under the control of Respondent relating to any matter contained in this Order; and B. Upon five (5) days’ notice to Respondent and without restraint or interference from them, to interview their officers, directors, or employees, who may have counsel present, regarding any such matters.

VOLUME 144 Initial Decision ATTACHMENT 1 EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 375 Opinion of the Commission OPINION OF THE COMMISSION By Majoras, Chairman I. INTRODUCTION1 In 2000, Evanston Northwestern Healthcare Corporation (“Evanston”) merged with Highland Park Hospital (“Highland Park”). Prior to the merger, Evanston owned Evanston Hospital and and Glenbrook Hospital.2 The Commission issued an administrative complaint challenging challenging Evanston’s acquisition of Highland Park under Section 7 Section 7 of the Clayton Act four years after the transaction closed. closed. Given that the merger was consummated well before the Commission commenced this case, we were able to examine not 1 This opinion uses the following abbreviations: CB – Complaint Counsel’s Brief on Appeal and Cross-Appeal CFF – Complaint Counsel’s Proposed Findings of Fact CX – Complaint Counsel’s Exhibit DX – Demonstrative Exhibit ID – Initial Decision of the Administrative Law Judge IDF – Numbered Findings of Fact in the ALJ’s Initial Opinion JX – Joint Exhibits RB – Respondent’s Appeal Brief RFF – Respondent’s Proposed Findings of Fact RFF Reply – Respondent’s Reply Findings of Fact RPTB – Respondent’s Post-Trial Brief RRB – Respondent’s Brief in Reply and Opposition to Cross-Appeal RX – Respondent’s Exhibit TR – Transcript of Trial before the ALJ. 2 In this opinion, unless otherwise noted, we adopt complaint counsel’s convention of referring to the pre-merger Evanston Northwestern Healthcare Corporation entity (including Glenbrook Hospital) as “Evanston” or “Evanston Hospital.” “Highland Park” refers to the pre- and post-merger Highland Park Hospital facility, as well as Lakeland Health Services, Inc., the parent corporation of Highland Park Hospital prior to the merger. “ENH” refers to the post-merger entity that includes Evanston Hospital, Glenbrook Hospital, and Highland Park Hospital.

VOLUME 144 Opinion of the Commission only pre-merger evidence, but also evidence about what happened after the merger.

There is no dispute that ENH substantially raised its prices shortly after the merging parties consummated the transaction. There is disagreement about the cause of those price increases, however. Complaint counsel maintains that the merger eliminated significant competition between Evanston and Highland Park, which allowed ENH to exercise market power against health care insurance companies. Respondent argues that, during the due diligence process for the merger, ENH obtained information about Highland Park’s prices that showed that Evanston had been charging rates that were below competitive levels for a number of years. Respondent contends that most of ENH’s merger-related price increases simply reflect its efforts to raise Evanston Hospital’s prices to competitive rates. Respondent also maintains that some portion of the mergerrelated price increases reflects increased demand for Highland Park’s services due to post-merger improvements at the hospital. Chief Administrative Law Judge Stephen J. McGuire (“ALJ”) found in his Initial Decision that the transaction violated Section 7 of the Clayton Act and ordered ENH to divest Highland Park. We affirm the ALJ’s decision that the transaction violated Section 7 of the Clayton Act. Considered as a whole, the evidence demonstrates that the transaction enabled the merged firm to exercise market power and that the resulting anticompetitive effects were not offset by merger-specific efficiencies. The record shows that senior officials at Evanston and Highland Park anticipated that the merger would give them greater leverage to raise prices, that the merged firm did raise its prices immediately and substantially after completion of the transaction, and that the same senior officials attributed the price increases in part to increased bargaining leverage produced by the merger.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 377 Opinion of the Commission The econometric analyses performed by both complaint counsel’s and Respondent’s economists also strongly support the conclusion that the merger gave the combined entity the ability to raise prices through the exercise of market power. The economists determined that there were substantial merger-coincident price increases and ran regressions using different data sets and a variety of control groups that ruled out the most likely competitively-benign explanations for substantial portions of these increases. The record does not support Respondent’s position that the merger-coincident price increases reflect ENH’s attempts to correct a multi-year failure by Evanston’s senior officials to charge market rates to many of its customers, or increased demand for Highland Park’s services due to post-merger improvements.

We do not agree with the ALJ, however, that a divestiture is warranted. The potentially high costs inherent in the separation of hospitals that have functioned as a merged entity for seven years instead warrant a remedy that restores the lost competition through injunctive relief.

II. PROCEDURAL HISTORY A. Pleadings The Commission issued a three-count complaint on February 10, 2004. The first count alleged that the merger violated Section 7 of the Clayton Act in specified relevant product and geographic markets. Compl. ¶¶ 16-17, 27. The complaint alleged that the relevant product market was “general acute care inpatient hospital services” and that the relevant geographic market consisted of the “area directly proximate to the three ENH hospitals and contiguous geographic areas in northeast Cook County and southeast Lake County, Illinois.” Id. ¶¶ 16-17.

Count II charged that the transaction violated the Clayton Act because it enabled ENH to raise its prices to private payors above VOLUME 144 Opinion of the Commission the prices that the hospitals would have charged absent the merger. merger. Id. ¶ 32. Unlike Count I, however, Count II did not allege a a particular product or geographic market and did not incorporate the complaint’s earlier product market and geographic market allegations by reference. Id.. ¶¶ 29-32. Respondent denied the material allegations of Counts I and II. Respondent also asserted a number of defenses, the most pertinent of which is that the merger yielded significant efficiencies and improvements in the quality of patient care that outweigh any alleged anticompetitive effects. Second Am. Answer, ¶¶ 1-15, 20-21. The complaint’s third count alleged that ENH had engaged in price fixing on behalf of physicians whom it employed and other affiliated physicians. Compl. ¶¶ 33-44. This count was resolved by a by a consent agreement, which became final on May 17, 2005.3 Count III is not at issue in this appeal. B. Initial Decision The case was assigned to the ALJ, who conducted an eight-week trial. Forty-two witnesses testified, and the ALJ admitted more than 1600 exhibits into evidence.

The ALJ issued his Initial Decision on October 17, 2005. The ALJ first made careful and extensive findings of fact about the merging parties, the health care sector, and the transaction’s competitive effects. The ALJ then started his legal analysis by holding that the Clayton Act requires complaint counsel to prove the the relevant product and geographic markets. ID 131. Complaint counsel argued at trial that the relevant product market was general general acute care inpatient services sold by hospitals to private 3 In re Evanston Northwestern Healthcare Corp., Dkt. No. 9315 (FTC May 17, 2005), available at http://www.ftc.gov/os/adjpro/d9315/050520do .pdf. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 379 Opinion of the Commission health insurance companies, which typically are referred to as managed care organizations or “MCOs.” Id. ENH maintained that the product market also included hospital-supplied outpatient services. ID 131-32; RPTB 16-17. The ALJ rejected ENH’s position position and found that MCOs cannot substitute outpatient for inpatient services, determining that ENH had set its inpatient rates without concern that patients would switch from inpatient to outpatient services. ID 133.4 The ALJ then defined the relevant geographic market. ID 135- 49. Complaint counsel argued that the geographic market consisted of the geographic triangle immediately surrounding the three merging hospitals, which contained only the ENH hospitals. ID 137. Respondent advocated that the geographic market included the three ENH hospitals and at least six other hospitals (Lake Forest, Advocate Lutheran General, Rush North Shore, St. Francis, Condell, and Resurrection). Id. The ALJ held that the geographic market was larger than that proposed by complaint counsel, but smaller than the market advocated by respondent, finding that the geographic market consisted of the area that covered the three ENH hospitals and four other hospitals – Lake Forest, Advocate Lutheran General, Rush North Shore, and St. Francis. ID 143.

The ALJ next assessed the competitive effects of the merger. ID ID 150-69. Using the seven-hospital geographic market, the ALJ found that the ENH hospitals had a 35% pre-merger market share based on inpatient revenues. IDF ¶ 317. The ALJ then calculated a 4 Relying on the U.S. Court of Appeals for the Seventh Circuit’s decision in United States v. Rockford Memorial Corp., 898 F.2d 1278, 1284 (7th Cir. 1990), the ALJ also held that the fact that inpatient and outpatient services have a common provider does not compel a finding that they are in the same relevant product market. ID 133-34.

VOLUME 144 Opinion of the Commission pre-merger Herfindahl-Hirschman Index (“HHI”)5 of 2355, and a post-merger HHI increase of 384 to 2739. IDF ¶¶ 314-19. The ALJ ALJ found that, under § 1.51 of the Department of Justice’s and Federal Trade Commission’s Merger Guidelines, the HHI change and post-merger HHI created a presumption that the merger was likely to create or enhance market power. IDF ¶¶ 314-25; ID 150-52. 150-52.

The ALJ also considered direct evidence of the transaction’s effect on competition. The ALJ found that senior officials at Evanston and Highland Park had predicted before the merger that the transaction would put the combined firm in a better bargaining position with the MCOs, that ENH’s revenues increased substantially after the merger, and that ENH management believed that the merger had “translated to better managed care contracts.” ID 155-60, 165. The ALJ also relied heavily on the econometric evidence presented at trial, which he found, viewed in conjunction with other evidence, supported a finding that market power was “the only plausible, economically sound, and factually well-founded explanation for ENH’s post-merger relative price increases.” ID 166- 69.

The ALJ also concluded that entry by new hospitals, or expansion by existing hospitals, was not likely to replace the competition lost due to the merger. ID 194-95. Finally, the ALJ concluded that the merger had not produced significant improvements in the quality of care at Highland Park that offset the anticompetitive exercise of market power. ID 175-92. Based on his findings of fact and conclusions of law, the ALJ ruled that the transaction violated the Clayton Act, as alleged in 5 The HHI is calculated by summing the squares of the individual market shares of all the participants. U.S. Dept of Justice & Federal Trade Commu, Horizontal Merger Guidelines § 1.5 (1992, revised 1997) (“Merger Guidelines”), available at http://www.ftc.gov/bc/docs/horizmer.htm. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 381 Opinion of the Commission Count I of the complaint. ID 200.6 The ALJ dismissed Count II as moot, but held that, if it were not moot, he would have dismissed it it because complaint counsel had not established as part of Count II II that respondent possessed a substantial share of a relevant market. market. ID 200-01. As stated, the ALJ ordered ENH to divest Highland Park. ENH appealed the ALJ’s Initial Decision to the Commission. Complaint counsel cross-appealed the ALJ’s decision decision not to make a ruling against respondent under Count II and and also requested that the Commission supplement and revise the ALJ’s divestiture order.7 6 The ALJ rejected several other arguments made by ENH. First, he rejected ENH’s contention that its nonprofit status reduced the likelihood of competitive harm, finding that there was no evidence in the record that ENH’s nonprofit status had restrained its efforts to negotiate higher prices. ID 192-94. Second, the ALJ rejected ENH’s argument that the merger was necessary for Highland Park’s economic survival, concluding that, at the time of the merger, Highland Park was able to meet its financial obligations for the near future, and was in no danger of entering bankruptcy or exiting the market. ID 197. Finally, the ALJ rejected ENH’s position that the merger of Evanston and Highland Park could not violate the Clayton Act because, at the time of the merger, the two hospitals were not separate entities. ID 197-99. He found that the two hospitals were separate entities and that the transaction was subject to the Clayton Act. ID 197-99. Respondent did not identify this last issue as one of the “questions presented” on appeal, RB 23, and only briefly referenced it at the end of its brief in the context of discussing the appropriate remedy. RB 86. Accordingly, the Commission views the issue as not properly before us. In any case, for the reasons set forth by the ALJ, the Commission also finds that the transaction was subject to the Clayton Act.

7 Complaint counsel also requested that the Commission vacate the ALJ’s order of September 24, 2006, which denied complaint counsel’s motion to compel the production of certain documents on respondent’s electronic back-up tapes. The Commission denies this request because the issue is now moot. VOLUME 144 Opinion of the Commission III. STANDARD OF REVIEW Pursuant to 16 C.F.R. § 3.54 (2007), the Commission reviews the the record de novo by considering “such parts of the record as are cited or as may be necessary to resolve the issues presented and . . . . exercis[ing] all the powers which [the Commission] could have exercised if it had made the initial decision.”8 IV. FINDINGS OF FACT A. Third-Party Payor Insurance System In many markets, vendors set or negotiate a price, which is paid paid in full by their customers. The costs and benefits of the product product or service are fully internalized by the vendors and the customers. The market for hospital services is more complex. Hospitals and patients rarely negotiate directly over the price of hospital services, and patients almost never pay directly the full cost cost of the hospital services that they receive. TR 480 (Mendonsa); (Mendonsa); TR 2456-58, 2461, 2464-65 (Haas-Wilson); TR 5906 5906 (Noether). Instead, various types of “third-party payors” (primarily public and private insurance entities) negotiate the prices prices in advance on a periodic basis, and pay the bulk of the hospitals’ charges. TR 480 (Mendonsa); TR 2457-58, 2461 (Haas- (Haas-Wilson); RX 1743 at 6-7. Private insurance companies then sell health care policies to employers and individuals, who pay premiums for the policies. TR 2461-62 (Haas-Wilson). Individual 8 We adopt the ALJ’s findings of fact to the extent those findings are not inconsistent with this opinion. In addition, unless otherwise noted, any Commission citation to any trial testimony, exhibit, or deposition segment in this opinion constitutes a determination by the Commission that the cited testimony, exhibit, or deposition segment is relevant, material, and reliable evidence, and therefore admitted into the record of this proceeding. 16 C.F.R. § 3.43(b). Each such determination shall be conclusive, with respect to determining the contents of the record of this proceeding, notwithstanding any objection or response thereto registered by either complaint counsel or counsel for respondent. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 383 Opinion of the Commission members often also pay a co-payment amount or a deductible when they use hospital services. TR 477-78 (Mendonsa); TR 2464 (Haas- Wilson).

The primary public third-party payors are the federal government’s Medicare and the joint federal and state Medicaid programs. Medicare provides health insurance for the elderly, and Medicaid provides coverage for low-income persons. TR 2454 (Haas-Wilson). ENH obtains slightly less than half of its revenues from patients who are covered by the Medicare or Medicaid programs. IDF ¶¶ 127, 134-35. We do not discuss the Medicare and Medicaid systems further because complaint counsel did not allege that the merger increased the prices paid by Medicare or Medicaid for hospital services.

Approximately half of ENH’s revenues come from private insurers. IDF ¶ 134.9 The United States has a largely employer-based health care system in which a majority of consumers who have private health insurance obtain it through their employers. Typically, consumers select an insurance plan from one or more private insurance companies with which their employers have contracted. TR 2460-61 (Haas-Wilson).

The private health insurance market has changed substantially over the past two decades. In the 1980s, the predominant type of insurance in Chicago was indemnity insurance. IDF ¶ 153; TR 1831- 32 (Hillebrand). In a typical indemnity plan, the consumer could select any hospital (or doctor), and the insurance company reimbursed the individual a set amount based on the care provided. IDF ¶ 155. Because indemnity plans allowed their insureds to select any hospital or provider, hospitals did not need to compete to be covered by the plans. TR 2466 (Haas-Wilson). Concerns about rising costs, among other factors, gave rise to MCOs, which now account for the vast majority of private insurance 9 The remaining portions of ENH’s volume are charity care and a very small percentage of patients who self-pay. IDF ¶ 137. VOLUME 144 Opinion of the Commission insurance in the Chicago market. TR 1832-33 (Hillebrand). There are two broad categories of MCO plans: health maintenance organization plans (“HMOs”) and preferred provider organization plans (“PPOs”). An HMO plan provides coverage to members through a “network” of physicians, hospitals, and other health care care providers that contract to furnish such services. RX 1743 at 6. 6. An HMO is generally a fully insured product: employers and consumers pay premiums to the provider of the HMO, and the provider assumes the risk that those premiums will be sufficient to cover the members’ healthcare expenses. TR 585 (Neary). Because Because the insurance company assumes the risk, HMO plans often often have a smaller network of physicians and hospitals than do risk-sharing plans, and they provide benefits only to members who who receive care from in-network providers. TR 1759-60 (Hillebrand); TR 477 (Mendonsa).

PPOs include elements of both managed care and fee-for-service fee-for-service arrangements.. RX 1743 at 6. A typical PPO plan has has contracts with a range of health care providers that is larger than than the range of providers in an HMO network. TR 2460 (Haas- (Haas-Wilson). PPOs generally offer members substantial financial financial incentives to obtain their health care “in network” or from from “preferred providers.” TR 477-78 (Mendonsa); RX 1743 at 6. 6. PPO members, however, can obtain health care from other providers at additional cost. IDF ¶ 148; TR 477-78 (Mendonsa). Many MCOs offer both HMO and PPO plans. TR 477 (Mendonsa).10 Depending on the type of insurance plan, when consumers receive services from an in-network hospital, they pay a deductible deductible and/or a co-payment, RX 1743 at 6, which usually 10 There are also point of service plans (POS). A POS is a variation of a PPO that contracts with a limited number of hospitals and doctors and extends terms of coverage to enrollees based on terms that vary depending on the provider from which the enrollee seeks care. CFF ¶¶ 187-88. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 385 Opinion of the Commission constitutes a small portion of the total price for the services that the patient receives. PPOs generally are more expensive than HMOs because they provide coverage or reimbursement for a larger set of providers. In the Chicago area, the use of HMOs has declined substantially in favor of PPOs. TR 1834 (Hillebrand); TR 479-80 (Mendonsa).

B. Competition Among Hospitals for MCO Contracts MCOs enter into two basic types of contracts with hospitals – “per diem” and “discount off charges.” In per diem contracts, there is an all-inclusive per day charge, based on the class of services, for each day that the patient is in the hospital, regardless of the amount or the total cost of the services that the patient receives. IDF ¶ 178; JX 8 at 8-9. Under discount off charges contracts, the MCO agrees to pay the hospital a rate for each service performed. The paid rate is equal to the hospital’s list price of the service, discounted by an agreed upon percentage. IDF ¶ 173. The list prices are contained in the hospital’s “chargemaster.” IDF ¶ 175. Thus, the prices paid by MCOs increase as a hospital increases the prices in its chargemaster. All else being equal, MCOs usually prefer per diem contracts because they allow for greater certainty about MCOs’ costs. IDF ¶¶ 179-80; TR 5740 (Sirabian).

MCOs do not typically select every hospital in a geographic region for their HMO networks, IDF 158, and they do not designate designate every provider as preferred for their PPOs. IDF ¶¶ 158-67; 158-67; TR 2457-60 (Haas-Wilson). Rather, physicians and hospitals compete to be included in HMO and PPO networks. IDF ¶ ¶ 109. The central terms of competition are price, quality of service, service, and geographic proximity to the MCO’s members. IDF ¶¶ 109, 121. The use of a business model that potentially excludes some providers allows MCOs to leverage competing providers against each other to negotiate lower prices. TR 2470 (Haas- Wilson); TR 6189 (Noether). Through this competitive process, MCOs seek to assemble high-quality networks at competitive rates VOLUME 144 Opinion of the Commission that include a sufficient number of hospitals and physicians to attract employers and their employees. IDF ¶¶ 109, 121, 158. C. Competition Among MCOs to be Selected by Employers As stated, a majority of people in the United States who have private health insurance obtain it through their employers. TR 2454 2454 (Haas-Wilson).11 Typically, the employer selects which MCOs and plans to offer its employees. TR 2460-61 (Haas-Wilson). Because employees sometimes consider the quality of health care benefits when they decide where to accept employment, many employers try to provide health care plans that are attractive to their employees. IDF ¶ 120; TR 2407 (Elzinga). Thus, employer demand for MCO services is a partially derived demand from employee preferences. TR 5936-37 (Noether); TR 2407 (Elzinga). As a general matter, employees prefer health plans that offer a broad choice of hospitals (and physicians) that are geographically convenient for them and their families. TR 2461 (Haas-Wilson); TR 485 (Mendonsa); TR 568 (Mendonsa), in camera. At the same time, employees (and employers) want to limit the amount of money that they spend on employee health benefits. TR 2461 (Haas-Wilson). Consequently, MCOs compete to have employers offer their plans based on price, quality, the geographic convenience of the hospitals and physicians in their networks, and other factors relevant to employees and employers. IDF ¶¶ 114, 117 252-53; TR 2407-08 (Elzinga); TR 2803 (Haas-Wilson), in camera. Similarly, because some employers offer their employees several plans from which to choose, TR 491 (Mendonsa), an MCO needs to offer an attractive 11 As respondent notes, employers generally are self-insured or fully-insured. RFF ¶ 54. Self-insured employers are responsible for the actual medical expenses of their employees but pay MCOs to access and manage the network and to process claims. TR 480 (Mendonsa). Fully-insured employers are liable only for premiums but not for the actual healthcare dollars spent by employees. RX 1743 at 6.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 387 Opinion of the Commission network to convince employees to enroll in its plan as opposed to a plan from one of its competitors. TR 2461 (Haas-Wilson); TR 5948 (Noether).

D. Consumer Harm from Increases in Hospital Prices Consumers are harmed when hospital prices increase due to the exercise of market power, even though they usually do not pay directly the full price of a hospital visit. TR 239 (Ballengee), in camera; TR 483-84 (Mendonsa); TR 549 (Mendonsa), in camera. When a hospital succeeds in raising its prices to an MCO, the MCO generally passes on those costs to the employers, which in turn pass them on to the employees. TR 483-84 (Mendonsa); TR 171-72, 179, 196-97 (Ballengee); TR 2463 (Haas-Wilson). Similarly, self-insured employers often pass on higher hospital costs to their employees. IDF ¶ 189. Thus, if a hospital can increase its market power by merging with a close competitor, the resulting price increases harm consumers.

Significantly, consumers who use a particular hospital will not necessarily pay for all of a price increase imposed by that hospital. Much of the cost may be borne by consumers who always use other hospitals. This is because consumers usually pay only the deductible or co-payment when they use a hospital, and MCOs do not necessarily vary these amounts for in-network or preferred providers, even when there is substantial variation among these providers’ prices to the MCO. TR 2464 (Haas-Wilson). Rather, MCOs often pass on the higher costs to employers and then consumers through higher premiums or across-the-board increases to deductibles and/or co-payment amounts. TR 483-84 (Mendonsa); TR 171-72 (Ballengee). This dynamic does not reduce the anticompetitive effects of hospital price increases to MCOs due to market power, but it does alter who incurs the costs of those effects. E. Types of Hospital Services VOLUME 144 Opinion of the Commission Hospitals provide a wide range of services, ranging from minor outpatient procedures to complex organ transplants and experimental treatments. TR 158-59 (Ballengee); TR 622 (Neary); TR 6159-60 (Noether). There is not precise agreement about how to categorize hospital services, but the record reflects that it is appropriate to classify hospital services into three broad categories: primary, secondary, and tertiary services. Primary services generally consist of internal medicine, obstetrics, and minor surgery. IDF ¶ 197; TR 6159 (Noether); TR 1293 (Neaman). Some primary services are provided on an outpatient basis. Outpatient services generally are considered to be any service for which a patient remains in the hospital for less than twenty-four hours. TR 302 (Newton); TR 144 (Ballengee).

Secondary services largely consist of inpatient medical services services provided by a specialist, including standard surgery, and generally require more skill, expertise, or equipment than primary care services. IDF ¶ 198; TR 1294 (Neaman); TR 6159 (Noether). Tertiary services refer to major surgical or medical procedures that that are done within a hospital setting. IDF ¶ 199; TR 1294 (Neaman).12 F. Parties 1. Evanston Northwestern Healthcare 12 There are even more complex medical services, which sometimes are referred to as “quaternary services.” TR 1294 (Neaman); TR 2009 (Hillebrand); TR 2491 (Haas-Wilson); TR 2701 (Haas-Wilson), in camera. The record does not indicate that there is a consensus about how to categorize these services, but they include procedures such as solid organ transplants and treatment for severe burns, TR 2491 (Haas-Wilson), and require very specific human and physical capital. TR 2701 (Haas-Wilson), in camera. Neither Evanston nor Highland Park provides these types of medical services. TR 298 (Newton); TR 1295, 1378 (Neaman); TR 2009-10 (Hillebrand); TR 2665 (Haas-Wilson). Other hospitals in the Chicago area, such as Northwestern Memorial Hospital and the University of Chicago, do offer these very advanced services. TR 1378 (Neaman). EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 389 Opinion of the Commission Evanston owned two hospitals, Evanston Hospital and Glenbrook Hospital (“Glenbrook”), prior to merging with Highland Park Hospital. Evanston Hospital is located in Evanston, Illinois. It is a 400-bed facility that provides a range of primary, secondary, and tertiary services. For example, Evanston offers obstetrical and pediatric services, psychiatric care, neurosurgery, radiation therapy, cardiology services, orthopedics, trauma centers, and the Kellogg Cancer Care Center. CX 84 at 8; CX 681 at 2; TR 299 (Newton); TR 1291-93 (Neaman); TR 2083-84 (Spaeth).

Glenbrook is a 125-bed facility located in Glenview, Illinois. IDF ¶¶ 2, 9, 11. Glenbrook provides primary and secondary services. IDF ¶ 12.

In fiscal year 1998, Evanston Hospital and Glenbrook together generated $441 million in revenue. CX 84 at 16. That year, 51% of Evanston’s revenue came from private MCOs, 37% from Medicare and Medicaid, and 12% from other sources. CX 84 at 8. 2. Lakeland Health Services Highland Park Hospital was the sole subsidiary of Lakeland Health Services, Inc. The hospital is located in Highland Park, Illinois, and has approximately 150 to 200 beds. IDF ¶¶ 20, 22. Before the merger, Highland Park offered primary and secondary services, but not tertiary services. IDF ¶¶ 22, 202, 203. The services offered included obstetrical service, a level II prenatal center, pediatric services, diagnostic services, a fertility center, psychiatric care, neurosurgery, radiation therapy, cardiology services, and a level II trauma center. CX 84 at 13, 15; CX 699 at 24; TR 299 (Newton); TR 2083-88 (Spaeth).

In fiscal year 1998, Highland Park generated $101 million in revenue. CX 84 at 16. Forty-five percent of Highland Park’s revenues that year came from MCOs, 43% from Medicare and Medicaid, and 12% from other sources. CX 84 at 13. VOLUME 144 Opinion of the Commission G. Other Hospitals in the Geographic Region Evanston, Glenbrook, and Highland Park Hospitals are located in the affluent suburban towns north of Chicago, generally referred to as the North Shore suburbs. IDF ¶ 227; TR 516-17 (Mendonsa), in camera; TR 901-02 (Foucre); TR 360 (Newton); TR 602 (Neary). The North Shore suburbs start at Evanston and include Glencoe, Wilmette, Winnetka, Kenilworth, Highland Park, and Lake Forest. TR 162-63 (Ballengee); TR 484 (Mendonsa). Regarding the hospital coverage in the area, one of the MCO witnesses testified that a person traveling up the North Shore from Chicago “would stop at Evanston” and then “Highland Park would be the next hospital.” TR 1426 (Holt-Darcy).

The three ENH hospitals form a triangle, one long side of which runs along Lake Michigan between Highland Park and Evanston Hospitals. Evanston is approximately 13.7 miles and 27 minutes south of Highland Park. IDF ¶ 21. Glenbrook is located 12.6 miles and 26 minutes west of Evanston Hospital and approximately 7 miles southwest of Highland Park. IDF ¶ 10. There are approximately 100 hospitals in the Chicago metropolitan area, TR 5982 (Noether), but no other hospitals within the triangle formed by the three ENH hospitals. TR 901-02 (Foucre); TR 167-68 (Ballengee). There are, however, other nearby hospitals, including nine hospitals that are closer to Evanston, Glenbrook, or Highland Park than they are to each other. RX 1912 at 20, 21, in camera; RB 29. These hospitals include:

1. Advocate Lutheran General Advocate Lutheran General is 10.2 miles west of Evanston Hospital, approximately a 21-minute drive. IDF ¶ 272; RX 1912 at EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 391 Opinion of the Commission at 20, in camera. Advocate Lutheran General is a 521-bed hospital that provides primary, secondary, and tertiary care. IDF ¶¶ 273-74. Advocate Lutheran General is the largest hospital in the Advocate system, which itself consists of eight hospitals. IDF ¶ 273; RX 1503 at 22, in camera; RX 1912 at 60.

2. Rush North Shore Rush North Shore is 3.7 miles southwest of Evanston Hospital, approximately a 9-minute drive. IDF ¶ 281; RX 1912 at 20, in camera. Rush North Shore has 150 to 200 beds and provides primary, secondary, and some level of tertiary services. IDF ¶ 282. 3. St. Francis St. Francis is 3 miles south of Evanston Hospital, approximately an 8-minute drive. IDF ¶ 87; RX 1912 at 20, in camera. St. Francis has 300 to 400 beds. IDF 288. St. Francis provides primary, secondary, and some level of tertiary services. IDF ¶ 289. 4. Resurrection Resurrection Medical Center is 12.1 miles southwest from Evanston, approximately a 25-minute drive. IDF ¶ 298; RX 1912 at 20, in camera. Resurrection has 350 beds. IDF ¶ 299; RX 1912 at 60, in camera.

VOLUME 144 Opinion of the Commission 5. Holy Family Holy Family is 11.3 miles west of Evanston, approximately a 23minute drive. RX 1912 at 20-21, in camera. Holy Family has 260 staffed beds. IDF ¶ 305.

6. Swedish Covenant Swedish Covenant is an urban hospital located 6.8 miles south of Evanston, approximately a 19-minute drive. IDF ¶ 306; RX 1912 at 20, in camera. Swedish Covenant has 325 beds, IDF ¶ 306, and provides primary, secondary, and tertiary services. CFF 1935. 7. Northwestern Memorial Northwestern Memorial is located in downtown Chicago, roughly 13 miles south of Evanston, approximately a 26-minute drive. IDF ¶ 308; RX 1912 at 20, in camera. Northwestern has more than 700 beds, and provides primary, secondary, and tertiary services. IDF ¶ 308. Northwestern Memorial is affiliated with the Northwestern Medical School. Id.

8. Lake Forest Lake Forest is 6.1 miles northwest of Highland Park, approximately a 13-minute drive. IDF ¶ 266. Lake Forest is a 142bed hospital, and provides primary and secondary services, including a significant level of obstetric services. IDF ¶ 267; TR 1304 (Neaman).

9. Condell Condell is 12.7 miles northwest of Highland Park, approximately approximately a 24-minute drive. IDF ¶ 293; RX 1912 at 20, in EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 393 Opinion of the Commission camera. Condell is a 163-bed hospital and provides primary, secondary, and some level of tertiary services. IDF ¶¶ 294-95. H. Parties’ Pre-Merger Objectives The parties signed a letter of intent to merge on July 1, 1999, and entered into the merger agreement in October 1999. IDF ¶¶ 81, 83. The parties completed the merger on January 1, 2000. IDF ¶ 85. The record reflects, and we find, that the parties had three objectives for the merger – raising prices, achieving economies of scale, and developing new programs at Highland Park. Mark Neaman, who joined Evanston in 1973 and has served as its Chief Executive Officer since 1992, TR 1278 (Neaman), testified that he hoped that Evanston’s merger with Highland Park would allow it to obtain better prices from MCOs. TR 1036 (Neaman). The parties’ premerger business records state that Evanston’s most senior officials thought that the merger would allow Evanston to do just that. At a January 4, 1999 meeting between Evanston and Highland Park’s board members and medical staff leaders, Evanston representatives identified the merger as an opportunity to “strengthen negotiation capability with managed care companies through merged entities” and not to “'compete with self’ in covered zip codes (e.g., 60% to 70% market shares) such as Evanston, Glenview, Highland Park, and Deerfield.” CX 1 at 3. Likewise, the minutes of an April 5, 1999 meeting record an Evanston representative as saying that “[t]his would be an opportunity to join forces and grow together rather than compete with each other.” CX 2 at 7. In September 29, 1999, Neaman told his managers and his Board that the merger would “[i]ncrease our leverage, limited as it might be, with the managed care players and help our negotiating posture.” IDF ¶ 335; CX 1566 at 9.

Neaman and Ronald Spaeth, the President and Chief Executive Executive Officer of Highland Park before the merger, also wrote that a goal of the transaction was to “strengthen their negotiating positions with managed care” organizations. CX 19 at 1; TR 1036- 1036-37 (Neaman). A Spring 1999 report by Highland Park’s VOLUME 144 Opinion of the Commission Chairman explains: “Everybody progresses [sic] to see the community benefit that would be derived as well as the economic benefit of not being out there doing battle with one another in what what will be a common battle ground if you want to call it that.” CX CX 4 at 1. Most significantly, Spaeth’s bottom-line conclusion about about the transaction was that “it would be real [sic] tough for any of of the Fortune 40 companies in this area whose CEOs either use this this place [Highland Park] or that place [Evanston Hospital and Glenbrook] to walk from Evanston, Highland Park, Glenbrook and and 1700 of their doctors.” Id. at 2.

We find that the testimony of Mark Newton, a former senior official at Highland Park, also supports the conclusion that Highland Park thought that the transaction would give it greater leverage to negotiate higher prices from payors. Newton testified that, before the merger, he had prepared an outline for a strategic planning retreat that identified various ways that Highland Park could increase its market share. TR 345-49 (Newton). The document identified the possibility of a merger between Highland Park and Evanston, Northwest Community, Lake Forest, or Condell Hospitals. CX 1869 at 6. Newton concluded that the merger between Highland Park and Evanston would produce the entity with the greatest negotiating strength with payors based on “the array of services, the numbers of the medical staff, as well as the communities that were being served.” TR 350-51 (Newton). He explained that “ [o]f the options that we had looked at in terms of merger . . . the power in the relevant market would be higher with Highland Park and Evanston than with those others.” TR 354 (Newton). The reasons included “the proximity of the institutions, the cultural relationships that exist in that community,[and] the placement of the medical staffs.” TR 354 (Newton).

Finally, we find that Evanston’s consultants also expressed confidence, prior to the closing, that the merger would give the combined company greater bargaining leverage with MCO customers. Evanston engaged the Bain consulting firm in the fall of EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 395 Opinion of the Commission of 1999 to assist in strategic planing related to the merger. TR 1159 1159 (Neaman). In an August 30, 1999 proposal letter from Bain to to Neaman, Bain wrote: “As a consequence of the merger, ENH will will have broad geographic coverage on the North Shore, with three three hospitals and an extensive physician network. The merger provides the opportunity to reduce costs, refocus activities at the three hospitals, shift activity from the overcrowded Evanston Hospital, and negotiate contracts with payors from a stronger position.” CX 2072 at 1. In October 1999, in a document entitled “Growth Opportunities from the Highland Park Merger,” Bain wrote wrote that “[b]etter integration with the ENH Medical Group and the the addition of Highland Park will substantially improve ENH’s leverage.” CX 74 at 19.

In October and November of 1999, Bain reviewed and analyzed Evanston’s and Highland Park’s contracts. CX 74; CX 75. Bain concluded that the merger would enable Evanston to grow net income by increasing revenue, due in part to higher prices and greater market share and to reduce costs through economies of scale, elimination of duplicate costs, and capital investment savings. CX 74 at 3. Bain also determined that the combined Evanston and Highland Park Hospitals would have “significant leverage with payors as [it has] the largest [number of] admissions” among other Chicago area hospitals. CX 74 at 15. An Evanston senior official testified at trial that he felt that Bain’s analyses were accurate and helpful. TR 1161(Hillebrand).

I. ENH’s Post-Merger Price Increases After the merger closed, ENH rapidly increased the prices that it it charged to most of its MCO customers to the higher of Evanston’s Evanston’s or Highland Park’s pre-merger rate for a particular service. IDF ¶¶ 348-54. ENH then set about negotiating a single contract for all three of its hospitals with each MCO. IDF ¶¶ 355-66; 355-66; TR 1528 (Holt-Darcy), in camera. ENH did not offer the MCOs the option to enter into separate contracts for the hospitals, or or to decline to use one or more of the three hospitals. IDF ¶¶ 355- VOLUME 144 Opinion of the Commission 355-66. In addition, ENH sought to raise its prices through the conversion of portions of some of its contracts from per diem to discount off charges payment structures. IDF ¶¶ 373-77. The record reflects that ENH’s post-merger negotiation strategy was highly successful. ENH negotiated with its MCO customers a single contract for all three of its hospitals with substantial price increases, and converted a number of its contracts from per diem to discount off charges structures. CX 5174 at 11, in camera; CX 5 at 5; TR 252 (Ballengee), in camera. In addition, from 2002 to 2003, ENH increased its chargemaster rates four times. IDF ¶ 384; RX 1687 at 3, in camera.

As we describe in detail below in our findings about the econometrics, the actual amount of ENH’s price increases depends depends on the calculation method. Using data that included all patients in Illinois, complaint counsel’s economist, Deborah Haas- Haas-Wilson, computed that from 1998 through 2002, ENH increased its per day average net prices by 48% for all patients; 46% 46% for the commercial and self-pay patients; and 46% for commercial, self-pay, self-administered, and HMO patients. CX 6279 at 7, in camera.13 On a per case basis, the corresponding average net price increases from 1998 to 2002 were 30%, 27%, and and 26%, respectively. Id., in camera.

Using data from individual MCOs, Haas-Wilson calculated the the level of ENH’s per case post-merger average net price changes changes for Aetna, Blue Cross/Blue Shield (“BCBS”), Humana, United Healthcare of Illinois (“United”), and Great West. She determined that ENH increased its per day average net prices by the the following amounts: Aetna (48% to 56%); BCBS (- (-12%(decrease) to 15%); Great West (79%); Humana (57% to 13 As we explain below, Haas-Wilson used various techniques to construct and estimate a “net price,” which consisted of the sum of (1) the payment from the MCO to the hospital, and (2) the payment from the patient to the hospital. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 397 Opinion of the Commission 82%); and United (77% to 202%). CX 6279 at 3, in camera; CX 6282 at 5, in camera. The corresponding per case average net price price increases were: Aetna (28% to 89%); BCBS (10% to 27%); Great West (42%); Humana (27% to 73%); and United (62% to 128%). CX 6279 at 3, 5, in camera. The ranges of price increases reflect that the price increases varied by the type of plan offered by by the MCOs (e.g.,HMO or PPO).

Respondent’s economist, Jonathan Baker, did not compute a market-wide price increase. Instead, Baker used two different methods to compute price changes from 1998 to 2003 for Aetna, BCBS, Humana, and United. The first calculation found the following per case average net price increases for Evanston, Glenbrook, and Highland Park: Aetna (35%); BCBS (13%); Humana Humana (83%); and United (138%). RX 2040 at 1, in camera; DX DX 7068 at 43, in camera. The per case average net price increase increase across all four payors was 42%. RX 2040 at 1, in camera; camera; DX 7068 at 43, in camera. The second calculation found the following per case average net price increases for only Evanston Evanston and Glenbrook: Aetna (25%), BCBS (2%), Humana (60%),United (140%), and an average per case increase across all four payors of 29%. RX 2040 at 1, in camera; DX 7068 at 43, in camera.14 Post-merger ENH documents indicate that ENH executives believed that the merger gave ENH the market power needed to achieve these price increases. The minutes of a September 27,2000 meeting of the ENH board’s finance committee state that ENH’s 14 Baker also performed these calculations omitting obstetrics cases because, as discussed below, there were some ambiguities in the data with respect to obstetrics. The corresponding per case average net price increases for Evanston, Glenbrook, and Highland Park Hospitals were: Aetna 34%, BCBS 5%, Humana 84%, and United 111%, with an average across all four payors of 37%. RX 2040 at 2, in camera; DX 7068 at 44, in camera. The corresponding per case price increases for only Evanston and Glenbrook were: Aetna 31%, BCBS 3%, Humana 82%, and United 124%, with an average across all four payors of 35%. RX 2040 at 2, in camera; DX 7068 at 44, in camera. VOLUME 144 Opinion of the Commission President Neaman attributed the price increases, at least in part, to the transaction: “[T]he larger market share created by adding Highland Park Hospital has translated to better managed care contracts.” CX 16 at 1. The next month, Neaman issued a memorandum entitled “Final Report – Merger Integration Activities” Activities” that stated: “Some $24 million of revenue enhancements enhancements have been achieved – mostly via managed care renegotiations,” and “none of this could have been achieved by either Evanston or Highland Park alone. The 'fighting unit’ of our three hospitals and 1600 physicians was instrumental in achieving these ends.” CX 17 at 1-2 (emphasis added). Portions of the trial testimony from Highland Park’s officials were consistent with these documents. Highland Park’s CEO before the merger, Spaeth, contrasted the post-merger price increases against Highland Park’s pre-merger negotiations, testifying that before the merger he did not see an opportunity to raise rates. TR 2172-73 (Spaeth). Terry Chan, Highland Park’s primary negotiator before the merger, testified that the merger gave ENH additional bargaining power. TR 709-10 (Chan); IDF ¶ 367. To summarize, we find that the documentary evidence and testimony support the conclusion that senior officials at Evanston and Highland Park anticipated that the merger would give them greater leverage to raise prices to MCOs, the merged firm did raise its prices to MCOs immediately and substantially after consummation of the transaction, and the same senior officials attributed the price increase in part to increased bargaining leverage with payors produced by the merger.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 399 Opinion of the Commission J. MCO Testimony Complaint counsel presented testimony from five MCOs at trial.15 1. Private Healthcare Systems (“PHCS”) PHCS develops networks of hospitals, doctors, and other ancillary services, and markets these networks to insurance companies, third-party administrators, and employers. TR 142-43 (Ballengee). Jane Ballengee, PHCS’ Regional Vice President for Network Development, testified about PHCS’ post-merger negotiations with ENH. Ballengee was PHCS’ Territory Director for the Chicago region when PHCS renegotiated its contract with ENH after the merger, although she did not participate in the negotiations. TR 146-47 (Ballengee).

Throughout the 1990s, PHCS had negotiated new rates with Evanston approximately every one and one-half years. TR 168-69 (Ballengee). Ballengee testified that PHCS viewed Highland Park as as Evanston’s “primary alternative” and that, before the merger, PHCS believed that it could select Evanston or Highland Park and “work them off against each other.” TR 166-68 (Ballengee). Prior to to the merger, PHCS had never threatened to drop either Evanston or or Highland Park, but PHCS believed that its ability to do so was understood and that this ability restrained the hospitals’ prices. TR 171 (Ballengee). PHCS had dropped other hospitals from its 15 We limit our findings about the MCO testimony to the MCOs’ descriptions of the role played by the ENH hospitals in their networks, their post-merger negotiations with ENH, ENH’s post-merger price increases, and which hospitals the MCOs viewed as competitors to the ENH hospitals before and after the merger. Both sides have presented extensive evidence about the tone and the rhetoric used during the MCO negotiations. We have carefully reviewed and considered these portions of the record and, while such information can be probative in antitrust cases, we have concluded that in this case this testimony neither supports nor undermines the conclusion that the merger gave ENH market power.

VOLUME 144 Opinion of the Commission network when it was not satisfied with the offered prices. TR 154-56 154-56 (Ballengee). Ballengee further testified that if Evanston had had made unacceptable price demands pre-merger, PHCS could have have eliminated it from the network and used Highland Park as the the alternative, and vice-versa. TR 167 (Ballengee). Ballengee testified that she believed that competition between Evanston and Highland Park had kept price increases to an average average of 4% to 8% for each contract renegotiation. TR 168-71 (Ballengee). By comparison, post-merger, ENH sought and obtained obtained what Ballengee testified was approximately a 60% price increase, primarily through increases in Evanston’s prices. TR 179 (Ballengee).16 Ballengee testified that PHCS accepted the increase because some of its customers had informed PHCS that they could not market their health plans without ENH in the network “[b]ecause there would be a large [geographic] area that would be uncovered.” TR 179-81 (Ballengee). Ballengee’s assessment of the market conditions is consistent with a document prepared for ENH by Bain at the time of the merger, which stated that ENH had “significant leverage in negotiations with PHCS as they have [a] strong North Shore presence and need us in their network.” CX 1998 at 44. On cross-examination, Ballengee also stated that she believed that Advocate Lutheran General and St. Francis were significant competitors to Evanston, and that Lake Forest was a significant competitor to Highland Park. TR 211-12 (Ballengee). She also stated that for purposes of forming a network, Advocate Lutheran and possibly Rush North Shore and Advocate Northside were comparable to Evanston. TR 191-93 (Ballengee). 16 Data analyzed by complaint counsel’s economist appeared to show that, post-merger, ENH increased its prices to PHCS (as a percentage per case) approximately 60%. CX 6279 at 4-5 (62.3% as calculated using the data received from the FTC’s Civil Investigative Demand to ENH and 59.6% as calculated using data received from the consulting firm NERA), in camera; TR 2522 (Haas- Wilson), in camera; CX 6279 at 4-5.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 401 Opinion of the Commission We find that Ballengee’s testimony, viewed in conjunction with with the Bain document, supports the conclusion that Evanston and and Highland Park were close substitutes that likely constrained each other’s pricing to PHCS before the merger. Ballengee’s testimony that Advocate Lutheran General, St. Francis, and possibly possibly Rush North Shore and Advocate Northside were significant significant competitors to Evanston, and that Lake Forest was a significant competitor to Highland Park, does not undermine this conclusion. The issue is not whether other hospitals competed with with the merging parties, but whether they did so to a sufficient degree to offset the loss of competition caused by the merger. The fact that PHCS retained ENH after it substantially raised prices at a a rate that exceeded the average rate increase of other hospitals, rather than drop ENH and use other hospitals, also supports the finding that, for PHCS, competition from these other hospitals was was not sufficient to constrain ENH from exercising market power.17 2. Aetna Robert Mendonsa, who was an Aetna general manager responsible for sales and network contracting, testified about Aetna’s Aetna’s negotiations with ENH after the merger. TR 475-76 (Mendonsa). Prior to the merger, Evanston and Aetna had last negotiated a contract in 1996, and the prices that Aetna negotiated at at that time had remained in effect through 2000. IDF ¶ 437; TR 533-34, 563 (Mendonsa), in camera. Mendonsa testified that the ENH hospitals had been part of Aetna’s network for many years because it was “extremely important” to include them. TR 516 (Mendonsa), in camera. Mendonsa also testified that it is very important to have hospital coverage in the North Shore suburbs 17 For the same reason, we find not particularly informative a PHCS statement to its customers during its post-merger negotiations with ENH, about the existence of other hospitals in the same geographic area as Evanston and Highland Park. RX 712 at 2-3. Further, Ballengee testified that her customers “made it very clear to [her] that they didn’t believe that they could have a marketable network, that they could compete in the marketplace without having the new ENH entity in it.” TR 180 (Ballengee). VOLUME 144 Opinion of the Commission because executives of employers live there who are involved in the companies’ decisions. TR 516-17 (Mendonsa), in camera. Mendonsa was concerned about the merger because it had resulted in “three extremely important hospitals negotiating together in a very important geography” and because it would “severely compromise[ ]” Aetna’s ability to sell its plans without the three hospitals. TR 530, 518 (Mendonsa), in camera.

On January 18, 2000, ENH wrote a letter to Aetna, requesting that it assign Highland Park’s rates to ENH until it negotiated a new hospital agreement with Aetna. RX 769 at ENH JL 2817. ENH’s letter also contained an initial proposal for a new contract. Id., in camera. Because Evanston’s rates for Aetna had not increased since 1996, Mendonsa expected ENH to ask for a price increase of approximately 10%. TR 534 (Mendonsa). By Aetna’s estimates, however, ENH sought a 65% increase. TR 533 (Mendonsa), in camera.

On March 14, 2000, ENH invoked the termination clause of the the existing pre-merger contract, giving Aetna notice that it would terminate the contract if the parties could not reach an agreement. CX 123 at 1; TR 546-47, 531 (Mendonsa), in camera. In June 2000, 2000, Aetna and ENH ultimately agreed to a contract that Aetna calculated increased ENH’s prices by approximately 45% to 47% over a three-year period. TR 539-40 (Mendonsa), in camera.18 Mendonsa testified that Aetna signed the post-merger contract with ENH because Aetna thought that people who lived in the communities around the ENH hospitals would not want to travel to to other hospitals. TR 541-43 (Mendonsa), in camera. He explained explained that he believed that “[s]omeone that’s going to Evanston Evanston is not going to drive all the way out to Park Ridge, which 18 Respondent does not dispute Aetna’s arithmetic about the post-merger price increase, but argues that it is more reasonable to calculate the increase on an annual percentage basis starting in 1996. RRB 44. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 403 Opinion of the Commission is where [Advocate] Lutheran General is, and . . . neither are they going to do that with Northwest Community Hospital.” TR 542 (Mendonsa), in camera. Mendonsa further testified that Aetna believed that it “couldn’t walk away” from ENH post-merger because it would have “devastated” Aetna and “shut down” its marketing to local employers. TR 518, 520 (Mendonsa), in camera. In addition, Mendonsa testified that before the merger Evanston was “extremely desirable” and that Aetna’s “walk-away point would have been pretty high . . . [but that Aetna] would have walked away[] because we still had Highland Park and we had Northwestern in the city and we had coverage.” TR 530 (Mendonsa), in camera. He also stated that “there probably would have been a walk-away point with the two independently. But with the two together, that was a different conversation.” TR 520 (Mendonsa), in camera. Aetna had terminated hospital contracts in the past when it had concluded that the prices were too high. TR 544 (Mendonsa), in camera. To do so with ENH, however, “would have killed [Aetna’s] marketing to any middle market or national accounts.” TR 530 (Mendonsa), in camera.

On cross-examination, Mendonsa testified that Evanston competed with Northwestern and Lutheran hospitals on tertiary services, and that Evanston also competed with St. Francis and Rush North Shore. TR 561 (Mendonsa), in camera. Mendonsa also testified that Highland Park competed with Lake Forest. TR 562 (Mendonsa), in camera.

We find that Mendonsa’s testimony that Aetna could have walked from Evanston pre-merger “because [it] still had Highland Park and . . . Northwestern in the city,” TR 530 (Mendonsa), in camera, and that “[s]omeone that’s going to Evanston is not going to to drive all the way out to Park Ridge, which is where [Advocate] Lutheran General is” located, TR 542 (Mendonsa), in camera, loosely suggests that Evanston and Highland Park were relatively close substitutes from Aetna’s perspective. His testimony that Evanston competed with Northwestern, Lutheran, St. Francis, and VOLUME 144 Opinion of the Commission Rush North Shore, and that Highland Park competed with Lake Forest, neither supports nor undermines complaint counsel’s case because it does not indicate whether competition from those hospitals could offset the loss of competition caused by the merger. merger.

3. One Health Patrick Neary testified on behalf of One Health, which today is called Great West. When Evanston and Highland Park merged, Neary was Director of Network Development and Provider Relations, and he negotiated One Health’s contract with ENH after the merger. TR 582 (Neary).

In December 1999, the month before the merger closed, Evanston contacted One Health to request the renegotiation of its contract. TR 594-95 (Neary). One Health’s previous contracts with Evanston and Highland Park were from 1996 and 1995, respectively. TR 596-97 (Neary). Bain had advised Evanston of what Bain believed was a “substantial difference” between One Health’s premerger rates at Highland Park and Evanston. CX 75 at 9-10. Neary testified that he thought that, after the merger, One Health was not in a strong negotiating position because he believed that Evanston had purchased “its main competitor” that “drew patients from the same general area.” TR 600-01 (Neary). Neary also testified that Advocate Lutheran General was “one of several” alternatives to ENH in 2000, along with St. Francis, Condell, and Northwestern Memorial. TR 631 (Neary).

Neary further stated “that it had been several years since the [Evanston Hospital] contracts had been renegotiated and that it was was appropriate to . . . increase some of the rates,” and One Health Health was willing to give a price increase based on an index. IDF ¶ 423; TR 608, 762-63 (Neary), in camera; CX 2085, in camera. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 405 Opinion of the Commission When ENH requested a larger increase than One Health thought was warranted after the merger, however, One Health and ENH failed to reach an agreement. One Health believed that ENH had proposed an increase of “26% to 219% of the current rate agreements in place.” CX 2085, in camera; TR 762 (Neary), in camera. One Health allowed the contract to lapse on August 31, 2000. TR 609-11 (Neary).

Neary testified that, shortly after its contract with ENH lapsed, One Health’s customers started to complain about their lack of access to ENH, and that One Health’s membership reports reflected a loss of membership. IDF ¶¶ 427-28; TR 615-17 (Neary); see also TR 1452, 1487-88 (Dorsey). At that time, One Health also had in its network Condell, Lake Forest, Northwest Community, Advocate Lutheran General, Rush North Shore, and St. Francis. TR 611 (Neary); TR 1459 (Dorsey). One Health ultimately agreed in the second half of 2000 to a contract with ENH that contained price increases that were “similar” to those in ENH’s initial proposal. TR 763-64 (Neary), in camera.

On cross-examination, Neary testified that Advocate Lutheran General, St Francis, and Condell were “several” main alternatives to ENH. In addition, he testified that Northwestern Memorial Hospital was also an “alternative” to ENH. TR 630-31 (Neary). Neary’s testimony that Evanston had purchased its “main competitor” and that One Health briefly had dropped ENH after ENH requested substantial price increases, and then entered into a contract with ENH at similar levels, provides some indication that pre-merger competition between Evanston and Highland Park prevented them from individually exercising market power. We assign only a small amount of weight to the testimony, however, because Neary provided less information about the substitutability of Evanston and Highland Park than did Ballengee and Mendonsa. Neary’s testimony about the existence of other “main” alternatives to ENH also lacks sufficient detail to allow for firm conclusions. VOLUME 144 Opinion of the Commission Kevin Dorsey also testified about One Health’s post-merger contract negotiations with ENH. Dorsey was employed at One Health from 1997 to 2003, first as a Director of Development and then as a Vice President. TR 1429-30 (Dorsey). Dorsey managed Neary and oversaw One Health’s post-merger negotiations with ENH. Dorsey testified that One Health did not play one hospital off against another in negotiations, that he believed that Lake Forest was Highland Park’s primary competitor, and that he viewed St. Francis as Evanston’s primary competitor. TR 1470-72 (Dorsey). Dorsey’s generalized testimony is not particularly informative because he supported it with only minimal supporting facts. TR 1470-72 (Dorsey).

4. Unicare Lenore Holt-Darcy testified for Unicare. TR 1412-13 (Holt- (Holt-Darcy). Holt-Darcy is a Unicare Regional Vice President. Id. Id. (Holt-Darcy). At the time of the merger, Unicare had both an HMO and a PPO contract with Evanston. The HMO contract had been negotiated in 1994 and contained a one-year term, with automatic annual renewals. CX 5085; CX 5091. Either party could could terminate the agreement with ninety days’ notice. CX 5091. The PPO contract had been negotiated in 1999. TR 1548, 1599, 1604-05 (Holt-Darcy), in camera; CX 216 at 12.19 In 2000, Unicare entered into contract renegotiations with ENH. ENH. Holt-Darcy testified that Unicare preferred to have rate 19 At the time of the merger, Highland Park also had a PPO contract with Rush Prudential, which was negotiated in 1994. CX 215; CX 5076. This contract also renewed annually, with each party having the right to terminate the contract with ninety days’ notice. CX 215 at 15. In 1998, Rush Prudential sought unsuccessfully to contract with Highland Park for its HMO plan. RX 392. While Highland Park did not have any contracts with Unicare before the merger, CX 114, Unicare acquired Rush Prudential in 1999. As a result, Unicare had access to Highland Park. Id.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 407 Opinion of the Commission increases below 10%, but if a hospital’s rates needed to “catch up,” the annual rate increase could exceed 10%. TR 1503 (Holt-Darcy), in camera. Holt-Darcy added that before the merger, Unicare could have developed a network with “adequate coverage” of the North Shore region with Evanston or Highland Park, and a combination of other hospitals. TR 1517-19 (Holt-Darcy), in camera. Unicare did not need both Evanston and Highland Park to “serve the geography.” Id. (Holt-Darcy), in camera.

The ALJ found that during the post-merger negotiations with Unicare, ENH officials stated that “[t]hey had sewn up” the North Shore suburbs for hospitals and physicians. IDF ¶ 455; see also TR 1546 (Holt-Darcy), in camera. The negotiations produced a contract on September 16, 2000, which contained substantial price increases. TR 1536, 1563-64 (Holt-Darcy), in camera. Holt-Darcy testified that the contract contained an 80% price increase in the rates that Evanston Hospital charged for Unicare’s PPO, TR 1539-40, 1563 (Holt-Darcy), in camera and that prices for Unicare’s HMO increased by 7%, 30%, and approximately 25% at Glenbrook, Highland Park, and Evanston Hospitals, respectively. TR 1543 (Holt-Darcy), in camera. Holt-Darcy also testified that Unicare had agreed to the substantial price increases because it viewed ENH as a “key provider,” and that not to have ENH in the network could have caused major employers, such as Kraft, to select other health plans. TR 1551-53 (Holt-Darcy), in camera. Holt-Darcy further explained that the ENH hospitals “had a contiguous service area that would have been hard, painful, for [Unicare’s] customers to see them leave.” TR 1602 (Holt-Darcy), in camera. On cross-examination, Holt-Darcy testified that Unicare does not not overtly play one hospital off against another during contract negotiations. TR 1593-94 (Holt-Darcy), in camera. She added, however, that it was not necessary to identify alternatives during negotiations because most hospitals know their competitors. TR 1602-03 (Holt-Darcy), in camera. Holt-Darcy also testified that Highland Park competes with Lake Forest and Condell Hospitals, VOLUME 144 Opinion of the Commission and that Evanston competes with a significant number of tertiaryservice hospitals in the Chicago area, including Rush North Shore, St. Francis, Loyola, University of Chicago, University of Illinois, and Northwestern Hospital. TR 1595-96 (Holt-Darcy), in camera. Similar to Mendonsa’s testimony, we find that Holt-Darcy’s testimony that Unicare could have developed a network with “adequate coverage” of the North Shore region with either Evanston or Highland Park, and a combination of other hospitals, TR 1517-19 (Holt-Darcy), in camera, loosely supports the inference that there was significant pre-merger competition between Evanston and Highland Park. Her testimony about the significance of the “contiguous service area.” TR 1602 (Holt-Darcy), in camera, of the ENH hospitals also suggests that Evanston and Highland Park were close geographic competitors, but because she offered relatively few specifics to support her testimony, we assign it only limited weight. Holt-Darcy’s testimony on cross-examination about competition between Evanston and Highland Park and other hospitals is not particularly probative because it did not explain whether and, if so, why this competition was sufficient to defeat a price increase by ENH.

5. United Jillian Foucre testified for United. Foucre worked at United from 1999 through 2004, and in August of 2001 became United’s Chief Operating Officer. TR 877-78 (Foucre). Foucre managed a team that negotiated with United’s network providers, including hospitals. TR 879 (Foucre).

United and ENH agreed on a new contract on January 1, 2000. TR 886-87 (Foucre). The new prices were substantially higher than than in United’s prior contract with Evanston. United’s documents show that it believed that ENH’s reimbursement rate (on allowed dollars per day basis) increased by 65.1% from 1999 to 2000, and by by 28.7% from 2000 to 2001. TR 1076-78 (Foucre), in camera; CX EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 409 Opinion of the Commission CX 21 at 9, in camera. Foucre was not involved in the negotiation of the 2000 contract. The United employee who was responsible for the negotiations was deceased at the time of trial. TR 887 (Foucre). In 2002, United analyzed ENH’s prices, concluded that they were higher than United’s average hospital reimbursement rate, and and sought to renegotiate them. TR 888, 890 (Foucre).20 United decided that it could not afford to drop ENH because when you look look at the three hospitals that make up the Evanston Northwestern Northwestern Healthcare system and look at . . . the triangle that they create, that area of Chicago . . . is very heavily populated by some of the most affluent communities in the Chicago area, and a result of that, the senior executives and the decision-makers of not only our existing customers but also our prospective customers would be residing within that area, and because, while there might be hospitals to the south and to the north, there are no other facilities, it did not seem feasible that we could have a viable network without Evanston Northwestern Healthcare. TR 901-02 (Foucre). Consequently, United did not believe that it could satisfy its customers without ENH, IDF ¶ 408; TR 901-02, 925-26 (Foucre), even though Lake Forest, Rush North Shore, St. Francis, and several other nearby hospitals were in its network at the time. IDF ¶ 408; TR 931-34 (Foucre).

Foucre testified that United was sufficiently concerned about ENH’s price levels that she met with local large employers, including Kraft, to discuss them. TR 904 (Foucre). The customers advised Foucre that they did not believe that it was feasible to remove the ENH hospitals from the network. TR 905-06 (Foucre). In In May 2003, United arranged a meeting between a number of local local employers and ENH officials to discuss the pricing levels. TR 20 United had three objectives for the 2002 negotiations: (1) change the format of the contract; (2) increase the percentage of the total revenues that it paid to ENH on a per diem basis and reduce the percentage that it paid pursuant to discount off charges terms; and (3) reduce its total payments under the contract. TR 892 (Foucre).

VOLUME 144 Opinion of the Commission TR 908 (Foucre). In 2004, United and ENH agreed to a contract that reduced ENH’s rates but, in United’s view, did not eliminate ENH as an outlier in terms of its prices. TR 1103 (Foucre), in camera. On cross-examination, Foucre testified that she viewed Condell and Lake Forest as the primary competitors to Highland Park, and that Evanston competes with Advocate Lutheran General, Rush North Shore, and St. Francis. TR 942-44 (Foucre). She testified that with respect to Evanston, “Lutheran General is the most comparable facility from type of services, quality of services, [and] size of facility; however, it is the furthest away. It’s got a bit of geographic disadvantage, but it’s not terribly far away.” TR 944 (Foucre). Foucre’s testimony that hospitals to the north of Highland Park and the south of Evanston were less desirable to residents of the North Shore suburbs suggests that the geographic proximity of Highland Park and Evanston made them close competitors, but because the testimony lacks detail we assign it only modest weight. Foucre’s very general testimony that Evanston and Highland Park competed with other hospitals, by itself, is not particularly informative.

K. ENH Officials’ Testimony Two of ENH’s senior executives, Neaman and Spaeth, presented presented testimony about pre-merger competition among North Shore hospitals. Neaman testified in general terms that he did “[n]ot “[n]ot really” view Highland Park as a competitor to Evanston because Evanston was “a lot bigger than Highland Park . . . [and] offered a much broader array of services.” TR 1306-07 (Neaman). He did not explain in detail, however, why Evanston and Highland Highland Park were not close competitors for the large number of primary and secondary services that they both provided. Accordingly, we find that Neaman’s testimony is not probative as to EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 411 Opinion of the Commission to the level of pre-merger competition between Evanston and Highland Park.

Spaeth testified that he considered Lake Forest Hospital to have have been Highland Park’s “primary competitor” before the merger merger because they are only six miles apart and have “major overlap” between their medical staffs. TR 2239, 2163 (Spaeth).21 Spaeth also testified that Highland Park competed with Evanston for for patients to the south of Highland Park, and that Evanston was competing for patients in Highland Park’s core area. TR 2157, 2241 2241 (Spaeth). In addition, he testified that, after Lake Forest, Evanston was Highland Park’s closest competitor: Q. Let’s talk about Highland Park’s closest competitors before the merger beyond the market share that we just looked at for your core area. Leaving aside Lake Forest, Evanston was Highland Park’s next closest competitor before the merger, correct? A. Leaving aside Lake Forest? I believe they were, yes, they were among the next one or two competitors.

Q. They were the next closest competitor, correct? A. They probably were.

* * * * Lake Forest would be first because of the major overlap in medical staffs. There were probably 200- 200-plus physicians that were on each other’s staff. 21 Terry Chan, a Highland Park employee tasked with analyzing Evanston’s and Highland Park’s prices shortly before the merger, also testified that she viewed Lake Forest as Highland Park’s closest competitor because there are many physicians who are on the staff of both hospitals. TR 730 (Chan). VOLUME 144 Opinion of the Commission Then the next set of competitors clearly put Evanston – Q. Well, the next competitor clearly was Evanston, correct? A. Yes.

TR 2162-63 (Spaeth).

On cross-examination by Respondent’s counsel, Spaeth testified that he also viewed Lake Forest, Condell, Rush North Shore, Advocate Lutheran General, St. Francis, and the downtown Chicago hospitals (along with Evanston) as competitors to Highland Park because of their “reasonably close” geography and because “[t]hey are all certainly substitutable for Highland Park.” TR 2239-40, 2299 (Spaeth). This competition, he explained, allowed MCOs to “go down the street” to Highland Park’s competitors to find substitutes for Highland Park in their networks. TR 2299 (Spaeth). Lastly, Spaeth testified that Evanston and Highland Park did not offer similar services because “there is a vast difference between an academic medical center and a community hospital” and because Highland Park did not offer heart care, sophisticated neonatal care or pediatrics, major oncology surgery, or neurosurgery. TR 2285-86 (Spaeth).

We find that Spaeth’s testimony that Evanston competed in Highland Park’s core service area, and that Evanston was Highland Park’s closest competitor (after Lake Forest), indicates that Evanston and Highland Park were close competitors for some services for patients who lived to the south of Highland Park and to the north of Evanston. The fact that Highland Park did not provide heart care or sophisticated neonatal care or pediatrics is not inconsistent with the existence of substantial competition between the two hospitals for primary and secondary services.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 413 Opinion of the Commission L. Econometric Evidence Complaint counsel and respondent each presented extensive econometric evidence. Complaint counsel’s primary economist was was Deborah Haas-Wilson, Professor of Economics at Smith College. Respondent’s economists were Jonathan Baker, a Professor Professor of Law at American University and Senior Consultant at Charles River Associates Incorporated, and Monica Noether who was then the Vice President and Head of the Competition Practice at at Charles River Associates.22 The econometric analyses of complaint counsel and respondent were designed to determine whether ENH charged higher prices than the merging hospitals would have charged if the merger had not occurred, and, if so, whether the price increases were due to an increase in market power produced by the merger. To answer these questions, Haas-Wilson and Baker used a thee-step process to predict the prices that ENH would have charged had the merger not occurred. First, they calculated the amount of ENH’s post-merger average net price increases to MCOs. Their second step was a difference-in-differences analysis, which consisted of a comparison of ENH’s pre- to post-merger change in average net price to the preto post-merger changes in average net price for various control groups. Their third step was a series of linear regressions using the same control groups.

Haas-Wilson ultimately concluded that, coincident with the merger, average net prices increased by higher-than-predicted levels for four of the five MCOs in the following ranges:23 Aetna (21.3% to 32.5%); Humana (12.3% to 16.6%); United (75.3% to 93.2%); and Great West (25.1% to 39.5%). CX 6279 at 18-19, in camera; CX 22 Dr. Noether is currently the Head of the Litigation and Applied Economics Platform and Professor Baker is a Senior Consultant at CRA International, Inc., a successor corporation to Charles River Associates. 23 The ranges are due to variations in the measured increases across econometric specifications.

VOLUME 144 Opinion of the Commission 6282 at 6, in camera; TR 2619-31 (Haas-Wilson), in camera. The results were statistically significant. Id. For BCBS, Haas-Wilson found that ENH’s actual post-merger average net prices were not statistically-significantly higher than her predicted post-merger average net ENH prices. CX 6279 at 18, in camera. Haas-Wilson also estimated that there were market-wide, higher-than-predicted merger-coincident average net price increases of 11% to 18%. CX 6279 at 20, in camera. She concluded that these price increases were due to market power created by the merger because she believed that she had factored out, through empirical and non-empirical analyses, the effects of the most likely competitively-benign explanations for the price increases. TR 2451, 2657 (Haas-Wilson); TR 2586-88, 2645-48, 2698-2733 (Haas-Wilson), in camera. Baker also found substantial higher-than-predicted average net price increases in acute inpatient services of 9% or 10%. TR 4620 4645-46 (Baker), in camera; RX 2040 at 3, in camera; DX 7068 at 21, ¶ 47, in camera. Because respondent maintained that hospitalbased outpatient services were also in the market, Baker also performed the same calculation for both inpatient and hospital-based outpatient services combined. Baker estimated a higher-thanpredicted average net price increase of 11% or 12% for these services combined. TR 4602-03 (Baker); TR 4617-18 (Baker), in camera; DX 7068 at 21, ¶ 46, in camera. Baker testified that these estimates did not account for ENH’s learning-about-demand and for potential post-merger changes in quality. TR 4602-03 (Baker). We address these issues below.

We describe the details of Haas-Wilson’s and Baker’s analyses separately, explain how they were similar and how they differed, and then state the findings and conclusions that we draw from their work.

1. Haas-Wilson’s Empirical Analyses Haas-Wilson tried to determine whether any of the following ten factors caused a post-merger price increase by ENH: EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 415 Opinion of the Commission 1. increases in costs that also affected other hospitals in the Chicago area, TR 2482 (Haas-Wilson);

2. changes in regulation that also affected other hospitals in the Chicago area, TR 2483-84 (Haas-Wilson);

3. increases in hospital demand that affected other hospitals in the Chicago area, TR 2484 (Haas-Wilson); 4. increases in quality at ENH relative to other hospitals in the Chicago area, TR 2485 (Haas-Wilson);

5. changes in the mix of patients (i.e., the complexity and type of the cases at each hospital) at ENH relative to other hospitals in the Chicago area that resulted in greater “resource intensity,” and thus greater costs, TR 2485-86 (Haas-Wilson); TR 2594 (Haas-Wilson), in camera; 6. changes in the mix of customers to more Medicare/Medicaid patients at ENH relative to other hospitals in the Chicago area, TR 2486 (Haas-Wilson);

7. increases in teaching intensity (i.e., the number of residents and interns per bed) at ENH relative to other hospitals in the Chicago area, TR 2486-87 (Haas-Wilson); TR 2604 (Haas- Wilson), in camera;

8. decreases in the prices of outpatient services charged to MCOs, TR 2487-88 (Haas-Wilson);

9. ENH’s learning-about-demand for hospital services from Highland Park’s pricing data, TR 2488 (Haas-Wilson); and VOLUME 144 Opinion of the Commission 10. an increase in market power due to the merger, TR 2488-89 (Haas-Wilson).24 Haas-Wilson used four data sources to conduct her analyses: (1) commercial payor claims data from MCOs in the Chicago area (“payor data”); (2) data received from the consulting firm NERA; (3) data received from the FTC’s Civil Investigative Demand to ENH; and (4) data from the Illinois Department of Public Health (“Illinois data”). TR 2495-500 (Haas-Wilson). Because only the payor and Illinois data were sufficiently comprehensive for Haas- Wilson to perform her regressions – which is the critical part of her analyses – we limit our discussion to Haas-Wilson’s analyses of these two data sets.

a. Simple Price Change Statistic Haas-Wilson began her analysis by calculating a simple postmerger price statistic. TR 2489 (Haas-Wilson). (1) Price Changes Calculated from the Payor Data The payor data were relatively comprehensive. The data (a) covered a five-year period from 1998 to 2002, CX 6279 at 3, in camera;25 (b) included data for the three ENH hospitals and many, if 24 Haas-Wilson acknowledged that this was not an exhaustive list of potential explanations for the post-merger price increases at ENH. TR 2481 (Haas- Wilson). Other explanations would include: (a) an increase in demand at ENH relative to other hospitals, and (b) an increase in costs at ENH relative to other hospitals. TR 2681-82 (Haas-Wilson), in camera; TR 4650-53 (Baker), in camera.

25 The exception was the data from Aetna, which ended in August 2002. TR 2512 (Haas-Wilson), in camera. Additionally, it appears from DX 7010 at 1, in camera, that the data from Humana also ended in August 2002. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 417 Opinion of the Commission not all, of the other general acute care hospitals26 from the Chicago metropolitan area; (c) were at the patient level for each hospital, and included the date of admission, the date of discharge, and in many cases the diagnosis, the age, and the gender of the patient; (d) included “the (dollar) amount that the managed care organization reimbursed the hospital for the care of the patient,” TR 2496-97 (Haas-Wilson), and the “diagnostic [or diagnosis] related group [(“DRG”)] indicating the nature of the hospital service,” DX 7068 at 15, in camera;27 and (e) covered seven of the at least fourteen MCOs that appeared to have had contracts with the ENH hospitals; including Aetna, BCBS, Humana, and United. CX 6279 at 5, in camera.28 Haas-Wilson used the payor data for Aetna, BCBS, Humana, and United purchases. Collectively, these four MCOs accounted for greater than 70% of ENH’s MCO patients on a per case basis in 2002, see CX 6279 at 5, in camera, but “less than 60 percent of MCO payments to ENH.” DX 7068 at 8, ¶ 20, in camera. By how 26 The record does not appear to contain a complete list of the other hospitals covered by the payor data. Haas-Wilson described the data as including “information on . . . the care received at many hospitals in the Chicago area.” TR 2497 (Haas-Wilson). Later, she stated that her largest control group of hospitals “included all general acute care hospitals in the Chicago PMSA.” TR 2548 (Haas-Wilson), in camera. From this testimony, we can infer that the payor data included, at the very least, all general acute care hospitals in the Chicago PMSA. Haas-Wilson denoted the area covered by the hospitals as the PMSA. The Commission could not locate a definition of “PMSA” in the voluminous record, but presumably it stands for Primary Metropolitan Statistical Area. 27 Diagnosis Related Groups (“DRGs”) refer to a system created for Medicare used to classify patients into groups expected to require similar hospital resources. There are roughly 500 DRGs. TR 2594 (Haas-Wilson), in camera; TR 5912-13 (Noether).

28 For an unknown reason, the payor data for all four MCOs contained more mothers than babies for obstetrics cases, which Baker and Noether labeled the “missing babies” problem. TR 4625 (Baker), in camera; DX 7126 at 74, ¶ 184; id. at 103, ¶ 267. The record does not appear to indicate that Haas-Wilson addressed this issue. Baker and Noether dealt with this issue by implementing a correction and by omitting obstetrics cases (both mothers and babies) from some of their analyses. DX 7126 at 74-75, ¶¶ 185-186; TR 4628 (Baker), in camera; DX 7068 at 12, ¶ 29, in camera.

VOLUME 144 Opinion of the Commission much less than 60% is not specified in the record. Haas-Wilson also used payor data that covered Great West (formerly known as One Health).29 For the payor data, Haas-Wilson delineated the pre- and postpost-merger periods for each MCO by the date of its first contract renegotiation after the merger. TR 2511 (Haas-Wilson), in camera. camera. Consequently, each payor had different pre- and postpost-merger periods.30 Haas-Wilson appeared to construct a hospital service “net price” that consisted of the sum of (1) the payment from the MCO to the hospital, and (2) the payment from the patient to the hospital. TR 2496-97, 2576, in camera (Haas-Wilson).31 From this measure, she apparently then calculated, on a per-patient basis, (1) an average net price per case, and (2) an average net price per day. TR 2514 (Haas-Wilson), in camera. We believe that the price-percase metric is more relevant than the price-per-day calculations because presumably MCOs are more focused on their total cost for a for a procedure rather than the amount of time that it takes to 29 The Great West data included payments only from the MCOs to the hospitals but not from the patients to the hospitals, while the other payors included the total payments to the hospitals. TR 2576 (Haas-Wilson), in camera. Haas-Wilson testified that the Great West data “does not allow me to look at the total reimbursement to the hospital for inpatient care,” but she did not explain why she used the data. Id. (Haas-Wilson), in camera. 30 For United, the contract effective date was January 1, 2000, which is the same date as the merger. TR 2512 (Haas-Wilson), in camera. For Aetna, the contract effective date was June 1, 2000. TR 2512 (Haas-Wilson), in camera. The record does not appear to indicate the contract effective dates for BCBS, Humana, and Great West. According to DX 7010 at 1, in camera, however, it appears that the contract effective date for the BCBS HMO was July 1, 2000, and the effective dates for the BCBS PPO and the Managed Care Network Provider plans were January 1, 2001; Humana and Great West had contract effective dates of September 15, 2000, and January 1, 2001, respectively. 31 Baker and Noether also appear to have constructed a total “net” price, which consisted of the payment made by the payor to the hospital, and any payment made directly by the patient. DX 7126 at 76. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 419 Opinion of the Commission perform. Consequently, we report Haas-Wilson’s per case calculations in text, and her per day calculations in footnotes. Haas-Wilson calculated that ENH’s average net price per case increased post-merger for all five of the MCOs that she examined: Aetna (28% to 89%); BCBS (10% to 27%); Humana (27% to 73%); 73%); United (62% to 128%), CX 6279 at 3, in camera; and Great Great West (42%), CX 6282 at 5, in camera. The ranges of percentages reflect that Haas-Wilson performed the calculations for for multiple plans for the MCOs.32 (2) Price Changes Calculated from the Illinois Data The Illinois data also were relatively comprehensive. The data set contained data on all patients from all hospitals in Illinois for the the periods 1998-99 and 2001-02. TR 2500 (Haas-Wilson); CX 6279 6279 at 7, in camera. Unlike the payor data, which contained data for individual MCOs, the Illinois data set identified the payor by general categories of payment types: commercial insurance, self-pay, self-pay, or HMO, as well as others. TR 2532 (Haas-Wilson), in camera; CX 6279 at 7, in camera. The data also contained only list list prices from the chargemaster (i.e., gross payments) but not the net prices (i.e., negotiated prices or net payments). TR 2500 (Haas- (Haas-Wilson). Haas-Wilson dealt with this limitation by using Medicare cost reports to derive an estimate of the net prices. TR 2527 (Haas-Wilson), in camera. The Medicare cost reports contain aggregate data on both net payments and gross payments by 32 Haas-Wilson also calculated the increase in ENH’s average net price per day post-merger for each of the MCOs’ plans: Aetna (48% to 56%); BCBS (-12% (decrease) to 15%); Humana (57% to 82%); United (77% to 202%), CX 6279 at 3, in camera; and Great West (79%), CX 6282 at 5, in camera. VOLUME 144 Opinion of the Commission hospitals for inpatient and outpatient services. IDF ¶ 576.33 Haas- Wilson calculated the ratio of the net receipts of the hospitals to their gross billing amounts and then multiplied that ratio by the billing information in the Illinois data set to estimate the actual net price. TR 2529 (Haas-Wilson), in camera. Haas-Wilson testified that, while there is potential bias in such an approach, any bias would be small. TR 2529-30 (Haas-Wilson), in camera; IDF ¶ 579. For the Illinois data, Haas-Wilson calculated the post-merger increases in the average net price per case for three broad categories categories of patients: all patients (30%); commercial and self-pay patients (27%); and commercial, self-pay, self-administered, and HMO patients (26%). CX 6279 at 7, in camera.34 b. Difference-in-Differences Analysis Haas-Wilson correctly recognized that her calculations of the simple changes in average net price using the payor and Illinois data data sets did not demonstrate that the changes in net prices resulted resulted from post-merger market power because they did not control for other factors that might explain the increases. TR 2540- 2540-41 (Haas-Wilson), in camera. Haas-Wilson’s second step was was to use a difference-in-differences (“DID”) analysis to attempt to control for her first three competitively-benign potential causes of the price increases: changes in cost, demand, and regulation 33 “The Medicare cost reports are reports that hospitals are required to file with the Federal Government if they receive Medicare dollars.” TR 2527 (Haas- Wilson), in camera. They include “information on both net payments and gross payments by hospital for inpatient and outpatient services.” TR 2529 (Haas- Wilson), in camera. The reports also appear to have included “the percent of patients receiving care at [each] hospital that are covered by Medicaid or the percent of patients at [each] hospital that are covered by Medicare.” TR 2600 (Haas-Wilson), in camera.

34 The average net per day price increases were: all patients (48%); commercial and self-pay patients (46%); and commercial, self-pay, selfadministered, and HMO patients (46%). CX 6279 at 7, in camera. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 421 Opinion of the Commission common across both ENH and her control groups. TR 2542-44 (Haas-Wilson), in camera. The DID analysis consisted of a comparison of ENH’s pre- to post-merger change in average net price to the pre- to post-merger changes in average net price for each each of three control groups. TR 2546-47 (Haas-Wilson), in camera; camera; DX 7027 at 1. Haas-Wilson compared the average percentage changes in ENH’s prices to those of the control groups because hospitals are differentiated and thus a simple cross-section cross-section comparison of price levels may be less informative. TR TR 2492-95 (Haas-Wilson). (Baker also measured percentage changes in ENH’s prices. RX 2040 at 1-3, in camera.) The reasoning underlying this approach was that changes in cost, demand, and regulation probably had a simultaneous and equal impact on the net prices charged by the ENH hospitals and hospitals that were similarly situated. If so, and if the control groups were reasonable and there were no other changes, her DID analysis enabled her to factor out the influence of the three competitivelybenign variables. TR 2548 (Haas-Wilson), in camera. Haas-Wilson used three control groups for her DID analyses: (1) (1) all general acute care hospitals in the Chicago PMSA;35 (2) all general acute care hospitals in the Chicago PMSA that were not involved in a merger between 1996 and 2002; and (3) all general acute care hospitals in the Chicago PMSA involved in some teaching teaching activity during the study period. TR 2548-49 (Haas- (Haas-Wilson), in camera. The purpose of using multiple control groups is that if results are consistent across a number of different econometric specifications, all other things equal, the regression analyses are more likely to be correct.36 35 As stated, “PMSA” presumably refers to Primary Metropolitan Statistical Area.

36 For more on specification tests, see G.S. MADDALA, INTRODUCTION TO ECONOMETRICS, Ch. 12 (2d ed. 1992).

VOLUME 144 Opinion of the Commission For the payor data, Haas-Wilson excluded hospitals with fewer than 100 admissions, during both the pre- and post-merger periods, for each payor plan.37 Consequently, a control group might be composed of different hospitals depending on the particular payor plan. TR 2557, 2560 (Haas-Wilson), in camera. Using the payor data, Haas-Wilson tried to use the DID analysis analysis to determine whether changes in cost, demand, or regulatory changes common across both ENH and her control groups explained all of ENH’s post-merger increases in average net net price. CX 6279 at 8-9, in camera; CX 6282 at 5, in camera; TR TR 2583 (Haas-Wilson), in camera. Haas-Wilson found that ENH’s ENH’s post-merger average net price per case increased, at statistically significant levels, for most of the payors’ plans by more more than that of the control groups: Aetna (30% to 73%); BCBS’ HMO and PPO (1% to 16%); Humana (5% to 53%); United (34% to to 113%); and Great West (13% to 27%). CX 6279 at 9, in camera; camera; CX 6282 at 5, in camera. The ranges reflect Haas-Wilson’s use of different control groups and plans.38 The only payor plan that 37 Haas-Wilson explained that she “selected only those hospitals that had more than 100 admissions in both pre- and post-merger periods” in order to “make sure that outlier admissions wouldn’t drive the result at any particular hospital. By 'outlier admission,’ [she] mean[t] an admission where the price, because someone had to stay an especially long time in the hospital, was extremely high, much higher than average.” TR 2556-57 (Haas-Wilson), in camera.

38 At varying levels of statistical significance, Haas-Wilson also found that ENH’s post-merger average net price per day increased for most payor plans beyond that of the control groups: Aetna (18% to 45%); BCBS’ PPO (1% to 4%); Humana (34% to 71%); United (43% to 167%); and Great West (57% to 58%). CX 6279 at 8, in camera; CX 6282 at 5, in camera. The only payor plans that did not experience at least one statistically significant increase in average net price per day beyond the control groups were BCBS’ HMO and POS plans, which experienced a 1% to 38% decrease. CX 6279 at 8, in camera. The Commission calculated these price changes from CX 6279 at 8 and 9, in camera, by subtracting from the figures in the “ENH” column either the “Chicago PMSA,” the “Non-Merging Chicago,” or the “Teaching Chicago” column. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 423 Opinion of the Commission did not experience at least one statistically significant increase in average net price per case beyond the control groups was BCBS’ POS plan, which experienced a 12% to 15% decrease. CX 6279 at 9, in camera.

Haas-Wilson’s application of the DID methodology to the Illinois data produced consistent results, as she again found statistically significant increases in ENH’s post-merger average net net prices beyond that of the control groups.39 The average net price per case increases for the three categories of patients were as follows: all patients (21% across all three control groups); commercial and self-pay patients (14% to 16%, depending on the control group); and commercial, self-pay, self-administered, and HMO patients (13% to 15%, depending on the control group). CX 6279 at 11, in camera.40 These results were all statistically significant at the one percent level. Thus, Haas-Wilson concluded that her DID analysis using the Illinois data also supported rejection rejection of the first three competitively-benign hypotheses for an increase in average net prices. TR 2585-86 (Haas-Wilson), in camera.

She also used the DID framework to test whether a decrease in in the average net price of outpatient services (her eighth potential alternative) was the cause of the substantial post-merger price 39 Again, Haas-Wilson excluded hospitals with fewer than 100 admissions during both the pre- and post-merger period. CX 6279 at 10-11, in camera. 40 These numbers were calculated from CX 6279 at 11, in camera, by subtracting from the figures in the “ENH” column either the “Chicago PMSA Control Hospitals,” the “Non-Merging Chicago Control Hospitals,” or the “Teaching Chicago Control Hospitals” column. The average net price per day increases for the three categories of patients were: all patients (34%, across all three control groups); commercial and self-pay patients (26% to 29%, depending on the control group); and commercial, self-pay, self-administered, and HMO patients (27% to 29%, depending on the control group). CX 6279 at 10, in camera. These numbers were calculated from CX 6279 at 10, in camera by subtracting from the “ENH” column either the “Chicago PMSA Control Hospitals,” the “Non-Merging Chicago Control Hospitals,” or the “Teaching Chicago Control Hospitals” column.

VOLUME 144 Opinion of the Commission increases at the ENH hospitals. TR 2607 (Haas-Wilson), in camera. The record indicates that MCOs negotiate the prices of both inpatient and outpatient services at the same time. Thus an MCO might agree to higher prices for inpatient services in exchange for reduced prices for outpatient services. Using the payor data only (because the Illinois data did not contain sufficient information on outpatient cases), Haas-Wilson found that the average net prices for outpatient services at ENH increased by at least as much as they did at hospitals in her control groups. CX 6279 at 17, in camera. These results, which were statistically significant, implied that the measures of the average net price changes for inpatient cases alone likely would understate the total post-merger price increases at ENH, if the case mix of outpatient services at ENH did not increase relative to the control groups. TR 2610-15 (Haas-Wilson), in camera.

Finally, Haas-Wilson used the DID method to determine whether whether changes in patient mix customer mix, and teaching intensity intensity were significantly different between ENH and the control control hospitals (i.e., her explanations five, six, and seven). CX 6279 at 13-16, in camera; TR 2594-2606 (Haas-Wilson), in camera. camera. Patient mix measured the complexity (and type) of the cases cases at each hospital, TR 2594 (Haas-Wilson), in camera;41 customer mix measured the percentage of patients receiving Medicare or Medicaid assistance at each hospital, TR 2600 (Haas- (Haas-Wilson), in camera; and teaching intensity measured the number of residents and interns per bed at each hospital, TR 2604 (Haas-Wilson), in camera. The results indicated that changes in patient mix customer mix, and teaching intensity differed significantly between ENH and the control hospitals and, therefore, therefore, that these factors could explain some of ENH’s post- 41 Haas-Wilson used four different measures for patient mix: All Patient Refined DRGs (“APRDRGs”), APRDRGs with a “length of stay” variable, DRG weights, and DRG weights with a “length of stay” variable. TR 2622-23 (Haas- Wilson), in camera.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 425 Opinion of the Commission post-merger price increases. TR 2607 (Haas-Wilson), in camera. Depending on the data set, the payor, the plan, and the control group, group, the percentage changes in case mix complexity at ENH differed substantially from those at the control group hospitals (from (from 9% below to 45% above).42 CX 6279 at 13, in camera. The average net price increases in the percent of patients on Medicaid and Medicare were greater at ENH (45% and 12%, respectively) than they were at hospitals in the control groups (30% to 34% and 7% to 8%, respectively, depending on the control group). Id. at 15, 15, in camera. The increase in teaching intensity was greater at ENH ENH (32%) than it was at hospitals in the teaching hospital control control group (8%). Id. at 16, in camera. Haas-Wilson’s finding that patient mix, customer mix, and teaching intensity differed between ENH and the control groups potentially invalidated her earlier use of the DID methodology to reject shared cost, demand, and regulation changes as explanations explanations for the post-merger price increases. This is because her her rejection of shared cost, demand, and regulation changes as explanations for the post-merger price increases was premised on the the ENH hospitals and the control groups having equivalent patient patient mix, customer mix, and teaching intensity.43 Nonetheless, as we now explain, this flaw does not invalidate Haas-Wilson’s ultimate conclusion because her linear regression results, which did did control for patient mix, customer mix, and teaching intensity, also implicitly eliminated shared cost, demand, and regulation changes as sufficient explanations for the post-merger price increases.

42 Again, the Commission calculated these percentages from the data on CX 6279 at 13, in camera, and CX 6279 at 14, in camera, by subtracting from the “ENH” column either the “Chicago PMSA Control Hospitals,” the “Non- Merging Chicago PMSA Control Hospitals,” or the “Teaching Chicago Control Hospitals” column.

43 Haas-Wilson’s DID analysis suffers from “omitted variable bias.” This problem occurs when a regression omits a relevant explanatory variable. See WILLIAM H. GREENE, ECONOMETRIC ANALYSIS 334-37 (4th ed. 2000). Her subsequent regressions demonstrated that the variables she omitted were relevant. VOLUME 144 Opinion of the Commission c. Linear Regression Analysis Haas-Wilson’s third step was to apply a linear regression model to test whether changes in patient mix, customer mix, and teaching intensity explained ENH’s post-merger increase in net prices. TR 2615 (Haas-Wilson), in camera; DX 7056 at 1. Regression is a statistical technique used to characterize the relationship between a variable of interest, such as price, and several other variables, such as changes in teaching intensity or increases in market concentration. Linear least squares regression, one of the most common forms of regression, characterizes the relationship by (1) assuming that it can be expressed as a straight line and (2) choosing a line of “best fit” that minimizes the sum of the squared differences between the predicted values (those on the line) and the actual values of the variable of interest. See Maddala, supra note 36, Ch. 3. Haas-Wilson’s regressions tested for whether changes in patient patient mix, customer mix, and teaching intensity explained ENH’s ENH’s post-merger increases in net prices, and also implicitly, by using control groups, tested for whether market-wide changes in cost, demand, and regulation could explain the price increases. In Haas-Wilson’s regression model, net prices at ENH and the control control hospitals were the dependent variables, and patient mix, customer mix, and teaching intensity were included in the independent variables. TR 2619 (Haas-Wilson), in camera. Using both the payor data and the Illinois data, Haas-Wilson regressed ENH’s and the control groups’ per case net prices on patient mix, customer mix, teaching intensity, and a dummy variable for the merger.44 DX 7056 at 1; TR 2619-22 (Haas-Wilson), in camera. Haas-Wilson used the same three control groups of hospitals as with 44 Haas-Wilson’s exclusive use of per case prices for her regression model (rather than per day prices) can be inferred by referencing CX 6279 at 19, in camera, and CX 6282 at 6, in camera.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 427 Opinion of the Commission with her DID analysis. TR 2620 (Haas-Wilson), in camera. She used four different measures of patient mix, which she regarded as a specification test. TR 2622-23 (Haas-Wilson), in camera.45 She ran the regressions separately for the payor data and the Illinois data. TR 2621-22 (Haas-Wilson), in camera.

Tellingly, Haas-Wilson’s regressions for the payor data indicated that ENH’s actual post-merger average net prices, at varying levels of statistical significance, were higher than her predicted postmerger ENH average net prices for four of the five payors: Aetna (21.3% to 32.5%); Humana (12.3% to 16.6%); United (75.3% to 93.2%); and Great-West (25.1% to 39.5%). CX 6279 at 18-19, in camera; CX 6282 at 6, in camera; TR 2619-31 (Haas-Wilson); in camera. The ranges reflect Haas-Wilson’s use of different control groups and measures of resource intensity. For BCBS, Haas-Wilson found that ENH’s actual post-merger average net prices were not statistically-significantly higher than her predicted post-merger average net ENH prices. CX 6279 at 18, in camera. Haas-Wilson found similarly higher-than-predicted increases in ENH’s average net price using the Illinois data for all three categories of patients: all patients (13.2% to 17%, depending on the control group and measure of resource intensity); commercial and self-pay patients (11.1% to 17.0%, depending on the control group and measure of resource intensity); and commercial, self-pay, selfadministered, and HMO patients (11.9% to 17.9%, depending on the control group and measure of resource intensity). These results were statistically significant at the one percent level. Id. at 20, in camera. d. Learning-About-Demand/Changes in Quality 45 Again, Haas-Wilson used four different measures of patient mix: APRDRGs, APRDRGs with a “length of stay” variable, DRG weights, and DRG weights with a “length of stay” variable. TR 2622-23 (Haas-Wilson), in camera. VOLUME 144 Opinion of the Commission Haas-Wilson did not formulate empirical tests to evaluate Respondent’s position that Evanston’s learning about market demand from Highland Park’s pricing data and improvements to Highland Park (factors four and nine) were responsible for the substantially higher-than-predicted merger-coincident price increases. TR 2586 (Haas-Wilson), in camera. She testified correctly that devising an econometric model to test for the learning-aboutdemand hypothesis is very difficult. TR 2643-44 (Haas-Wilson), in camera. Haas-Wilson rejected Respondent’s learning-about-demand argument and Respondent’s argument about improvements to Highland Park based on other portions of the record. TR 2586-88, 2645-48, 2698-2732 (Haas-Wilson), in camera; DX 7046, in camera; DX 7047, in camera; DX 7057, in camera; DX 7058, in camera; DX 7060, in camera; DX 7061 in camera. We do not discuss this portion of Haas-Wilson’s testimony because, as discussed below, the Commission has determined, based on its own review of the record (including many of the portions that Haas- Wilson relied upon), that neither possibility is a plausible explanation for ENH’s higher-than-predicted merger-coincident price increases.

2. Baker’s Empirical Analyses Baker used the same basic methodology as Haas-Wilson to analyze the changes in ENH’s prices to MCOs against the price changes of various control groups. Significantly, like Haas-Wilson, Baker found substantial higher-than-predicted merger-coincident price increases for ENH.

Baker, however, differed from Haas-Wilson in how he organized organized the data and in limiting his analysis to the payor data. First, Baker calculated prices only on a per case basis, while Haas- Haas-Wilson used both per case and per day prices in the majority majority of her analysis. TR 4628-29 (Baker), in camera. Second, consistent with Respondent’s position that the relevant product market includes inpatient and hospital-based outpatient services, EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 429 Opinion of the Commission Baker used both inpatient and outpatient cases (together) to measure price, although, for comparison with Haas-Wilson, Baker also performed his analysis using only inpatient cases.46 TR 4620 (Baker), in camera; DX 7068 at 10, ¶25, in camera. Third, Baker and Noether defined the post-merger period as the time after January 1, 2000, while Haas-Wilson used each payor’s contract renegotiation date as the start of the post-merger period. TR 4635 (Baker), in camera; DX 7068 at 9, ¶23, in camera. Baker testified that using the date of the merger as the starting point of the post-merger period was a more accurate method of calculating the post-merger price increases. TR 4636-67 (Baker), in camera. Fourth, Baker and Noether analyzed the data at the payor level for Aetna, BCBS, Humana, and United, but testified only on the results averaged across all payors. TR 4621, 4631-32 (Baker), in camera; DX 7068 at 8, ¶20, in camera; id. at 10, ¶24, in camera. Baker testified that he preferred using the overall price changes because he believed that it was more reliable, and also more appropriate given that complaint counsel had alleged that the relevant market involved the entire managed care market. TR 4648 (Baker), in camera.

a. Simple Price Change Statistic To calculate the simple price change statistic, Baker used two different measures of price: (a) “ENH constructed,” which was used to examine price increases across Evanston, Glenbrook, and Highland Park Hospitals taken together; and (b) “ENH,” which was used to examine price increases only at Evanston and Glenbrook Hospitals but not Highland Park Hospital. TR 4633 (Baker), in camera; DX 7068 at 10-11, ¶26, in camera. 46 No data were available on the case mix of outpatient cases. Baker and Noether dealt with this issue by assuming that the case mix changes for both inpatient and outpatient cases were identical. TR 4642-43 (Baker), in camera. VOLUME 144 Opinion of the Commission Baker found that inpatient average net prices increased across all all four payors after the merger using both the “ENH” and “ENH constructed” measures. RX 2040 at 1, in camera; DX 7068 at 43, in in camera. Using the “ENH constructed” measure, Baker calculated calculated the following average net price increases by ENH: Aetna Aetna (35%); BCBS (13%); Humana (83%); and United (138%). RX 2040 at 1, in camera; DX 7068 at 43, in camera. Overall, the four payors experienced an average 42% inpatient price increase from ENH. RX 2040 at 1, in camera; DX 7068 at 43, in camera. Using the “ENH” measure, Baker calculated the following average average net price increases by ENH: Aetna (25%); BCBS (2%); Humana (60%); and United (140%). RX 2040 at 1, in camera; DX DX 7068 at 43, in camera. Overall, the four payors experienced an an average 29% inpatient net price increase from ENH. RX 2040 at 1, in camera; DX 7068 at 43, in camera.47 Baker did not report levels of statistical significance for any of these calculations. RX 2040 at 1, in camera; DX 7068 at 43, in camera. 47 Baker also performed these calculations omitting obstetrics cases because of the “missing babies” problem. The corresponding per case average net price increases for “ENH constructed” (i.e., Evanston, Glenbrook, and Highland Park Hospitals) were: Aetna (34%); BCBS (5%); Humana (84%); and United (111%), with an average net price increase across all four payors of 37%. RX 2040 at 2, in camera; DX 7068 at 44, in camera. The corresponding per case average net price increases for “ENH” (only Evanston and Glenbrook) were: Aetna (31%); BCBS (3%); Humana (82%); and United (124%), with an average across all four payors of 35%. RX 2040 at 2, in camera; DX 7068 at 44, in camera. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 431 Opinion of the Commission b. Difference-in-Differences Analysis Next, to control for factors that could change prices across all hospitals, Baker differenced ENH’s price change with a control group’s price change. Baker used a control group of eighteen hospitals that Noether selected. DX 7126 at 71, ¶ 174; TR 4637-38 4637-38 (Baker), in camera; DX 8039, in camera. Noether did not not explain with precision how she chose the eighteen hospitals, and and her list does not match any set of hospitals in any particular document or any particular industry standard. TR 6149-51 (Noether). Again, using the “ENH constructed” measure, after differencing, Baker found that ENH’s average net prices increased above those of the control group at Aetna by 26%, at Humana by 58%, and at United by 103%. Baker found that BCBS’ prices did not not increase. RX 2040 at 1, in camera; DX 7068 at 43, in camera. The combined average net price increase by ENH for all four payors payors was 25% above that of the control group.48 Using the “ENH” measure, Baker found average net price increases to three of the payors and a price decrease for BCBS. After After differencing, Baker found that ENH’s average net prices increased above that of the control group at Aetna by 16%, at Humana by 35%, and at United by 105%. Baker found that BCBS’ BCBS’ average net prices decreased by 11%. RX 2040 at 1, in camera; DX 7068 at 43, in camera. With the “ENH” measure, the average net ENH price increase for all four payors was 12% above above that of the control group.49 48 Again, Baker performed these calculations omitting obstetrics cases. The corresponding per case average net price increases above the control group were: Aetna (26%); Humana (61%); United (83%); and a decrease of 3% for BCBS. The corresponding average net price increase across all four payors of 23% above that of the control group. RX 2040 at 2, in camera; DX 7068 at 44, in camera. 49 The corresponding per case ENH average net price increases above the control group, omitting obstetrics, were: Aetna (23%); Humana (59%); United (96%); and a decrease of 5% for BCBS. The corresponding combined price increase for all four payors was 21% above that of the control group. RX 2040 at 2, in camera; DX 7068 at 44, in camera.

VOLUME 144 Opinion of the Commission Again, Baker did not report the statistical significance of any of these differences. Presumably, if the increases were not statistically significant, Baker would have had an incentive to disclose that information. Regardless, Baker’s DID results, like Haas-Wilson’s, are not particularly informative because (1) they do not account for the substantial changes that Haas-Wilson found in case mix, patient mix, customer mix, and teaching intensity between ENH and other hospitals and, (2) as explained below, his DID results differed substantially from his linear regression results. c. Linear Regression Analysis The third step in Baker’s analysis is probative. As with Haas- Haas-Wilson, Baker’s third step was a regression model, which he used to control for changes in cost, demand, and regulation common common to both ENH and hospitals in his control group. Because Baker was unable to adjust for variations in outpatient cases, he included only inpatient cases in his regression analysis. TR 4642 (Baker), in camera; DX 7068 at 15, in camera. Baker also used his his regression to control for a range of variables that could affect price, including a patient’s age, gender, length of stay, type of health health care plan, and hospital. DX 7068 at 16, ¶ 38, in camera.50 To control for changes in case mix, Baker estimated his model separately for each DRG and for each payor. Id., in camera.51 Baker then calculated a weighted average ENH net price change over all the DRGs. Id. at 17, ¶ 39, in camera. Significantly, from this regression model, Baker concluded that, relative to the control group, ENH’s inpatient average net prices increased by 9% or 10% more than the predicted level, depending on whether obstetrics cases 50 Differing slightly from his earlier work, Baker used the natural logarithm of prices, rather than the prices themselves, as the dependent variable. RX 2040 at 3, in camera; DX 7068 at 45, in camera. 51 Haas-Wilson used different specifications to control for changes in case mix. Presumably she could have used an identical specification but chose not to. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 433 Opinion of the Commission were included. Id. at 19-20, ¶ 43, in camera; RX 2040 at 3, in camera; DX 7068 at 45, in camera. For inpatient and outpatient cases combined, Baker found that average net prices increased by higher-than-predicted levels of 11% or 12%. TR 4602-03 (Baker); DX 7068 at 21, in camera. As before, Baker did not report statistical significance, which is very unusual for regression results. One can presume again that Baker’s results were statistically significant because, if the results were insignificant, Baker would have had a strong incentive to report this.

3. Summary and Findings of Fact Concerning the Econometrics We find that the econometric work of both Haas-Wilson and Baker supports our finding that the higher-than-predicted mergermerger-coincident increases in ENH’s prices reflect the exercise of of market power caused by the merger. The economic testimony is marked by both agreement and disagreement over the correct way to to estimate the price changes associated with the merger, but significantly for purposes of resolving this case, the results of the analyses differed very little.52 Every empirical analysis conducted by Haas-Wilson and Baker found higher-than-predicted mergercoincident increases in ENH’s average net price for Humana and United. All but one of the empirical analyses conducted by Haas- Wilson and Baker found higher-than-predicted merger-coincident increases in ENH’s average net price for Aetna. Nearly every empirical test found little or no unexplained merger-coincident average net price increase for BCBS.

In addition, Haas-Wilson’s calculation of average market-wide changes in net price for all payors and Baker’s calculation of the average net price increase for Aetna, BCBS, Humana, and United, 52 There are essentially four different regression analyses: Haas-Wilson’s regressions using the Illinois Data, Haas-Wilson’s regressions using the payor data, Baker’s regressions using the payor data, and Baker’s learning-aboutdemand regressions, described infra 43-45, which also relied on the payor data. VOLUME 144 Opinion of the Commission produced very similar results. As we discuss in our findings about Respondent’s “learning-about-demand” argument, only when Baker Baker used a particularly contrived and narrow control group of six six academic hospitals was he able to account for the mergermerger-coincident price increases.

Haas-Wilson’s work demonstrated that Aetna, Humana, United, and Great West likely experienced higher-than-predicted price increases as a result of the merger, while BCBS did not. TR 2501, 2540 (Haas-Wilson), in camera. Using the payor data, the magnitude of the estimated average net price increase beyond that of the control groups ranged from 12.3% to 93.2%, depending on the payor and the control group used for the regression. CX 6279 at 18-19, in camera. Using the Illinois data, Haas-Wilson estimated that the merger caused market-wide average net price increases of 11% to 18%. Id. at 20 in camera.

Baker obtained very similar results. Again, Baker calculated that, relative to his control groups, average net inpatient prices across the four MCOs that he examined increased by 9% or 10% more than the predicted level due to the merger. TR 4620, 4645-46 (Baker), in camera; RX 2040 at 3, in camera; DX 7068 at 21, ¶ 47, in camera. Respondent briefly maintains that the Commission cannot rely on the econometrics because they were based on “imperfect” data. RB 50. Data inevitably have some flaws, and Haas-Wilson acknowledged that the data were not perfect. TR 2496-500 (Haas- (Haas-Wilson). The question is whether the data are sufficiently reliable that they are suitable for analysis. Both Haas-Wilson and Baker relied heavily on the payor data, and the record indicates that that this reliance was sensible because the data were comprehensive comprehensive enough to permit sound analyses. The data contained contained information from many (if not all) of the acute care hospitals in the Chicago metropolitan area from 1998 to 2002, and included data from more than 70% of ENH’s MCO patients on a per EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 435 Opinion of the Commission per case basis in 2002. See CX 6279 at 5, in camera. Further, Baker obviously felt sufficiently confident about the data to use them for most of his econometric analysis. The fact that Haas-Wilson obtained similar results from her regressions using the Illinois data also suggests that the payor data, as well as the Illinois data, were sufficiently reliable to instill confidence in Haas-Wilson’s and Baker’s results.

We also reject Respondent’s argument that Haas-Wilson did not account for every possible factor that might explain ENH’s substantially higher-than-predicted merger-coincident price increases, such as increases in marketing or advertising. RB 57. Rarely is it possible to consider every imaginable factor that might cause a price increase, and that is not necessary to have confidence in the results of econometric analysis. The issue is whether Haas- Wilson (and Baker) took into account the factors that were reasonably likely to have caused the substantial post-merger price increases. We find that both of them ruled out those other factors with econometric analysis. Further, as we describe below, the parties’ documents and the balance of the record indicate that it is very unlikely that the higher-than-predicted portions of the price increases were due to other competitively-benign causes. It is true that neither Haas-Wilson nor Baker provided the Commission with their exact econometric model, and Haas-Wilson Haas-Wilson did not provide a full explanation of how she calculated prices from the Illinois data. If the results of the regressions were more mixed, this lack of detail might make us less less confident about their reliability. However, except for our rejection of a portion of the results obtained from Baker’s narrow “learning-about-demand” six-hospital control group (discussed below), we need not pick and choose among the economists’ various various regressions, or the data sources, because they all produced essentially the same result: there were substantial higher-thanhigher-than-predicted merger-coincident average net price increases, increases, and it is likely that a significant portion of these increases VOLUME 144 Opinion of the Commission increases did not result from the most likely competitively-benign causes. The consistent results of such a wide range of tests utilized utilized by both sides’ experts, combined with our other findings of of fact, warrant our finding that it is very likely that the unexplained unexplained portions of the merger-coincident price increases were were due to ENH’s exercising market power created by the merger. merger.

M. Learning-About-Demand Respondent vigorously maintains that ENH’s post-merger increases in the prices for Evanston Hospital were caused by ENH’s obtaining information about Highland Park’s prices during the due diligence process, rather than the exercise of market power. RB 47- 59. This information allegedly showed ENH that some of Evanston’s pre-merger prices were below those charged by Highland Park. RB 18 47-59. Respondent argues that because Evanston offered more “comprehensive and advanced” services than Highland Park, and because more advanced hospitals allegedly receive higher prices, Evanston concluded that its pre-merger prices were below competitive rates. RB 51. If Evanston had been charging competitive prices, respondent reasons, its pre-merger prices would have exceeded those charged by Highland Park. RB 50-51. As we discuss in our legal analysis, respondent does not cite any any case in which a party has argued that its price increases are an attempt to correct a systematic failure to charge competitive prices, prices, and our research has not produced any such authorities. As we also discuss, we need not resolve all of the doctrinal issues associated with Respondent’s argument because none of the four types of evidence offered by respondent to support the learninglearning-about-demand position indicates that Evanston was systematically charging below-competitive prices to MCOs before the merger. We also note at the outset of our analysis that Respondent’s learning-about-demand argument does not apply to the EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 437 Opinion of the Commission the merger-coincident price increases at Highland Park, which respondent appears largely to attribute to post-merger improvements improvements in quality at the hospital. RB 51. 1. ENH Officials’ Testimony Evanston’s Chief Operating Officer, Jeffrey Hillebrand, testified that from 1990 to 1998 Evanston’s strategy was to have a relationship with every health insurer, and that this goal affected Evanston’s negotiating style. TR 1835 (Hillebrand). He also testified that during the 1990s there were fewer financial pressures on hospitals; that Evanston had a target rate of return; and that “as long as we were able to achieve that, management and our board felt that whatever pricing we were getting was sufficient.” TR 1836 (Hillebrand). According to Hillebrand, Evanston did not renegotiate a number of its contracts for approximately five years before the merger, TR 1850 (Hillebrand), which purportedly resulted in Evanston’s “short-changing itself for years in negotiations with MCOs.” RRB 49.

Respondent assigns some of the responsibility for its allinclusive strategy with MCOs to Jack Sirabian, who was Evanston’s principal negotiator from 1990 to 2000 and in that position reported to Hillebrand. TR 5697-98, 5701 (Sirabian). Sirabian and Kim Ogden of Bain testified that Sirabian wanted to have Evanston included in every network, lacked negotiation experience and support staff, and “was not comfortable taking a tough stand.” TR 5697-98 (Sirabian); RX 2047 at 34 (Ogden). Respondent asserts that it hired Bain in 1999 to conduct an analysis of its contracts and assist it with the merger. Bain’s analysis analysis purportedly demonstrated that Highland Park had higher rates than Evanston for the majority of its MCO contracts. Hillebrand testified that he was surprised and “embarrassed” to learn learn this fact. TR 1853 (Hillebrand). Neaman similarly testified that that he was “shocked” by the purported price disparity between VOLUME 144 Opinion of the Commission Evanston’s and Highland Park’s prices. TR 1344-45 (Neaman). Based on Bain’s conclusions, ENH decided that it would use more aggressive negotiating tactics with MCOs, including risking being dropped from the MCOs’ networks. TR 1854-55 (Hillebrand); TR 1218 (Neaman).

The testimony by Hillebrand, Sirabian, and Neaman is not persuasive. First, respondent stated in the proposed findings of fact fact submitted to the ALJ that during the 1990s Hillebrand participated in negotiations with larger MCOs, such as BCBS, which which caused Sirabian to pay “closer attention” to these pre-merger pre-merger contract negotiations and resulted in contracts with higher prices. RFF ¶¶ 604, 757; RB 52; see also TR 1700, 1836 (Hillebrand). As a result, ENH did not need to impose a “relative post-merger price increase” on BCBS. RB 52. Thus, the import of Respondent’s argument is that Evanston allowed Sirabian to forgo millions of dollars for the better part of a decade for those contracts contracts that he negotiated alone, but charged market rates when Hillebrand (to whom Sirabian reported) participated in the negotiations. This argument and the supporting testimony lack credibility.53 Second, Neaman testified that Hillebrand was an effective negotiator, with a good understanding of the marketplace and Evanston’s relationships with health plans. TR 1220 (Neaman). 53 Respondent also claims that it needed to raise its prices in 2000 because it faced new financial pressures, including that the Balanced Budget Act reduced its revenues and MCOs began to exert increased negotiating pressure. RB 49 (citing RFF ¶¶ 106, 110, 624, 630-33, 637). These events were not unique to respondent; they affected many hospitals, including Highland Park. Respondent asserts in its proposed findings of fact that Highland Park also felt the impact of the Balanced Budget Act, RFF ¶ 632, and Neaman testified that Balanced Budget Act cuts had hurt Highland Park’s financial performance, TR 1137 (Neaman), in camera. Respondent does claim that the Balanced Budget Act hit hospitals like Evanston harder because it had more clinical lines of service and teaching programs. RB 16. Respondent, however, cites only a few lines of conclusory testimony in support of this assertion. RFF ¶¶ 628-29. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 439 Opinion of the Commission Neaman also testified that he never criticized Hillebrand about Evanston’s pre-merger contracts with health plans. TR 1220 (Neaman). While not dispositive, such testimony contradicts Respondent’s argument.

Third, for those contracts that Sirabian allowed to remain in effect for a number of years without renegotiation, the record indicates that it is equally plausible that the prevailing competitive environment would not have allowed Evanston to raise prices. Spaeth testified that during the 1990s, Highland Park had “multiyear, no change contracts” and that before the merger he did not see an opportunity for Highland Park to raise prices. TR 2182, 2172-73 (Spaeth). As the ALJ found, “[t]he fact that Highland Park executives were concerned about contract terminations pre[-]merger [if they raised rates] is illustrative of the competitive environment that existed before 2000 and stands in contrast to the [post-merger] actions of ENH officials who, given their competitive situation, were not constrained by such prospects in their renegotiations with managed care representatives . . . .” ID 166. Finally, Respondent’s learning-about-demand argument is difficult to square with Respondent’s position that Evanston was and and is a state-of-the-art hospital, with superior management, that consistently provided high-quality services. RB 7. Respondent maintains that despite these many attributes, Evanston could not set set prices at market levels for some MCOs. In contrast, respondent respondent maintains that Highland Park failed to address quality issues properly, provided poor services to the point that it was threatening patient safety, was in severe financial distress, and would have deteriorated without the merger. RB 8-9, 63-67, 69. Despite these alleged shortcomings, Respondent’s learning-aboutlearning-about-demand argument rests on the premise that Highland Park officials were proficient at setting the hospital’s profitmaximizing price. This logical discrepancy is not determinative, but VOLUME 144 Opinion of the Commission when viewed in conjunction with the totality of the other evidence, supports rejecting Respondent’s learning-about-demand position.54 2. Baker’s Learning-About-Demand Analysis Baker sought to show through econometrics that at least some portion of ENH’s post-merger price increases was due to ENH’s learning that it had under-priced the market. Baker’s work, however, partially undermines Respondent’s position. To test Respondent’s learning-about-demand position, Baker performed a regression analysis that was conceptually similar to the regression model that he used to measure the post-merger net price changes. TR 4665-67 (Baker), in camera; DX 7068 at 29-30, ¶¶ 60-61, in camera. The primary difference was that, in his first regression, Baker used an eighteen-hospital control group; his learning-about-demand regression used a control group that consisted of only six hospitals selected by Noether, which Noether termed an “academic” group. DX 7068 at 27-30, ¶¶ 58-61, in camera; RX 2040 at 4, in camera; DX 7068 at 46, in camera. Baker’s rationale was that the information on market demand that ENH had obtained from the merger would enable it to price up to but not above the average prices charged by this group of hospitals, which Noether claimed were peers to Evanston. TR 5993-6000 (Noether); DX 7068 at 27, ¶ 56-57, in camera.

Baker estimated the average difference in the net prices between Evanston and each of the six academic hospitals for each year, after controlling for variation in the mix of patients across hospitals. DX 7068 at 28-29, ¶ 60, in camera. Baker then calculated the weighted 54 There is also an overall lack of merit to respondent’s contention that some of the MCO witnesses supported its learning-about-demand argument because they agreed that some of their contracts with Evanston were outdated. As the ALJ correctly found, these witnesses also testified that they thought that ENH’s postmerger price increases far exceeded reasonable market price benchmarks. ID 172; IDF ¶¶ 392-456.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 441 Opinion of the Commission average across the six hospitals of these predicted average differences. Baker found that the average net price (combining the four MCOs in his sample: United, Humana, BCBS, and Aetna) at ENH did not exceed the predicted level as compared to the control group. TR 4809-11 (Baker), in camera; RX 2040 at 4, in camera; DX 7068 at 30-31, 46, in camera. In contrast to all of his prior results, Baker also reported the statistical significance of his results.55 Baker’s regressions with the six-hospital control group are not reliable, however, because, as the ALJ found, and we agree, the narrow academic control group is highly flawed. The academic control group consisted of Advocate Lutheran General, Advocate Northside, Northwestern Memorial, Rush-Presbyterian-St. Luke, Loyola, and the University of Chicago. TR 6000 (Noether). Noether selected her academic control group based on three criteria: teaching intensity (rate of residents to beds); number of staffed beds; and breadth of services (number of DRGs). IDF ¶ 808; TR 5993-95 (Noether). Noether included in her academic control group only hospitals with at least 370 DRGs, more than .25 residents per bed, and more than 300 staffed beds. IDF ¶ 808; TR 5993-95. Noether’s criteria appear to be somewhat arbitrary and designed designed to exclude a number of hospitals that likely are Evanston’s Evanston’s peers. The teaching intensity classification is consistent consistent with the Medicare Payment Advisory Commission’s (“MedPac”) provision that defines a “major teaching hospital” as a a hospital with “at least .25 residents per bed,” but the DRG and number of bed criteria are not based on any specific established industry metric. IDF ¶¶ 809, 814, 817. The control group also included four of the most expensive hospitals in the city. Northwestern Memorial, University of Chicago, Rush-Presbyterian- Rush-Presbyterian-St. Luke, and Loyola each had higher average 55 The Commission could not determine whether Baker’s learning-aboutdemand regression analysis included obstetrics cases. VOLUME 144 Opinion of the Commission reimbursement per case than did Evanston. RX 1912 at 147, 150, in camera. Conversely, the control group excluded less expensive hospitals that could handle most of the cases handled by Evanston. Evanston. IDF ¶ 819; RX 1912 at 60; id. at 147-52, in camera. Additionally, four hospitals in the control group had a higher breadth of services (i.e., number of DRGs) than did Evanston. IDF IDF ¶¶ 821, 824-25; RX 1912 at 44, in camera. Also, four of the hospitals performed significant numbers of solid organ transplants, transplants, and two of them treated a significant number of extensive burn injuries. TR 2702 (Haas-Wilson), in camera; DX 7058, in camera. Evanston did not provide either service. TR 1378 1378 (Neaman). Four of the six hospitals in the control group had a a substantially greater number of residents per bed (i.e., more teaching intensity) than did Evanston. RX 1912 at 60. At the time, Evanston had 0.3386 residents per bed, while Loyola University had had 0.6060 residents per bed, Northwestern Memorial had 0.5670 residents per bed, Rush-Presbyterian-St. Luke’s had .7606 residents residents per bed, and University of Chicago had 0.7938 residents per bed. IDF ¶ 827; RX 1912 at 60.56 Further casting doubt on Noether’s criteria for selecting the narrow control group is that her standards resulted in the exclusion exclusion of two hospitals – Louis A. Weiss Hospital and St. Francis Francis Hospital – that met the MedPac criteria for a major teaching teaching hospital but that, according to Noether’s calculations, charged average prices below those charged by ENH from 2000 to 2003. TR 5921-22, 6170-71 (Noether); RX 1912 at 60; id. at 148, 151, in camera. Similarly, Noether excluded a number of hospitals that had a higher case mix index than did ENH, which she calculated 56 Moreover, three MCO witnesses testified that Evanston was not an academic hospital. TR 621 (Neary); TR 1444 (Dorsey); TR 936 (Foucre). We find this testimony to be credible. Further, none of the Bain documents upon which respondent relies so heavily, infra 46-47, references these very high-end academic hospitals as the appropriate benchmark for Evanston’s prices. TR 2052- 58 (Hillebrand).

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 443 Opinion of the Commission charged average prices below those charged by ENH from 2001 to 2003. TR 6168, 6170-72 (Noether); RX 1912 at 25; RX 1912 at 148- 49, 151-52, in camera. These hospitals were Alexian Brothers Medical Center, Northwest Community Hospital, and St. Francis Medical Center. TR 6168, 6170-72 (Noether); RX 1912 at 26, 148- 49, 151-52, in camera.57 Even assuming that Baker’s and Noether’s narrow academic control group was valid (which it is not), Baker’s regressions still partially undermine Respondent’s argument because he computed that Evanston’s post-merger prices to both Humana and United were were significantly higher than he predicted they would have been without the merger. TR 4682-85 (Baker), in camera; RX 2040 at 4, 4, in camera; DX 7068 at 46, in camera. For Humana, Baker computed that in 2002 the net prices ENH charged were 21% higher higher than he predicted they would have been had the merger not occurred. RX 2040 at 4, in camera; DX 7068 at 46, in camera. For For United, the net prices ENH charged in 2002 and 2003 were higher by 35% and 29%, respectively. RX 2040 at 4, in camera; DX 7068 at 46, in camera.58 The results for United are particularly significant because respondent repeatedly cites United as its primary example of a contract under which ENH’s pre-merger prices were substantially below market. E.g., RB 52. 57 Noether also excluded from her academic control group a number of hospitals that she listed as “best practice competitors” in her expert report, including Hinsdale Hospital, Christ Hospital, and MacNeil Hospital. TR 6152 (Noether); DX 7126 at 50. Conversely, Noether included in the academic control group Loyola University Medical Center and Rush-Presbyterian-St. Luke’s Medical Center, which are not listed in the documents that she relied on in her report to identify Evanston’s competitors. TR 6153-54 (Noether). 58 The Commission computed the numbers through straightforward calculations of the percentage differences in rows 7 vs. 9 (Humana) and rows 10 vs. 12 (United) in RX 2040 at 4, in camera; or DX 7068 at 46 (Table 4), in camera. The calculations are the following: Humana in 2002 – 21% = ($9,683 - $7,993)/$7,993; United in 2002 – 35% = ($10,373 - $7,708)/$7,708; United in 2003 – 29% = ($11,479 - $8,906)/$8,906. RX 2040 at 4, in camera; DX 7068 at 46 (Table 4), in camera.

VOLUME 144 Opinion of the Commission Baker’s findings for Aetna and BCBS were more favorable to ENH. He found that the prices ENH charged to these two payors were not statistically higher than prices at the academic hospitals. These results are not informative, however, because of Baker’s use of the flawed narrow control group.

3. Comparisons of Evanston’s and Highland Park’s Prices Respondent also tried to support its learning-about-demand position by introducing evidence that purportedly showed that Evanston charged lower prices than those charged by Highland Park for a number of MCOs before the merger. We find that this evidence does not support Respondent’s argument.

First, it is not entirely clear whether respondent is correct that Evanston’s theoretical equilibrium price was systematically higher than Highland Park’s, particularly for the primary and secondary services that both hospitals provided. Highland Park is located in the wealthiest part of the North Shore suburbs. TR 320-21 (Newton). Several of the MCOs explained that it was important to include Highland Park in their networks because many high-level officials who selected their company’s health plans lived in Highland Park and wanted access to the local hospital. As One Health’s Neary explained:

[I]n my opinion, . . . Highland Park knew that they had these influential people who were living in their community who would not be satisfied with a network that didn’t have Highland Park in their . . . healthcare plan. So, Highland Park had that as basically negotiating leverage, and they were able to say that these folks want us in their network, so you need to contract with us at higher rates. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 445 Opinion of the Commission TR 605-06 (Neary). Given the MCOs’ desire to satisfy major corporate decision-makers, some MCOs may have been willing to pay Highland Park higher rates than they would pay to Evanston. Assuming that Evanston’s theoretical, equilibrium price was greater than Highland Park’s price, Respondent’s pricing evidence that purported to show that Evanston’s pre-merger prices were often below Highland Park’s prices is not persuasive. Noether compared Evanston’s and Highland Park’s 1999 per diem rates for a number of payors at a number of different hospitals. RX 1912 at 34, in camera. The results showed that Highland Park’s prices were higher than Evanston’s for all the payors except BCBS and Unicare. TR 6079, 6088 (Noether). Noether’s results are flawed because, as she argued (and we agree), price per case is likely a more meaningful measure of price than price per day. DX 7126-104. Also, Noether’s table does not report statistical significance. RX 1912 at 34, in camera. In addition, Haas-Wilson’s calculations showed that pre-merger pre-merger prices were higher at Evanston than they were at Highland Park. TR 2646 (Haas-Wilson), in camera; DX 7047 at 1, 1, in camera. Baker implicitly arrived at the same conclusion as Haas-Wilson. TR 4744-47 (Baker), in camera. As the ALJ found, Baker calculated the average percentage price increase following the the merger for four health plans – Aetna, BCBS, Humana, and United – using two methodologies: (1) comparing Evanston’s and Glenbrook’s pre-merger prices to the ENH post-merger prices; and and (2) comparing Evanston’s, Glenbrook’s, and Highland Park’s combined pre-merger prices (Baker’s “constructed prices”) to the ENH post-merger prices. IDF ¶ 795 (citing TR 4633 (Baker), in camera). The constructed price calculation (which includes the prepre-merger prices at Highland Park) showed a larger average postpost-merger price increase than his calculation for the price increase increase (both with and without obstetrics) for just Evanston and Glenbrook. RX 2040 at 1-2, in camera; DX 7068 at 43-44, in camera. It follows that because the post-merger price increases were were larger when Baker included Highland Park’s prices in his VOLUME 144 Opinion of the Commission calculations, Highland Park’s average prices were lower than the average prices at Evanston and Glenbrook before the merger. TR 4744-47 (Baker), in camera. And finally, ENH’s Sirabian testified that no more than one-third of Highland Park’s contracts had higher higher rates than those contained in Evanston’s contracts. TR 5717 5717 (Sirabian).

Respondent attempts to dismiss Haas-Wilson’s and Baker’s calculations on the ground that they were based on econometric analyses that controlled for various factors, such as case mix, rather than the nominal contract rates. RB 50. Respondent appears to argue that even if Evanston’s prices, when adjusted for these relevant factors, were higher than those charged by Highland Park, ENH and some MCOs believed that they were lower based on a review of the nominal “contract rates,” RB 50 (emphasis added), and therefore that Respondent’s merger-coincident price increases could not have been due to market power. This reasoning is unconvincing. Even if we assume that Evanston’s unadjusted prices were below Highland Park’s, ultimately, business decisions are made based on actual rather than nominal prices. For example, we would not expect job seekers to decide between various employment opportunities using only nominal (i.e., “unadjusted”) wages; rather, we would expect them to consider the quality of the work, training opportunities, potential bonuses, the number of vacation days, and other factors along with wages. Therefore, we find that the appropriate way to compare prices is by controlling for the appropriate variables, which is the approach used by Haas-Wilson and Baker. Respondent also argued that Bain concluded that Highland Park Park had higher prices for certain contracts. E.g., RX 652; RX 684; 684; RX 1995, in camera. In several of these documents, Bain compared what it described as the “non-adjusted” contract terms of of Evanston’s and Highland Park’s pre-merger contracts. RX 684 at at 6. As the ALJ correctly found, the actual revenues received by a hospital are a function of both the discount rate in the contract and EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 447 Opinion of the Commission the hospital’s chargemaster. IDF ¶ 789-93; ID 173. A hospital with a higher chargemaster can have a lower discount rate and still charge higher prices. IDF ¶ 789-93; ID 173. Further, even the Bain documents that purport to compare Evanston’s and Highland Park’s contracts on an “adjusted” basis do not identify with precision the methodology that Bain used to make this determination. RX 1995 at 8, in camera. Therefore, the pre-merger pricing analyses that Bain performed shed relatively little light on Respondent’s learningabout-demand position. Id.

Finally, respondent cites Terry Chan’s assessment about the relationship between Evanston’s and Highland Park’s prices before the merger. As the Highland Park employee tasked with analyzing the two hospitals’ prices shortly before the merger, Chan authored a September 24, 1999 memo that stated that Highland Park’s contract rates “seem[ed] to be higher” than those charged by Evanston, but she acknowledged that her analysis did not include “information on [Evanston’s] charges and case mix.” TR 715 (Chan); RX 620 at 1, in camera. Chan also qualified her subsequent assessment that if Highland Park had applied the rates contained in Evanston’s contract rates in the previous year, Highland Park would have earned approximately $5 million less in inpatient revenue and $8 million less in outpatient revenue, RX 625 at 8294; RX 674 at 17915, with the caveat that “[f]uture environments under [Evanston’s] pricing structure and case mix might yield different results.” RX 674 at 17915. Chan noted in other memos that the “gross” rates in Evanston’s chargemaster were expected to be higher than those in the chargemaster used by Highland Park, RX 663 at 016939, in camera, and that “ENH’s charge master . . . is expected to generate higher gross charges than [the] gross charges generated by Highland Park Hospital’s current chargemaster.” CX 1373 at 14, in camera. These observations suggest that the net prices charged by Evanston may have been higher than those for Highland Park. VOLUME 144 Opinion of the Commission 4. Comparison of Evanston’s Prices and Other Hospitals’ Prices Finally, Noether attempted to validate ENH’s learning-aboutdemand argument by comparing Evanston’s average pre- and postmerger price levels with the average prices of groups of what she termed “community” and “academic” hospitals. TR 5993 (Noether). Her premise was that if Evanston had learned that its prices were low coincident with the merger, she would expect Evanston’s price to move from the average community hospital price to the average academic hospital price. TR 6060 (Noether), in camera. Noether reported her results using graphical plots of prices. RX 1912 at 62- 75, 108-52, in camera. In Noether’s graphs, ENH’s price appeared to move closer to the academic average price for a number of payors, but not all payors.

Noether’s price comparisons are unreliable, however, because they use the flawed academic control group. Further, even assuming that the control group is reasonable (which it is not) and that Noether’s calculations are correct, standing alone they do not support Respondent’s learning-about-demand position because they are equally consistent with the post-merger exercise of market power by ENH.

5. Summary of Findings of Fact on Learning-About- Demand While no one type of evidence in the record is dispositive, we find that the totality of the record warrants rejecting Respondent’s position that ENH’s learning-about-demand explains the substantially higher-than-predicted merger-coincident price increases.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 449 Opinion of the Commission N. Post-Merger Improvements and Cost Reductions 1. Merger-Specificity As stated above, in addition to believing that the merger would would and did allow ENH to raise prices to MCOs, Evanston’s senior officials viewed the merger as an opportunity to achieve cost cost reductions and economies of scale in various clinical and administrative areas,59 and to provide an additional teaching site for Evanston and Northwestern University Medical School. CX 359 at 22. Highland Park officials saw the merger as an opportunity for an infusion of capital at a time when the hospital was experiencing reduced income. TR 1327-28 (Neaman); TR 2266 (Spaeth). Those officials also viewed Evanston as an experienced partner that could help Highland Park implement new programs and enhance existing services – in particular cardiac surgery and oncology. TR 2273-74 (Spaeth); CX 6305 at 7. At Highland Park’s insistence, the merging parties’ letter of intent included specific commitments to implement these programs. RX 567 at 10, 12-13; CX 6305 at 9-10. Shortly after the merger, ENH established a cardiac surgery program at Highland Park and an interventional cardiology program program that supplemented Highland Park’s existing diagnostic cardiology program. IDF ¶¶ 952, 961. In mid-2000, ENH expanded expanded Highland Park’s existing oncology services by opening the the Kellogg Cancer Care Center at Highland Park, which provided a provided a multi-disciplinary approach to cancer care and brought together an array of oncology services in a single location. IDF ¶ 921. ENH also established a residency training program in family medicine at Highland Park, and obtained academic appointments at at Northwestern University Medical School for approximately sixty 59 Highland Park’s transaction counsel also advised Highland Park’s board that, although the merger might produce cost savings, “such savings are not the highest priority of the transaction” and “[t]he financial condition of both parties is such that neither require [sic] a financial reason for such affiliation.” CX 1923 at 2; TR 5840 (Kaufman); IDF ¶¶ 1039-40.

VOLUME 144 Opinion of the Commission Highland Park physicians, enabling them to participate in teaching activities (principally at Evanston). IDF ¶¶ 988, 990; TR 3124-25 (Romano), in camera.

In addition, ENH improved Highland Park’s physical facilities – – e.g., it constructed a new ambulatory care center, renovated the emergency department, and expanded the on-site laboratory – and upgraded some equipment. IDF ¶¶ 911-20, 929, 935-36, 941-43, 962 962 968. ENH also replaced all three hospitals’ existing electronic medical records systems with an integrated, entirely paperless computerized system called EPIC. IDF 976-81.60 All told, ENH spent approximately $120 million to make these changes at Highland Park. TR 1250, 1350 (Neary).

Complaint counsel and respondent each presented the testimony of a healthcare quality expert, who identified three widely recognized measures of quality: structure (e.g., facilities, staffing), process (e.g., surgical procedures, medication regimens), and outcome (e.g., mortality). TR 2986-87 (Romano); TR 5143-45 (Chassin). ENH’s evidence focused principally on structural changes (as well as some process changes) made by ENH, which its expert, Dr. Mark Chassin, testified constituted quality improvements because they “increase the likelihood of desired health outcomes.” TR 5141 (Chassin). Complaint counsel’s expert, Dr. Patrick Romano, focused principally on outcome measures. For the most part, ENH did not endeavor to show that the claimed improvements have actually improved health care outcomes at Highland Park. The ALJ found that ENH did not present any quantifiable evidence that improvements at Highland Park enhanced competition, competition, ID 177, and that ENH failed to show that quality 60 EPIC is a software system for managing patient records for both hospitals and physicians. It includes a physician order entry system and clinical decision support systems. IDF ¶¶ 978-79.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 451 Opinion of the Commission improved across the combined ENH system (not just at Highland Park) and relative to other hospitals. ID 179-81. The ALJ found that that Highland Park could have achieved the vast majority of the claimed improvements without the merger. ID 182-92. Our findings of fact differ in some respects from those of the ALJ, but we agree with the ALJ that Highland Park could have made the large majority of the quality improvements asserted by ENH without the merger. The record shows, and we find, that Highland Park was considered to be an excellent community hospital before the merger. IDF ¶¶ 850-52; TR 2095-98 (Spaeth); TR 4382 (Dragon); TR 5087-88 (Ankin). Highland Park had plans in place to improve its quality and expand services further without a merger, including many of the same improvements that ENH credits to the merger. Highland Park planned, for instance, to develop a cardiac surgery program in affiliation with Evanston or another hospital. IDF ¶¶ 952-59. In fact, in early 1999, Highland Park and Evanston entered into an agreement to develop a joint cardiac surgery program at Highland Park, with the understanding that implementation of the program did not depend on a merger. IDF ¶¶ 958-59. This agreement was similar to the affiliation agreements that ENH has with two other community hospitals – Swedish Covenant and Weiss – where it currently runs successful cardiac surgery programs without a merger. IDF ¶ 957; TR 4442-44, 4527-28 (Rosengart). Highland Park planned to improve its interventional cardiology services by expanding the diagnostic capabilities of its existing cardiac catheterization lab and to provide emergent angioplasty with the planned cardiac surgery program. IDF ¶ 964. Highland Park also had plans to enhance its existing “center for excellence” in oncology by launching a joint comprehensive oncology program with an institution other than Evanston, without a merger. IDF ¶¶ 924-28. In late 1997, Evanston’s CEO wrote to Highland Park’s CEO:

Our interest and expertise in developing an oncology oncology program with you along the same lines as VOLUME 144 Opinion of the Commission with cardiac surgery, is extremely high . . . . [H]aving the proven track record of already expanding our oncology program at both Glenbrook Hospital and Swedish Covenant Hospital, we believe we have developed a successful model that could rather quickly be implemented at your institution. CX 1865 at 2. Evanston also considered partnering with organizations other than Highland Park. IDF ¶ 925. Additionally, Highland Park’s strategic plans in 1998 and 1999 identified plans to enhance clinical services in maternal/fetal health, orthopedics, surgical services, and behavioral services; to improve physician collaboration; to improve workflow in all departments with particular focuses on radiology, cardiology, laboratory, and physical medicine; and to utilize technology to expand access to information to physician offices. CX 1868 at 13, 16, 18; CX 1908 at 13-14, 18, 20; IDF ¶¶ 869-70. In March 1999, Highland Park’s finance committee approved a long-range capital budget of $43 million for investments in strategic initiatives and master plan items such as cardiology services, ambulatory services, oncology, assisted living, and facility expansion, and $65 million for hospital construction, routine capital, and information technology. CX 545 at 3; IDF ¶¶ 872-74.

Prior to the merger, Highland Park already had begun to make a a number of the improvements that ENH contends the merger produced. For instance, in early 1998, Highland Park initiated an effort to improve the quality of care provided in its obstetrics and gynecology (“OB/GYN”) department by inviting the American College of Obstetricians and Gynecologists (“ACOG”) to conduct an on-site review of its birthing center and make recommendations recommendations for improvements. Highland Park then undertook undertook a comprehensive effort to implement these recommendations and address the issues that ACOG had identified. identified. IDF ¶¶ 883-86 (citing TR 3152-54 (Romano), in camera); EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 453 Opinion of the Commission camera); TR 389-93 (Newton). Highland Park also began the process of improving its nursing staff by hiring new, more effective effective nursing leaders and initiating a comprehensive effort to train, retain, and reward its nurses, and to improve communications communications between nurses and physicians. IDF ¶¶ 908-10; TR TR 3746-49 (Krasner); TR 5479-80 (Chassin); CX 6265 at 19, 21, in in camera. Before the merger, Highland Park also undertook an internal review of its quality assurance and quality improvement programs to identify ways to enhance these programs. IDF ¶ 898. The resulting report laid out a number of planned initiatives for improvements, including some of the same types of improvements that ENH asserts were produced by the merger. RX 417.61 The record also shows that a number of the changes that ENH made at Highland Park after the merger merely reflect emerging trends in the industry, rather than benefits unique to the merger. IDF IDF ¶ 895 (quality assurance program); IDF ¶¶ 901-02 (quality improvement program); IDF ¶ 950 (decentralized dispensation of medication); IDF ¶ 973 (use of intensivists); IDF ¶ 983 (electronic medical records systems); TR 3840-41 (Silver) (in-house physician physician coverage in obstetrics departments). Further, since the time of the merger, there has been a growing consensus regarding how best to monitor and improve healthcare quality measures that Highland Park likely would have incorporated into its quality assurance and quality improvement programs had it not merged with with Evanston. TR 2998-99, 3003-04 (Romano). And while respondent criticizes pre-merger Highland Park for not having had an intensivist program or a completely paperless electronic medical 61 For example, the report included recommendations that Highland Park consider developing an interdisciplinary steering committee to focus on operations and quality issues; develop a mechanism to improve reporting of adverse events; develop and adopt additional treatment protocols to address comorbidity and complications; review the quality tracking indicators used by Highland Park and identify critical indicators that ought to be tracked; and improve usage of national benchmarks. RX 417; see also CX 99 at 3 (outlining Highland Park’s plans, inter alia, to develop additional care maps that incorporate national benchmark data).

VOLUME 144 Opinion of the Commission medical records system, ENH’s decision to implement these programs at Highland Park was largely influenced by the publication, in 1999 and 2000, of recommendations by the Institute Institute of Medicine and the Leapfrog Group. TR 4065-66 (Wagner); TR 5079-87 (Ankin). In short, the record does not contain contain sufficient evidence to conclude that had it not merged with with Evanston, Highland Park could not, or would not, have been responsive to these emerging trends as well. In this respect, we disagree with the ALJ’s decision that ENH’s ENH’s installation of EPIC at Highland Park in late 2003 was a merger-specific quality improvement. Prior to the merger, Highland Highland Park was exploring ways to improve its information technology. CX 94 at 2-3; CX 1908 at 20. Although the ALJ concluded that Highland Park was unlikely, on its own, to have installed EPIC (in part because Highland Park already had an “excellent” electronic medical records system, and because, as a standalone hospital, it would not have had the same need as Evanston to integrate records from three hospitals, ID 190-91), Highland Park likely would have continued to improve its operations by investing in current information technology – if not EPIC, then through other appropriate systems.62 We find only one merger-specific improvement: the medical staff staff integration and affiliation with a teaching hospital. The record record shows that ENH physicians in several specialties now rotate rotate through all three hospitals, and that ENH facilitated faculty appointments at Northwestern Medical School for approximately 60 60 Highland Park physicians, who now participate in teaching activities at Evanston (for example, by giving “didactic lectures” to to medical students receiving their training at Evanston). IDF ¶¶ 62 Complaint counsel’s health care quality expert testified that there are no barriers other than cost for a community hospital to install EPIC, and some hospitals of similar size to Highland Park have partnered together to share the costs of installing and maintaining the system. TR 3162-63 (Romano), in camera. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 455 Opinion of the Commission 989-90; TR 3588-90 (Victor). The merger has not, however, transformed Highland Park (which has only one residency program, program, in family medicine) into a teaching hospital. IDF ¶¶ 988, 992. While studies have apparently shown that teaching hospitals have lower risk-adjusted mortality rates in certain clinical areas, there is no literature that shows that merely being owned by a teaching hospital is associated with improved quality of care. IDF ¶ ¶ 993; TR 3121-25 (Romano), in camera. ENH’s health care quality quality expert testified that the integration of medical staff and academic affiliation provides Highland Park physicians with greater greater opportunities to upgrade their skills and keeps them “on their their toes.” TR 5373-78 (Chassin). But this does not constitute verifiable evidence that any such improvement is of sufficient magnitude to offset the competitive harm that demonstrably has resulted from the merger.

2. Effect of Highland Park Improvements on Demand Respondent also maintains that some portion of the higher-thanpredicted merger-coincident price increases computed by both Haas- Wilson and Baker was caused by increased demand for Highland Park’s services due to post-merger improvements, and thus does not reflect the exercise of market power. RB 51, 58-59, 62, 72. We also find that the record does not support this assertion. Just as it is incorrect to conclude that nominal price increases by themselves reflect market power, it is also wrong to assume that nominal increases in quality are likely to lead to greater demand for the improved service. The relevant questions are whether Highland Park’s quality improved relative to that of other hospitals, and, if so, whether such above-market improvements increased demand for the hospital’s services.

As the ALJ found, quality of medical care is not easily defined or or measured, ID 179, and this difficulty is reflected in the differing differing approaches of complaint counsel’s and Respondent’s health care experts. The record is ambiguous as to whether Highland Park’s services improved more quickly than services at other hospitals in VOLUME 144 Opinion of the Commission the Chicago area. If they did, however, they likely did so by only a modest amount. The Joint Commission on Accreditation of Healthcare Organizations (“JCAHO”) regularly evaluates overall hospital quality nationally, including at Highland Park and Evanston, and JCAHO accreditation is necessary to qualify for Medicare, as well as for most managed care plans. ID 181. JCAHO assigns hospitals scores based on approximately 1200 elements of hospital performance. Id. In 1999, Highland Park received a preliminary score of 95 and a final score of 96. Id. In 2002, Highland Park received a JCAHO score of 94, a slight decline from 1999. Id. Further, as we have already found, a number of the post-merger changes at Highland Park reflect emerging trends in the industry. A comparison of the rate of Highland Park’s improvement to that of other hospitals is not critical, however, because even if the quality of care at Highland Park improved at a faster rate, the record does not support a finding that these improvements increased demand for Highland Park’s services. Again, hospital quality is difficult to measure, and demand for the services of one hospital compared to another is the product of a number of factors. Consequently, it does not follow that relative increases in the quality of one hospital always produce rapid increases in demand for that hospital’s services.

Here, the record indicates that relative demand for Highland Park’s services did not increase during the time period covered by the record. As the ALJ found, the record establishes that at the time time that ENH increased its prices, ENH did not mention that its price increases to MCOs were due to improvements at Highland Park. IDF ¶ 840. Hillebrand testified that he did not tell MCOs that that the substantial post-merger price increases were a function of improved quality at Highland Park. IDF ¶ 842; ID 178. Similarly, Neaman testified that he never saw any documents correlating the higher prices with the quality changes at Highland Park. IDF ¶ 843; 843; ID 178. Even after ENH implemented changes at Highland Park, ENH never identified any improvements at Highland Park to EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 457 Opinion of the Commission to MCOs (other than in a single press release). IDF ¶ 841-47; ID 178. We agree with the ALJ that if relative quality improvements were what drove ENH’s substantial post-merger price increases, logic suggests that ENH at least would have informed some MCOs MCOs on an individual basis about the improvements. ID 178. Such communications never occurred. The MCO representatives testified that the topic of quality improvements at Highland Park never came up during contract negotiations. IDF ¶¶ 844-47; ID 178. The MCOs also testified that they were not aware of a significant increase in quality at Highland Park after the merger. IDF ¶¶ 846-47, 851; ID 181. Additionally, many of the price increases were instituted in 2000, before some of the improvements were made. IDF ¶¶ 911, 916, 966, 981.

In short, we find that the record does not support the conclusion that Highland Park’s higher-than-predicted merger-coincident price increases were due to increased demand for Highland Park’s services relative to those offered by other hospitals. Rather, as the ALJ found, “the totality of the evidence strongly suggests that Respondent’s quality-of-care argument is a post hoc attempt to justify its postmerger price increases found to exist even by its own expert.” ID 179 (emphasis in original).

V. CONCLUSIONS OF LAW A. Section 7 of the Clayton Act Section 7 of the Clayton Act prohibits the acquisition of assets “in any line of commerce or in any activity affecting commerce in any section of the country, [where] the effect of such acquisition may be substantially to lessen competition, or to tend to create a monopoly.” 15 U.S.C. § 18. Congress used the phrase “'may be substantially to lessen competition’ to indicate that its concern was with probabilities, not certainties.” FTC v. H.J. Heinz Co., 246 F. 3d 708, 713 (D.C. Cir. 2001) (quoting Brown Shoe Co. v. United States, 370 U.S. 294, 323 (1962)). “Ephemeral possibilities” of VOLUME 144 Opinion of the Commission anticompetitive effects, however, are not sufficient. United States v. Marine Bancorp., Inc., 418 U.S. 602, 623 (1974). Merger enforcement is directed at market power. Heinz Co., 246 F.3d at 713; Merger Guidelines § 0.1 (“[M]ergers should not be permitted to create or enhance market power or to facilitate its exercise.”). The courts analyze whether a merger will produce or increase market power through the use of the now-familiar sequential approach. The plaintiff first establishes the relevant market, which itself consists of the relevant product and geographic markets. See United States v. Baker Hughes Inc., 908 F. 2d 981 982- 83 (D.C. Cir. 1990). Typically, the next step is to assess whether the transaction would produce a significant increase in concentration in the relevant market. Id. If the plaintiff makes such a showing, there is a structural “presumption” that the merger will substantially lessen competition. See Heinz, 246 F. 3d at 715; Baker Hughes, 908 F. 2d at 982-83. The burden of production then shifts to the defendant to produce evidence that shows that the market share statistics do not reflect the merger’s probable effects on competition. See Baker Hughes, 908 F.2d at 982-83. If the defendant successfully rebuts the structural presumption of illegality, “the burden of producing additional evidence of anticompetitive effect shifts to the government, and merges with the [government’s] ultimate burden of persuasion.” Id.

In practice, courts apply the burden-shifting paradigm by defining the relevant market, and then determining “the transaction’s transaction’s probable effect on competition in the product and geographic markets.” United States v. Sungard Data Sys., Inc., 172 172 F. Supp. 2d 172, 181 (D.D.C. 2001); see also Marine Bancorp., Bancorp., Inc., 418 U.S. at 618-23. In addition to examining evidence of existing competition between the merging parties and other firms, an integral part of the competitive effects analysis is determining whether new entry or expansion is likely to offset any reduction in competition between the merging firms. See FTC v. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 459 Opinion of the Commission Cardinal Health, Inc., 12 F. Supp. 2d 34, 58 (D.D.C. 1998) (citing Merger Guidelines § 3.0); FTC v. Staples, Inc., 970 F. Supp. 1066, 1086 (D.D.C. 1997).

If a court finds that a transaction is likely to produce a substantial reduction in competition that will not be averted by entry, courts generally consider whether efficiencies are likely to offset the reduction in competition. Although the Supreme Court held in FTC v. Procter & Gamble Co., 386 U.S. 568, 579 (1967), that “[p]ossible economies cannot be used as a defense to illegality,” subsequent lower court decisions have stated that “whether an acquisition would yield significant efficiencies in the relevant market is an important consideration in predicting whether the acquisition would substantially lessen competition.” FTC v. University Health, Inc., 938 F.2d 1206, 1222 (11th Cir. 1991); see also Cardinal Health, 12 F. Supp. 2d at 61. In University Health, 938 F.2d at 1223, the Eleventh Circuit held that a defendant could potentially overcome a “presumption that a proposed acquisition would substantially lessen competition [by] . . . demonstrat[ing] that the intended acquisition would result in significant economies and that these economies ultimately would benefit competition and, hence consumers.” The Merger Guidelines also recognize the role of efficiencies in determining the competitive effects of a transaction, stating that “[e]fficiencies generated through merger can enhance the merged firm’s ability and incentive to compete, which may result in lower prices, improved quality, enhanced service, or new products.” Merger Guidelines § 4.

Although the courts discuss merger analysis as a step-by-step process, the steps are, in reality, interrelated factors, each designed designed to enable the fact-finder to determine whether a transaction transaction is likely to create or enhance existing market power. See Baker Hughes, 908 F.2d at 984 (Section 7 inquiry is of a “comprehensive nature”). In the recently published Commentary on on the Horizontal Merger Guidelines (“Merger Guidelines Commentary”), the Federal Trade Commission and the Department VOLUME 144 Opinion of the Commission Department of Justice’s Antitrust Division emphasized “that the Agencies apply [] an integrated approach to merger review . . . [rather than] a linear, step-by-step progression that invariably starts starts with market definition and ends with efficiencies or failing assets.”63 Count I of the complaint alleges that the merger violated Section 7 of the Clayton Act in specified relevant product and geographic markets. Count II does not allege a particular relevant market; instead it alleges that the transaction violated the Clayton Act because the merger enabled ENH to raise its prices to private payors above the prices that the hospitals would have charged absent the merger. Under this count, complaint counsel maintains that it is not necessary to prove the relevant market because direct effects evidence shows that the transaction reduced competition substantially. CB 5. We first determine whether the record establishes that the transaction reduced competition substantially within a relevant antitrust market under Count I and then address complaint counsel’s thesis that it is possible to establish liability under Section 7 solely through the analysis of direct effects evidence under Count II.

63 FEDERAL TRADE COMM’N & U.S. DEP’T OF JUSTICE, COMMENTARY ON THE HORIZONTAL MERGER GUIDELINES 2 (2006), available at http://www.ftc.gov/os/2006/03/CommentaryontheHorizontalMergerGuidelinesM arch2006.pdf.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 461 Opinion of the Commission B. Defining the Relevant Market 1. Relevant Product Market The “boundaries of a product market are determined by the reasonable interchangeability of use [by consumers] or the crosselasticity of demand between the product itself and substitutes for it.” Brown Shoe, 370 U.S. at 325; see also United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 395 (1956). “Interchangeability of use and cross-elasticity of demand look to [1] the availability of products that are similar in character or use to the product in question and [2] the degree to which buyers are willing to substitute those similar products for the product.” FTC v. Swedish Match, 131 F. Supp. 2d 151, 157 (D.D.C. 2000) (citing du Pont, 351 U.S. at 393).

The Merger Guidelines use a related type of market definition test. Under the Guidelines, the product market is defined by asking whether a hypothetical monopolist of the proposed product market could impose a small but significant and nontransitory increase in price (“SSNIP”) and not lose an amount of its sales to alternative products that would make the price increase unprofitable. Merger Guidelines § 1.11. If so, then the proposed market constitutes a relevant product market. Id. The agencies often use a SSNIP amount equal to a 5% price increase, although this varies depending on the nature of the market. Id.; see Staples, 970 F. Supp. at 1076 n.8. The Merger Guidelines provide that “what constitutes a 'small but significant and nontransitory’ increase in price will depend on the nature of the industry, and the Agency at times may use a price increase that is larger or smaller than five percent.” Merger Guidelines § 1.11.

Courts are not required to follow the Merger Guidelines’ approach, but many modern courts have applied either the hypothetical monopolist test or some related test that defines markets by determining the set of products over which a dominant or or monopolist firm could exercise market power. See, e.g., United VOLUME 144 Opinion of the Commission States v. Microsoft Corp., 253 F.3d 34, 81 (D.C. Cir. 2001) (“To establish a dangerous probability of success, plaintiffs must as a threshold matter show that the browser market can be monopolized, monopolized, i.e., that a hypothetical monopolist in that market could enjoy market power.”); Coastal Fuels, Inc. v. Caribbean Petroleum Corp., 79 F.3d 182, 198 (1st Cir. 1996) (“The touchstone touchstone of market definition is whether a hypothetical monopolist monopolist could raise prices.”); Sungard, 172 F. Supp. 2d at 182, 186-92 (citing the Guidelines’ hypothetical monopolist test approvingly); Swedish Match, 131 F. Supp. 2d at 160-61 & n.8 (paraphrasing Merger Guidelines and informally applying the hypothetical monopolist test). The authors of the leading treatise also generally endorse the hypothetical monopolist approach. See II II Phillip E. Areeda, Herbert Hovenkamp & John L. Solow, Antitrust Antitrust Law ¶¶ 530a, 536, at 180-82 (2d ed. 2002).64 Complaint counsel asserts that the relevant product market is “general acute care hospital services, including primary, secondary, and tertiary services, sold to MCOs.” CB 37. Respondent argues, although not very strenuously, that the product market also includes “hospital-based” outpatient services because MCOs purchase both inpatient and outpatient services from hospitals. RB 26-27 & n.3. Respondent does not include non-hospital-based outpatient services in its relevant product market. Id.

The ALJ held that the record established that the relevant product market is that for acute inpatient hospital services, and we agree. ID 132-34. Current and former Evanston and Highland Park Park executives testified that ENH set inpatient rates independently independently of its outpatient rates and without concern that patients would switch to outpatient services. IDF ¶ 209; TR 330-31 330-31 (Newton); TR 1210-11 (Neaman). ENH’s Hillebrand 64 See generally Gregory J. Werden, The 1982 Merger Guidelines and the Ascent of the Hypothetical Monopolist Paradigm, 71 ANTITRUST L.J. 253 (2003). EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 463 Opinion of the Commission testified that he believed that inpatient hospital prices do not alter customer decisions to seek outpatient services because the physician physician makes that determination. TR 1755-56 (Hillebrand). Such Such pricing independence is strong evidence that the two sets of services are not in the same market because it suggests that there is a is a low cross-elasticity of demand between inpatient and outpatient outpatient services. Cf. United States v. Archer-Daniels-Midland Co., 866 F.2d 242, 248 (8th Cir. 1988) (finding that sugar and highhigh-fructose corn syrup not in the same product market in the absence of “evidence . . . demonstrating a high cross-elasticity of demand” between them).

Additionally, Noether testified that inpatient and outpatient services are not substitutes for patients and that MCOs cannot offer offer their patients outpatient services as a substitute for inpatient services when the patients need inpatient services. TR 6194 (Noether). Finally, the MCO witnesses who testified on the issue also stated that they could not, as a practical matter, substitute inpatient for outpatient services. TR 1422-23 (Holt-Darcy); TR 538- 538-39, in camera (Mendonsa); TR 591-92, 594-95 (Neary).65 Respondent’s position that outpatient services are in the market market is also inconsistent with all modern hospital merger cases. The courts have held repeatedly that acute inpatient hospital services services are a “cluster of services” that constitute a relevant product product market. See, e.g., FTC v. Freeman Hosp., 69 F.3d 260, 268 268 (8th Cir. 1995); University Health, 938 F.2d at 1211-12; United United States v. Rockford Mem’l Corp, 898 F.2d 1278, 1284 (7th Cir. Cir. 1990); United States v. Long Island Jewish Med. Ctr., 983 F. Supp. 121, 138-40 (E.D.N.Y. 1997); FTC v. Butterworth Health Corp., 946 F. Supp. 1285, 1290-91 (W.D. Mich. 1996). The rationale rationale is that while “the treatments offered to patients within this 65 Our descriptions of the testimony from Neaman, Hillebrand, Noether, Holt-Darcy, Mendonsa, and Neary are part of the Commission’s findings of fact. We did not include them in our findings of fact in Part IV only for ease of presentation.

VOLUME 144 Opinion of the Commission cluster of services are not substitutes for one another . . . the services and resources that hospitals provide tend to be similar across a wide range of primary, secondary, and tertiary inpatient services.” California v. Sutter Health Sys., 130 F. Supp. 2d 1109, 1119 (N.D. Cal. 2001). The record does not support our departing from this long line of cases.

Respondent argues incorrectly that complaint counsel’s “focus on MCOs as the consumers” warrants including hospital-based outpatient services in the market because MCOs simultaneously negotiate with hospitals for both inpatient and outpatient services. As the Seventh Circuit explained in Rockford Mem’l, the fact that a a customer purchases two sets of services from a supplier does not automatically lead to the conclusion that the two products are substitutes, or that one acts as a competitive constraint on the other. other. 898 F.2d at 1284.66 In short, we conclude that the evidence in the record establishes establishes that the relevant product market is acute inpatient hospital services. We also find that even if we included hospitalhospital-based outpatient services in the relevant product market, as 66 One could argue that there is no more substitutability between different types of inpatient services (e.g., a tonsillectomy and a heart transplant) then there is between inpatient and outpatient services, and that would certainly be correct. However, this does not justify including hospital-based outpatient services in the relevant product market, as respondent proposes. The record is not clear on the issue, but it is very likely that there are some types of outpatient services for which hospitals compete only with other hospitals, and other types of outpatient services for which hospitals compete with both hospitals and non-hospital providers. Respondent appears to agree because it limited the types of outpatient services that it included in its proposed product market to those provided by hospitals. RB 26-27 & n.3. If it were feasible to isolate the outpatient services that only hospitals provide, then it might make sense to define a broader “hospital services” product market, as respondent suggests. Such segmentation, however, is not practical here; nor is it necessary because, as the ALJ implicitly found, there plainly is a substantial volume of inpatient services for which neither hospitalbased nor non-hospital-based outpatient services are substitutes. IDF ¶¶ 206, 207, 209-11.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 465 Opinion of the Commission respondent proposes, it would not alter the outcome of this case. As we found above, both sides’ economists determined that ENH’s postmerger price increases for inpatient services were not offset by reductions (or smaller increases) in ENH’s prices for outpatient services. Baker actually calculated larger higher-than-predicted average merger-coincident net price increases for inpatient and hospital-based outpatient services combined (11% or 12%), than he did for inpatient services alone (9% or 10%). DX 7068 at 21, in camera.

2. Relevant Geographic Market The geographic market is “the ‘area of effective competition . . . in which the seller operates, and to which the purchaser can practicably turn for supplies.’” United States v. Philadelphia Natl Bank, 374 U.S. 321, 359 (1963) (quoting Tampa Elec. Co. v. Nashville Coal Co., 365 U.S. 320, 327 (1961)). The Merger Guidelines use the same hypothetical-monopolist approach to define the geographic market as they do for product market definition, stating that the relevant geographic market is a region in which a hypothetical monopolist could “profitably impose at least a ‘small but significant and nontransitory’ increase in price, holding constant the terms of sale for all products produced elsewhere.” Merger Guidelines § 1.21.

Complaint counsel asserts that the geographic market is the “geographic triangle formed by the three ENH hospitals.” CB 38. Respondent does not specify a precise geographic market but maintains that it is much larger.67 Whereas the north-south axis of 67 Respondent, of course, does not have the burden of proving the relevant product or geographic markets. Respondent cites a number of hospital merger cases in which the courts have defined geographic markets to include a county or several counties. RB 27-28. Precedent is a relevant consideration in defining markets, and we have partially relied on precedent to define the relevant product market. However, market definition fundamentally is a question of fact. This is particularly the case for geographic market definition, where population density, traffic patterns, and socio-economic factors vary substantially from region to region.

VOLUME 144 Opinion of the Commission complaint counsel’s market is approximately 13.7 miles, Respondent’s market has a north-south axis of at least 36 miles, and and includes hospitals such as Condell (approximately 13 miles north of Highland Park and 25 miles north of Evanston) and Northwestern Memorial (approximately 13 miles south of Evanston Evanston and 26 miles from Highland Park). RB 28-30. The record record is less clear about the respective lengths of the east-west axes axes of complaint counsel’s and Respondent’s geographic markets, markets, although it appears from a map in Respondent’s brief that that Respondent’s axis is at least approximately one-third longer. See See RB 29.

The ALJ defined the geographic market as the region covered by by the three ENH hospitals and four other hospitals – Lake Forest, Advocate Lutheran General, Rush North Shore, and St. Francis. The The ALJ found that “it is highly probable that the four non-ENH hospitals in the geographic market would have the ability to constrain prices at ENH, either now or in the future, and could be utilized by managed care organizations to create alternate hospital networks.” ID 144 (emphasis added). To the extent that the ALJ held held that MCOs could defeat a post-merger anticompetitive price increase by ENH by using one or more of these four other hospitals, hospitals, we reject this holding. Indeed, such a holding is inconsistent with the ALJ’s ultimate conclusion that the merger enabled ENH to exercise market power. Moreover, the ALJ’s opinion reflects that he made his conclusions about the geographic geographic market through rough inferences from the MCOs’ testimony and documents, and by making very general findings about driving distances. ID 142-43. The ALJ’s technique did not address the central issue in defining geographic markets – over what what geographic region could a hypothetical monopolist impose a SSNIP.68 68 We also find infirm the ALJ’s reliance on a portion of a survey conducted by Lake Forest Hospital about consumers’ willingness to travel for various types of hospital services, ID 142-43, because it is not possible to evaluate with confidence the survey’s reliability from the document alone. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 467 Opinion of the Commission As discussed above, some of the MCO testimony partially supports complaint counsel’s assertion that Evanston and Highland Park were close substitutes for some MCOs, and, therefore that the triangle formed by the ENH hospitals might constitute a geographic market. Standing alone, however, the MCO testimony was not precise enough to allow the Commission to draw firm conclusions. Conversely, the testimony from Respondent’s executives was not sufficiently detailed to conclude that the relevant geographic market is much broader than the market alleged by complaint counsel. Because it is not possible to define the geographic market solely through the testimony of the MCOs or Respondent’s executives, the question is whether the Commission can define the market based on the econometric evidence, which established that ENH could and did impose substantially higher-than-predicted merger-coincident price increases – 11% to 18% higher as computed by Haas-Wilson and 9% or 10% higher as computed by Baker. These price increases are larger than the 5% SSNIP that is often used under the Guidelines to define a market. See Merger Guidelines § 1.11. Respondent describes the concept of defining a relevant market market based on analysis of post-merger price increases as “circular” “circular” and a “tautology.” RRB 48-49. As we explain, defining markets based on such pricing evidence does not reflect a flawed circular analysis, but rather the fundamental relationship between market definition and competitive effects analysis in unilateral effects cases involving differentiated product markets. Complaint counsel, while alleging a geographic market, maintains that it is not necessary to define the relevant geographic market because, here, it is possible to show through direct evidence that the merger enabled ENH to exercise market power unilaterally. This argument, too, implicitly derives from the connection between market definition and competitive effects analysis in unilateral effects cases that VOLUME 144 Opinion of the Commission involve differentiated products. To explain, we turn to discussing unilateral effects analysis.

3. Market Definition and Unilateral Effects Modern merger analysis examines whether a merger is likely to lead to either or both coordinated and unilateral anticompetitive effects. Coordinated effects are reductions in competition caused by express or tacit interaction by the firms in a market, such as coordination on levels of price or output. See Merger Guidelines § 2.1. Generally, coordination is more likely in markets with homogeneous products because it is easier for competitors to reach agreement on the terms of coordination and to detect or punish deviations from those terms. See id. § 2.11. Determining that a merger has enabled the merged firm to raise prices does not necessarily aid in defining the relevant market in a coordinated effects case because the fact of the price increase may not readily enable the identification of the rivals in the market with which the merged firm is coordinating.

Unilateral effects are different. They result when a merger leads leads to higher prices due to the loss of competition between the two two merging firms, independent of the action of other firms in the market. See United States v. Oracle Corp., 331 F. Supp. 2d 1098, 1113 (N.D. Cal. 2004); Merger Guidelines § 2.2. There are a number number of different types of unilateral effects.69 Both complaint counsel and respondent agree that the type of unilateral effect that is relevant here is a reduction in competition in a differentiated product 69 The Areeda treatise classifies unilateral effects into four different types: “(a) creating a monopoly or dominant firm; (b) perpetuating a monopoly or dominant firm by eliminating a nascent rival; (c) giving one firm more secure control of its 'niche’ in a product-differentiated market; or (d) strengthening a firm’s power to make noncompetitive bids that buyers will be unable to refuse.” IV PHILLIP E. AREEDA, HERBERT HOVENKAMP & JOHN L. SOLOW, ANTITRUST LAW ¶ 910, at 55-56 (2d ed. 2006).

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 469 Opinion of the Commission market, meaning that the products under examination are not perfect substitutes for one another. See generally Merger Guidelines § 2.21. A merger between firms in a differentiated product market can enable the merged firm to raise prices unilaterally if customers accounting for “a significant share of sales” view the merging parties as their first and second choices for a particular need. Id. As As the agencies explained in the Merger Guidelines, anticompetitive anticompetitive unilateral effects occur when a sufficient amount of of the sales loss due to a post-merger price increase is diverted to the the product of the merger partner to make the price increase profitable. Id. Thus, whether a firm can profitably increase its prices prices unilaterally after a merger depends in part on the degree to which customers switch to the product of the other merged firm, as as opposed to switching to products of third-party firms. See id. § 2.21. The likelihood of unilateral effects in differentiated product markets also depends on the degree to which non-merging firms will will “reposition” their products post-merger to make them closer substitutes to those of the merging parties. Id. Unilateral effects are are less likely if other firms can quickly redesign or reformulate their products after a merger. See id.70 70 Thus, the Merger Guidelines provide that substantial unilateral price elevation in a market for differentiated products requires that there “[1] be a significant share of sales in the market accounted for by consumers who regard the products of the merging firms as their first and second choices, and [2] that repositioning of the non-parties’ product lines to replace the localized competition lost through the merger be unlikely.” Merger Guidelines § 2.21. The leading treatise contains a similar description of the factors relevant to assessing the likelihood of unilateral effects:

The degree to which a merger in a product-differentiated market might facilitate a unilateral price increase depends on (1) the relative “closeness” in product space of the merging firms to one another; (2) the relative distance between the post-merger firm’s product offering and the offerings of others in the market; and (3) the relative inability of other firms to redesign their products to make them close to the output of the merging firms.

VOLUME 144 Opinion of the Commission The portion of sales that constitute “a significant share of sales” sales” (and the number of customers that produce such sales) varies varies by market, and is a function of the relative closeness of the merging parties’ products or services, versus those of other competitors, and the relative margins of the merging firms. See IV Areeda, Hovenkamp & Solow, supra note 69, ¶ 914a, at 67; id. ¶ 914h, at 80-83; Merger Guidelines § 2.21. Notably, it is not necessary for the merged firms to b the closest substitutes for all customers, or even a majority of customers. IV Areeda, Hovenkamp Hovenkamp & Solow, supra note 69 ¶ 914h, at 82. Instead, what matters is that customers purchase enough of the merged firm’s products after a post-merger price increase to make the increase profitable. See id.; see also Merger Guidelines Commentary 27 (“A (“A merger may produce significant unilateral effects even though a a large majority of the substitution away from each merging product product goes to non-merging products.”)71 Because the focus of the analysis is on the unilateral loss of “localized” competition between the merging parties, there are substantial factual and analytical overlaps between the market definition process and competitive effects analysis in unilateral effects cases. Again, a market is the smallest possible group of competing products (or geographic area) over which a hypothetical hypothetical monopolist that sells those products (or competes in that area) could profitably impose a SSNIP. Merger Guidelines §§ 1.11, 1.21. Thus, if a merger enables the combined firm unilaterally IV AREEDA, HOVENKAMP & SOLOW, supra note 69, ¶ 914a, at 67. 71 See also Jonathan B. Baker & Carl Shapiro, Reinvigorating Horizontal Merger Enforcement 10 (June 2007), available at http://faculty.haas. berkeley.edu/shapiro/mergerpolicy.pdf (“[U]nilateral effects will arise so long as some customers of one of the merging firms consider its merger partner’s product as their second choice, even if more of the firm’s customers consider a third firm’s products to be their second choice.”). EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 471 Opinion of the Commission unilaterally to raise prices by a SSNIP for a non-transitory period due to the loss of competition between the merging parties, the merger plainly is anticompetitive, and the merging firms comprise a a relevant antitrust market because the merged entity is considered considered to be a “monopolist” under the Guidelines. As the authors of the leading treatise explain: In cases where a merger facilitates a significant “unilateral” price increase for a grouping of sales that was not a distinctive-looking market prior to the merger, the appropriate conclusion is that the merger has facilitated the emergence of a new grouping of sales capable of being classified as a relevant market. This formulation meets the statutory requirement that the effect of a merger is anticompetitive in some “line of commerce” and in some “section of the country.”

IV Areeda, Hovenkamp & Solow, supra note 69 ¶ 913b, at 64 (emphasis added);72 see also Gregory J. Werden, Simulating the Effects of Differentiated Product Mergers: A Practical. Alternative Alternative to Structural Merger Policy, 5 George Mason L. Rev. 363, 384 & n.97 (1997) (“If the products of the merging firms are next-closest substitutes for each other and the [merger] simulations simulations predict price increases of at least 5%, then the 72 The authors make the same point in the section of the treatise that discusses the criteria for identifying the likelihood that a merger will produce a unilateral price increase: “To the extent that . . . a merger enables the post-merger firm profitably to assess a significant price increase without losing sales to other firms, we would say that the merger facilitates the emergence of a new grouping of sales, or relevant market, in which the merging firms have either a monopoly or else a dominant share.” IV AREEDA, HOVENKAMP & SOLOW, supra note 69, ¶ 914f, at 77.

VOLUME 144 Opinion of the Commission Horizontal Merger Guidelines would support a market consisting of just the merging firms.”).73 The district court’s analysis in Staples is instructive. The district district court determined that office “superstores” constituted a relevant antitrust product market, relying heavily on its finding that that Staples’ and Office Depot’s pricing was disciplined more by the the presence of other superstores than by that of office supply stores stores generally. Staples, 970 F. Supp. at 1075-76, 1078. Staples’ prices were 13% lower in geographic markets where it competed 73 Respondent suggests that the government must show that the combined firm will have a dominant or monopoly share of the relevant market to establish that a merger is likely to cause anticompetitive unilateral effects in a differentiated product market. RB 37-38. This argument is incorrect, regardless of whether markets are defined through the Merger Guidelines’ approach, or by making general assessments about the functional substitutability of products or services. As Professor Baker explains:

[S]mall increases in concentration can generate higher prices in the localized competition model of mergers among sellers of differentiated products . . . . The reason: two brands may be close substitutes even if both have low market shares. Jonathan B. Baker, Unilateral Competitive Effects Theories in Merger Analysis, 11 ANTITRUST 21, 25 (1997) (emphasis added). Professor Baker made the same point in an article that he recently co-authored with Professor Carl Shapiro: the notion that “a plaintiff must demonstrate that the merging parties would enjoy a post-merger monopoly or dominant position [to raise prices unilaterally] . . . is incorrect and constitutes a clear error in economic reasoning.” Baker & Shapiro, supra note 71, at 10 (citations and quotations omitted); see also Gregory J. Werden, Simulating Unilateral Effects from Differentiated Markets, 11 ANTITRUST 27 (1997) (“. . . [C]ourts often delineate very broad relevant markets, yielding small market shares. But shares of these broad markets do not indicate what really matters – how often consumers of the product(s) of either merging firm view a product of the other merging firm as their next-best substitute, and how close other substitutes are in such cases.”). EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 473 Opinion of the Commission with Office Depot and OfficeMax than in markets where it did not face superstore competition.74 Id. at 1075-76. When the court turned to the competitive effects analysis, it looked at the same pricing evidence that it relied on to define the product market, explaining that “[m]uch of the evidence already discussed with respect to defining the relevant product market also also indicates that the merger would likely have an anti-competitive anti-competitive effect.” Id. at 1082. The court further explained that that “the evidence of the defendants’ own current pricing practices, practices, for example, shows that an office superstore chain facing facing no competition from other superstores has the ability to profitably raise prices for consumable office supplies above competitive levels,” id. (emphasis added), which also is essentially essentially the central issue examined in defining a relevant market. market. Logically, the court could have started its analysis by examining the transaction’s likely competitive effects, determined that competition between the firms reduced prices by more than a SSNIP, and then concluded that office superstores are a relevant product market under the Merger Guidelines.75 74 The data also showed that Office Depot’s prices were more than 5% higher in markets where it did not face superstore competition than in the markets where Office Depot competed with other superstores. Staples, 970 F. Supp. at 1077. 75 Practitioners have offered a similar assessment of the relationship between direct price-effects evidence and market definition in unilateral effects cases: If the Guidelines were not so wedded to the prima facie case developed for coordinated effects cases, the Division might have started with its econometric analysis. It might have argued that the combination of [Oracle and Peoplesoft] was going to raise price between 9.7 and 13.6 percent. The Division might have argued that the prima facie case is just an indirect means of proving the competitive effect it has established directly. So it is not really important to know the market definition and the market share. But if the court feels that it needs to have market definitions, those follow from the competitive effects. Marc G. Schildkraut, Oracle and the Future of Unilateral Effects, 19 ANTITRUST 24 (2005) (emphasis added).

VOLUME 144 Opinion of the Commission This case is somewhat different from Staples because prices in the hospital market are determined through bilateral bargaining. In bargaining markets, prices and other conditions of sale are set through individual negotiations between a buyer and seller. See Merger Guidelines Commentary 34. Because of the nature of the price-setting mechanism, bargaining markets can result in different prices for the same product, depending on the alternatives available to the negotiating parties.

Contrary to Respondent’s position, RRB 11, bargaining markets markets are quite common and fully consistent with unilateral effects theory. See Merger Guidelines Commentary 34-36. And most most economists who have recently studied the issue have concluded concluded that bargaining models are appropriate for hospital markets because bilateral negotiations between MCOs and hospitals hospitals determine prices that often are unique to the particular negotiation.76 The record in this case also demonstrates that hospital prices in the Chicago market are set through bilateral negotiations. CFF 245-83; TR 2470 (Haas-Wilson); TR 6189 (Noether); RB 51. The principles of unilateral effects analysis apply to bargaining bargaining markets, but their application is somewhat different in a a bargaining than in a single-price market. The unilateral exercise of of market power in a single-price market harms all customers because they each pay a higher price for the good or service. In a bargaining market, a merger may allow the merged firm to exercise exercise market power against a subset of customers who view the merging parties as their first and second choices, while the transaction will have no effect on other customers who do not view the merging firms as close alternatives or who have substantial “buy-side” market power. One or both of these possibilities likely 76 See generally Cory Capps et al., Competition and Market Power in Option Demand Markets, 34 RAND J. ECON. 737 (2003); Robert Town & Gregory Vistnes, Hospital Competition in HMO Networks, 20 J. HEALTH ECON. 733 (2001).

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 475 Opinion of the Commission explains, for example, why ENH appears to have been unable to exercise market power against BCBS after the merger. The potential for a merger in a bargaining market to have disparate effects on different customers potentially creates sticky and unsettled issues for merger analysis, most significantly, determining the percentage of a merged firm’s revenues that must come from customers who are harmed by the merger for the transaction to violate Section 7. The Commission need not delve into this issue in this case because, as we found above and discuss further below, the record demonstrates that the merger likely gave ENH sufficient market power to increase the average price that it charged to all MCOs.

We are mindful of the potential in both bargaining and nonnon-bargaining markets for defining overly narrow markets in cases cases involving differentiated products. “Demonstrating that the merging parties’ products are differentiated is not sufficient” to define a market, and there is a risk that “'localized competition’ analysis [will] devolve into an unstructured submarket-type analysis.” See Oracle, 331 F. Supp. 2d at 1119 (quoting IV Areeda, Areeda, Hovenkamp & Solow, supra note 69, ¶ 914a, at 60); see also du Pont, 351 U.S. at 393 (cautioning against viewing a manufacturer of every non-standardized commodity as having market power). At the same time, “a relevant market in an antitrust antitrust case may be smaller than a layperson would normally consider to be a market.” Oracle, 331 F. Supp. 2d at 1119; cf. Staples, 970 F. Supp. at 1074 (defining “office supply superstores” superstores” product market). Further, a set of products can constitute an antitrust market even when it is not possible to delineate a traditional “clean break” around the products or to devise devise a traditional market definition label. The keys to protecting against incorrectly narrow markets are, first, not to assume that a firm has economic power merely because its products are differentiated from those of its competitors; and, second, to ensure that the touchstone principle of market definition is satisfied: that the the degree of “product differentiation [is] sufficient to sustain a VOLUME 144 Opinion of the Commission small but significant and non-transitory price increase.” Oracle, 331 F. Supp. 2d at 1120.77 Thus, here, if complaint counsel has proven that the significant higher-than-predicted post-merger price increases resulted from market power gained through the merger, then complaint counsel has correctly defined the geographic market as the triangle formed by the three ENH hospitals. We turn now to the competitive effects analysis to determine whether the merger did enable ENH to exercise market power.

C. Competitive Effects Courts reviewing mergers pursuant to a Section 7 challenge assess the totality of the circumstances, weighing a variety of factors to determine the transaction’s effects on competition. See Baker Hughes, 908 F.2d at 984. We start our analysis with the extensive econometric evidence submitted by complaint counsel and respondent, and then discuss the other evidence. 1. Econometric Evidence 77 We are, of course, aware that some lawyers and economists have argued that the agencies and courts should focus solely on analyzing a transaction’s likely competitive effects, and not define markets, in unilateral effects cases involving differentiated products due to the fact that, viewed in isolation, market shares often are not always informative about the competitive proximity of the merging firms’ products. We also recognize that market definition can take on a conclusory quality in unilateral effects cases involving differentiated products because of the analytical and factual overlaps between the market definition and competitive effects analysis. See Jonathan B. Baker, Stepping Out in an Old Brown Shoe: In Qualified Praise of Submarkets, 68 ANTITRUST L.J. 203, 217 (1997). As we discuss in our treatment of Count II of the complaint, these and other considerations may justify holding at some point that it is not necessary to define a relevant market in certain Section 7 cases. Here, however, we need not decide this issue because, as we explain, it is readily possible to define the relevant product and geographic markets. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 477 Opinion of the Commission It is undisputed that ENH substantially and immediately raised its prices after the merger. Nominal price increases, however, do not by themselves establish the exercise of market power. Accordingly, as described above, Haas-Wilson and Baker sought to determine whether the post-merger increases were due to market power produced by the merger by calculating the amounts of ENH’s postmerger price increases, and then running a series of regressions to filter out the effects of the most likely competitively-benign factors that could have caused prices to rise after the merger. First, both Haas-Wilson and Baker found that ENH substantially substantially increased the actual prices that ENH charged to its customers. Haas-Wilson calculated, using the payor data, that ENH’s ENH’s average net price per case increased post-merger for all five five of the MCOs that she examined: Aetna (28% to 89%); BCBS (10% to 27%); Humana (27% to 73%); United (62% to 128%); and and Great West (42%). CX 6279 at 3, in camera; CX 6282 at 5, in camera.78 Using the Illinois data, Haas-Wilson calculated the postmerger increases in the average net price per case for three broad categories of patients: all patients (30%); commercial and self-pay patients (27%); and commercial, self-pay, self-administered, and HMO patients (26%). CX 6279 at 7, in camera. Similarly, Baker, using two different methods to calculate the price increases, found that ENH substantially raised its average net prices after the merger to the four payors that he examined: Aetna (25%, 35%); BCBS (2%, 13%); Humana (60%, 83%); and United (140%, 138%). RX 2040 at 4, in camera; DX 7068 at 43, in camera.79 The two different percentage amounts reflect that Baker used two different methods to calculate the price increases.

78 The ranges of the price increases for Aetna, BCBS, Humana, and United reflect that ENH raised prices by different levels for these MCOs’ various plans. 79 Baker also performed these calculations omitting obstetrics cases. The corresponding results were that ENH increased its prices to the four payors by the following amounts: Aetna (31%, 34%); BCBS (3%, 5%); Humana (82%, 84%); and United (124%, 111%). RX 2040 at 2, in camera; DX 7068 at 44, in camera. VOLUME 144 Opinion of the Commission Haas-Wilson next ran regressions using two data sources (the payor and the Illinois data) and three control groups, while Baker used only the payor data and two control groups. Although Haas- Wilson and Baker used different regression equations and different control groups, their calculations produced similar results. Haas- Wilson found, using the payor data, statistically-significantly higherthan-predicted post-merger ENH average net prices for four of the five payors: Aetna (21.3% to 32.5%); Humana (12.3% to 16.6%); United (75.3% to 93.2%); and Great West (25.1% to 39.5%). CX 6279 at 18-19, in camera; CX 6282 at 6, in camera; TR 2619-31 (Haas-Wilson), in camera. The percentage ranges reflect the use of different control groups and measures of resource intensity. For BCBS, Haas-Wilson found that ENH’s actual post-merger average net prices were not statistically-significantly higher than her predicted post-merger average net ENH prices. Haas-Wilson also found statistically-significantly higher-thanpredicted increases in average net price using the Illinois data: all patients (13.2% to 17%); commercial and self-pay patients (11.1% to 17.0%); and commercial, self-pay, self-administered, and HMO patients (11.9% to 17.9%). CX 6279 at 30, in camera. Again, the percentage ranges reflect the use of different control groups and measures of resource intensity.

Finally, Baker’s regressions found average net price increases of of 9% or 10% for the four payors that he examined, relative to his eighteen-hospital control group, depending on whether obstetrics cases were included. RX 2040 at 3, in camera; DX 7068 at 45, in camera; DX 7068 at 19-20, ¶ 43, in camera. In addition to the factors ruled out by Haas-Wilson, Baker’s model also controlled for for patient age, gender, length of stay, type of health care plan, and and hospital.80 80 As described supra 43-45, we find that Baker’s regressions using the narrow six-hospital academic control group are unreliable because the control group was not reasonable.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 479 Opinion of the Commission Because Haas-Wilson and Baker ruled out the most likely competitively-benign explanations for a substantial portion of the merger-coincident price increases, the size of the increases and the congruence of their results strongly suggest that the price increases were due to an increase in market power caused by the merger. As we found above, and discuss further below, the record does not support Respondent’s position that Evanston’s learning-aboutdemand or increased demand for Highland Park’s services as a result of post-merger improvements explains these portions of the mergercoincident price increases.

2. Documents and MCO Testimony The documentary evidence bolsters the conclusion that the higher-than-predicted merger-coincident price increases that both sides’ economists found were caused by market power produced by the merger. As both the ALJ and we have found, the merging parties’ documents reflect that a primary motivation of the senior officials in agreeing to merge the hospitals was to increase their bargaining leverage with MCOs in order to raise prices. The records of a January 4, 1999 meeting between Evanston’s and Highland Park’s board members and medical staff leaders state that Evanston representatives viewed the merger as an opportunity to not “'compete with self’ in covered zip codes (e.g., 60% to 70% market shares) such as Evanston, Glenview, Highland Park, and Deerfield,” CX 1 at 3, all of which are in the triangle. Similarly, the minutes of an April 5, 1999 meeting record an Evanston representative’s statement that the merger “would be an opportunity to join forces and grow together rather than compete with each other.” CX 2 at 7. After the merger, ENH’s Neaman tied the post-merger price increases in part back to greater negotiating leverage produced by the merger, telling the ENH board’s finance committee that “the larger market share created by adding Highland Park Hospital has translated to better managed care contracts.” CX 16 at 1. VOLUME 144 Opinion of the Commission The bottom-line conclusion of Highland Park’s Spaeth was that the way to “push back on the managed care phenomenon and get rates back to where they ought to be [was to become] 'big enough,’” at which point “it would be real tough for any of the Fortune 40 companies in this area whose CEOs either use this place or that place to walk from Evanston, Highland Park, [and] Glenbrook.” CX 4 at 2. It is difficult to imagine a clearer example of an executive using everyday language to explain how a merger will produce a firm that can exercise market power and whose services constitute a relevant antitrust market. Spaeth clearly thought that the merged firm would be able to raise prices because its customers would not be inclined to leave the ENH hospitals for other providers. Respondent’s efforts to downplay the significance of its documents are not persuasive. RB 59-62. The documents are probative because they reflect the merging parties’ unvarnished contemporaneous analyses of the parties’ market positions by their most senior officials. The statements are not simple bravado or unsubstantiated hyperbole from middle managers or sales representatives.

Respondent’s argument that “intent” does not establish a Section Section 7 violation is correct, but beside the point. RB 59-60. The documents are probative not because they reflect the desire of Neaman and Spaeth to raise prices, but because they contain the informed analysis of experienced executives about when, why, and how the transaction would enable the merged hospitals to increase prices. Antitrust courts frequently rely on such evidence. See, e.g., Cardinal Health, 12 F. Supp. 2d at 63-64 (relying on statements of senior executives that merger would reduce excess capacity and curb downward pricing pressures). We disagree with respondent that it does not “matter whether ENH executives later tied the merger to price increases.” RB 59. Antitrust courts often rely on the conclusions of senior executives about the goals and effects of their actions. See, e.g., Microsoft, 253 F.3d at 77 (“Microsoft’s internal documents and deposition testimony confirm both the EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 481 Opinion of the Commission anticompetitive effect and intent of its actions.”); University Health, 938 F.2d at 1220 n.27 (relying on evidence showing that the “appellees, by their own admissions, intend[ed] to eliminate competition through the proposed [hospital] acquisition”) (emphasis in original).

Respondent’s effort to expand upon the plain meaning of the documents also is not persuasive. Respondent argues, for example, that the merging parties’ use of the phrase “leverage” in one document was shorthand for seeking to obtain fair market value for their services. RB 61. Shortly before the merger, Evanston CEO Neaman told his managers and his board that the merger would “[i]ncrease our leverage . . . with the managed care players.” IDF ¶ 335; CX 1566 at 9 (emphasis added). This language reflects that Neaman thought that the merger would give Evanston additional bargaining power, not that the merger would allow Evanston to exercise bargaining leverage that it already possessed. Finally, we reject Respondent’s implied position that reliance on on the documents to infer anticompetitive effects is improper because the documents also indicate that the merging parties thought thought that the transaction would produce efficiencies. RB 60. Although some of the documents state that the merging parties thought that the merger would be efficient, this does not diminish the the fact that the documents also reflect the parties’ expectation that that the transaction would increase (and in their view that it had increased) the combined entity’s ability to raise prices. The exercise exercise of market power and the achievement of efficiencies are not not mutually exclusive or inconsistent.81 The MCO testimony also provides some (albeit modest) support support for the conclusion that the higher-than-predicted merger- 81 We analyze below whether the transaction enabled efficiencies and improvements that offset the anticompetitive effects of an increase in market power.

VOLUME 144 Opinion of the Commission merger-coincident price increases were due to market power, and it certainly is not inconsistent with that conclusion. The MCOs’ testimony suggests that they were reluctant to drop the ENH hospitals because they were highly desirable hospitals that served the North Shore suburbs. Aetna’s Mendonsa testified that he was concerned about the merger because it had resulted in “three extremely important hospitals negotiating together in a very important geography.” TR 530, 518 (Mendonsa), in camera. Similarly, United’s witness explained that the ENH hospitals were geographically significant because “when you look at the three hospitals that make up the Evanston Northwestern Healthcare system and look at . . . the triangle that they create, . . . it is very heavily populated by some of the most affluent communities in the the Chicago area . . . and because while there might be hospitals to to the south and to the north, there are no other facilities [within the the triangle], it did not seem feasible that we could have a viable network without Evanston Northwestern Healthcare.” TR 901-02 (Foucre). Unicare’s Holt-Darcy likewise testified about the strategic strategic significance of the “contiguous service area” covered by the three ENH hospitals. TR 1602 (Holt-Darcy), in camera. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 483 Opinion of the Commission 3. Respondent’s Positions Respondent offers a series of arguments as to why the Commission should conclude that factors other than market power caused the higher-than-predicted merger-coincident price increases. We address each argument in turn, and conclude that none of them is valid.

a. Learning-About-Demand Respondent’s primary rebuttal to the econometrics is its contention that a significant portion of the merger-coincident price price increases resulted from Evanston’s learning from Highland Park that Evanston supposedly was charging prices that were below below what respondent terms “the fully-informed competitive level.” RB 48. Respondent essentially is arguing that complaint counsel’s case reflects a “reverse” version of the Cellophane fallacy.82 Respondent, in essence, maintains that complaint counsel has defined the market too narrowly by applying a SSNIP to a price that is below the theoretical competitive level, and thus wrongly concluded that ENH’s ability profitably to impose such a price increase is due to market power.

Respondent cites no case to support its argument.83 Instead, respondent refers the Commission to a treatise and several articles 82 The “Cellophane fallacy” derives from the Supreme Court’s decision in United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377 (1956), in which the Supreme Court assessed the existence of market power by defendant du Pont by using as a baseline the existing supracompetitive price of a food wrap, rather than examining the profitability of a price increase from the baseline of a competitive price for the product. This analytical error caused the Court to find the absence of market power in a situation where the defendant already had been exercising market power.

83 Respondent asserts that the ALJ found that “Complaint Counsel failed to prove that ENH’s post-merger prices exceeded competitive levels” and that this finding is dispositive in respondent’s favor. RB 1 (citing ID 155). Respondent appears to be referring to the ALJ’s statement that “Complaint Counsel did not attempt to compare ENH’s price increases to a competitive level.” ID 155. As the VOLUME 144 Opinion of the Commission for the uncontroversial proposition that information about competitors’ prices can be costly to acquire, and as a result firms may not always price at fully-informed levels at all times. RB 48 n.8; RRB 2. While obviously true, it does not follow that firms systematically and substantially undercharge the majority of their customers for years, which is what respondent is claiming Evanston Evanston did in the 1990s.

The lack of authority for Respondent’s novel learning-aboutdemand position is not surprising. The argument runs at least partially counter to the Merger Guidelines. As respondent correctly points out, the Merger Guidelines provide that market power “is the ability profitably to maintain prices above competitive levels for a significant period of time.” Merger Guidelines § 0.1 (emphasis added). What respondent neglects to mention, however, is that the antitrust enforcement agencies typically apply the hypothetical monopolist test by “using prevailing prices of the products of the merging firms and possible substitutes for such products.” Id. § 1.11. The Merger Guidelines do mention two circumstances in which the agencies will use a price different from the prevailing price – (1) when pre-merger circumstances suggest that coordinated interaction has occurred and (2) in cases in which it is possible to predict changes in the prevailing prices with reasonable reliability. Id. Here, both complaint counsel and respondent agree that coordination among competitors is not at issue. And the econometric analysis used by Respondent’s and complaint counsel’s economists accounted for future changes in the prevailing price by factoring out the effects of the most likely competitively-benign factors that would cause prices to rise.

ALJ found, and as we agree, it is appropriate to determine that price increases reflect the exercise of market power by ruling out competitively-benign reasons for the price increases. On the same page to which respondent refers, the ALJ found that the “evidence therefore demonstrates that the relative price increases were the result of ENH’s enhanced market power, achieved through elimination of a competitor as a consequence of the merger.” Id. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 485 Opinion of the Commission In addition, while we are not aware of any court that has specifically discussed the appropriate baseline price to use for the hypothetical monopolist test or to measure the exercise of market power, courts have looked to actual prices when defining markets. Olin Corp. v. FTC, 986 F.2d 1295, 1300-02 (9th Cir. 1993) (applying Merger Guidelines and using actual prices); Sungard, 172 F. Supp. 2d at 186-92 (analyzing customer testimony about actual prices regarding possibility of 5% to 10% price increase); Staples, 970 F. Supp. at 1076-77 (using actual prices); New York v. Kraft Gen. Foods, Inc., 926 F. Supp. 321, 332-34, 359-61 (S.D.N.Y. 1995) (applying Merger Guidelines and referring back to previous analysis of relevant product market that contained references to actual customer prices); FTC v. Owens-Illinois, Inc., 681 F. Supp. 27, 38- 47 (D.D.C. 1988) (same), vacated as moot, 850 F.2d 694 (D.C. 1998); see also CF Indus. v. Surface Transport Bd., 255 F.3d 816, 824 (D.C. Cir. 2001) (“[N]ormal assumption in examining assertions of market power is that the current price is at least the competitive price.”).

Respondent’s argument also raises a number of practical issues. issues. It will almost always be true in markets where firms submit submit non-public bids or offers, such as hospital markets, that access by one firm to another firm’s prices will provide insight into into the demand structure that could allow a firm to price more closely to theoretical, long-run equilibrium levels on a sustained basis. It is also very likely, however, that systematic access by firms firms to their competitors’ pricing can undermine firms’ incentives incentives to price aggressively and can facilitate collusion. Customers often do not share one provider’s prices with another competing provider for this very reason. Presumably, Evanston did did not know Highland Park’s prices until Evanston received them during the due diligence process because MCOs thought that sharing sharing the pricing data might reduce Evanston’s incentives to compete aggressively for their business. Thus, caution is warranted warranted before assigning procompetitive or competitively neutral effects to competitors’ learning about each other’s pricing strategies VOLUME 144 Opinion of the Commission through mergers, and even more caution is needed when those mergers result in substantial price increases. We need not resolve all of the doctrinal or practical challenges presented by Respondent’s learning-about-demand argument, however, because, as we have discussed in detail in our findings of fact, giving respondent all benefit of the doubt, we agree with the ALJ that the facts in the record do not support the argument. First, the testimony of the ENH executives that their business and negotiating strategy caused them not to obtain competitive prices in negotiations with MCOs during the 1990s lacks credibility. Second, Evanston’s decision not to renegotiate certain contracts during the 1990s is equally consistent with Evanston’s deciding that it could not obtain higher prices. Third, Respondent’s learning-aboutdemand argument hinges heavily on the purported gap between Evanston’s pre-merger prices and those charged by Highland Park. As we found, while not unambiguous, the weight of the record evidence suggests that this gap did not exist. In addition, Baker’s regressions partially undermine the argument because even when he used an unrealistically narrow control group to test the learning-about-demand position, he found found that ENH’s post-merger prices to both Humana and United were statistically-significantly higher than the predicted levels.84 TR 4739, 4743, 4682-85 (Baker), in camera; RX 2040 at 4, in camera; DX 7068 at 46, in camera. For Humana, the average net prices that ENH charged were 21% higher in 2002 than he predicted they 84 We find it somewhat surprising that Baker chose to report the statistical significance of these results. We presume that this is due to the fact that when he originally reported the results before correcting a mathematical error, he explicitly reported that the results were not statistically significant. DX 7067 at 45, in camera.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 487 Opinion of the Commission would have been had the merger not occurred, and for United they were higher by 35% and 29% in 2002 and 2003, respectively.85 As we also found above, Respondent’s learning-about-demand argument is difficult to square with a number of Respondent’s other other positions. Respondent alleges that Evanston was and is a statestate-of-the art hospital, with superior management, that consistently consistently provided high-quality services. RFF ¶ 3. Yet respondent respondent also maintains that Evanston’s most senior officials did did not set prices at market levels for certain MCOs while simultaneously charging market rates for other MCOs, such as BCBS and Cigna. In contrast, respondent maintains that Highland Park provided such poor services that it was threatening patient safety, and that Highland Park was in severe financial distress, but at at the same time was highly proficient at setting a profit-maximizing profit-maximizing price. Again, this logical discrepancy is not determinative, but when viewed in conjunction with the totality of the other evidence, it supports our rejection of Respondent’s position position that Evanston was systematically charging belowbelow-competitive rates before the merger.86 b. Lack of Decline in Output Respondent also argues vigorously that complaint counsel’s position that the merger allowed supracompetitive pricing is deficient because complaint counsel did not show a decline in output. RB 56; RRB 5, 23-25. We disagree with Respondent’s 85 The Commission computed the numbers through straightforward calculations of the percentage differences in rows 7 vs. 9 (Humana), and rows 10 vs. 12 (United), in RX 2040 at 4, in camera; and DX 7068 at 46 (Table 4), in camera.

86 As described supra at 41, the learning-about-demand argument does not apply to the post-merger price increases at Highland Park. Respondent’s primary rebuttal to the econometrics as to Highland Park’s price increases is that they reflect increased demand for Highland Park’s services due to alleged post-merger improvements in the quality of the hospital. We address this argument, infra, at 70-72.

VOLUME 144 Opinion of the Commission reasoning. First, strictly speaking, the issue is not whether Respondent’s output declined in nominal terms, but whether it declined from what it would have been but for the merger. Despite a a merger-induced increase in ENH’s market power, its nominal level level of output still could have grown if demand for hospital services services in the Chicago area increased.

More fundamentally, respondent incorrectly assumes that there is a relatively constant relationship in the hospital market between quantity and price. The record reflects that this is not the case. When MCOs negotiate with hospitals, for the most part they are faced with an all-or-nothing decision about whether to include the hospital in their network because, as Hillebrand testified, it is “very, very difficult” for an MCO to steer its PPO members to particular in-plan hospitals through differential pricing. IDF ¶ 169; TR 1760-63, 1766 (Hillebrand). Steering also is not an option for HMO plans because HMOs charge members uniform rates for all hospitals in their networks and preclude members from using other hospitals. Thus, generally, output declines only after the hospital exceeds the price at which the MCO is willing to enter into any contract with the hospital, at which point the output drops very substantially. In other words, there is a substantial range of prices, including prices at supracompetitive levels, over which an MCO will decide to include a hospital in its networks without a material change in the level of the hospital’s services demanded by the MCO. The fact that complaint counsel did not prove a drop in market-wide output thus is not a deficiency in complaint counsel’s case. c. Quality Improvements at Highland Park Respondent also argues that some portion of the mergermerger-coincident price increases computed by both Haas-Wilson and Baker was caused by increased demand for Highland Park’s services due to post-merger improvements, rather than market power. RB 58-59, 62, 72. Complaint counsel responds that the ALJ EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 489 Opinion of the Commission ALJ found no evidence that the quality of care improved at ENH relative to other hospitals and, therefore, that Haas-Wilson’s and Baker’s estimates of the merger-coincident price increases do not require adjustment. CB 51.

Courts in merger cases usually consider efficiencies, including quality improvements, after the government has shown that the transaction is likely to reduce competition. See Heinz, 246 F.3d at 715, 720. Once the government has done so, the defendant can show that the loss of competition will not harm consumers by demonstrating that the transaction will produce “significant economies and that these economies ultimately would benefit competition and, hence, consumers.” University Health, 938 F.2d at 1223; see Merger Guidelines § 4.0 (“To make the requisite determination, the Agency considers whether cognizable efficiencies likely would be sufficient to reverse the merger’s potential to harm consumers in the relevant market, e.g., by preventing price increases in that market.”). The defendant has the burden of production to show that efficiencies offset any likely anticompetitive effects of the increase in market power produced by the merger. See Heinz, 246 F.3d at 715, 720 (finding that, to rebut presumptions of harm based on high concentration levels, defendants need to prove extraordinary efficiencies); Staples, 970 F. Supp. at 1088-89 (finding that defendants can use efficiency evidence to rebut presumption that merger will substantially lessen competition). Because of the manner in which complaint counsel presented its its case, however, here the issue of quality improvements at Highland Park is also relevant to determining whether the transaction increased the merging parties’ market power. Complaint Complaint counsel sought to prove that the merger increased ENH’s ENH’s market power by showing that there were large post-merger post-merger price increases that are not attributable to the most plausible competitively-benign factors. Respondent correctly points points out that one such plausible factor is that MCO demand for Highland Park’s services might have increased if (for whatever VOLUME 144 Opinion of the Commission reason) the quality of Highland Park’s services improved after the merger. RB 51. More formally, it is possible that the MCO demand demand curve for ENH’s services might have shifted outward after after the merger relative to demand for other hospitals due to a relative increase in the quality of the services at Highland Park. As we have found, however, the record does not support Respondent’s argument that improvements in quality at Highland Park caused the merger-coincident price increases at the hospital. First, because Evanston is more than twice the size of Highland Park, IDF ¶¶ 5, 22; ID 180, and generated roughly four times more revenue, CX 84 at 16, the large majority of commerce affected by ENH’s substantial post-merger price increases was from Evanston’s services, not those of Highland Park. Thus, even if respondent is correct that MCO demand for Highland Park’s services increased after the merger due to quality improvements, such increased demand likely accounted for well short of half of the substantial higher-than-predicted merger-coincident price increases identified by both Haas-Wilson and Baker.

Second, the record is ambiguous as to whether quality at Highland Park improved relative to that of other hospitals after the the merger. As we and the ALJ have found, however, even if Highland Park’s quality improved relative to that of other hospitals, hospitals, the record supports a finding that it did not increase demand for Highland Park’s services. ID 179. ENH did not mention mention to MCOs that its price increases were due to improvements improvements at Highland Park, IDF ¶¶ 840, 842; ID 178, and Neaman testified that he never saw any documents correlating the higher prices with the quality changes at Highland Park. IDF ¶ 843; 843; ID 178. Other than a single press release mentioning planned clinical service improvements, ENH never identified any improvements at Highland Park to MCOs. IDF ¶¶ 841-47; ID 178. 178. The MCO witnesses also testified that the topic of quality improvements at Highland Park never came up during contract negotiations, IDF ¶¶ 844-47; ID 178, and that they were not aware EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 491 Opinion of the Commission aware of a significant increase in quality at Highland Park after the the merger. IDF ¶¶ 846-47, 851; ID 181.

d. Merger Guidelines’ Unilateral Effects Standards Respondent also argues that complaint counsel has not satisfied the requirements for establishing that a merger enabled the combined firm unilaterally to exercise market power. Citing the Merger Guidelines, In re R.R. Donnelley & Sons Co., 120 F.T.C. 36, 195 (1995), and several other authorities, respondent argues that establishing a likelihood of unilateral effects requires showing that (1) the merging firms are viewed as the first and second choices by (2) customers accounting for significant sales in the relevant market. RRB 3-5. Respondent maintains that complaint counsel did not and could not introduce such evidence because Evanston was much larger than Highland Park, and was a teaching facility that offered a greater breadth of medical services than did Highland Park. RB 42- 43. Respondent also maintains that Evanston and Highland Park were geographically dissimilar because at least eight hospitals to the south of Evanston and two hospitals to the north of Highland Park are closer to Evanston and Highland Park, respectively, than Evanston and Highland Park are to each other. RB 43. Respondent’s position is not persuasive. An MCO’s demand for for hospital services is largely derived from an aggregation of the preferences of its employer and employee members. TR 5936-37 (Noether). When a hospital increases its price, the MCO can retain retain the hospital in its network and pay the higher price or drop the the hospital and replace it with another hospital or some combination of hospitals. TR 2470 (Haas-Wilson).87 If the MCO drops the hospital, it may cause some members who have a strong 87 See Town & Vistnes, supra note 76, at 734-36, 752; see also Gregory Vistnes, Hospitals, Mergers, and Two-State Competition, 67 ANTITRUST L.J. 671, 686 (2000).

VOLUME 144 Opinion of the Commission preference for that hospital to switch to another MCO, and cause employers with a significant number of such members to drop the MCO altogether. If a significant portion of an MCO’s members view view a hospital that raises its prices as particularly important, the MCO likely will be more willing to pay some or all of the increase. increase. TR 2475 (Haas-Wilson).88 For example, Bain advised ENH that it likely could increase its prices to PHCS due to the “significant leverage [that ENH had] in negotiations with PHCS as [PHCS] ha[s] [a] strong North Shore presence and need[s] [ENH] in their network.” CX 1998 at 44. Thus, whether the MCO decides to drop a hospital that raises its prices depends on a potentially complex assessment of the preferences of its employer and membership base. The record reflects that Evanston and Highland Park likely were close substitutes for MCOs’ members and employers, and thus for the MCOs. Evanston and Highland Park provided comparable primary and secondary services. TR 1291-93 (Neaman); CX 84 at 13, 15; TR 299 (Newton); TR 2083-88 (Spaeth). As Neaman testified, Evanston provided “[a]ll kinds of services, both inpatient and outpatient, sort of the basics, such as obstetrics, all the way up to the more intensive services, such as cardio-angiogenesis.” TR 1291 (Neaman). That Highland Park did not provide the tertiary services provided by Evanston does not negate the interchangeability of the two hospitals’ primary and secondary services, such as basic obstetrics and general surgery. Respondent’s implied argument to the contrary is at odds with common sense and its own documents, which reflect pre-merger competition between Evanston and Highland Park. In addition, the district court in Long Island Jewish Medical Center rejected an argument similar to Respondent’s position; the court held that two defendant merging academic hospitals (that provided tertiary services) competed with nearby community hospitals in the provision of primary and secondary care. 983 F. Supp. at 138-39.

88 See Town & Vistnes, supra note 76, at 734, 737. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 493 Opinion of the Commission Respondent’s position that the two hospitals were highly differentiated geographically has somewhat more force, but also is ultimately unpersuasive. Respondent argues that a number of other hospitals are closer to Evanston and Highland Park than the merging hospitals are to each other. RB 43. Respondent appears to side-step the fact that geographic substitutability is a function not merely of the geographic relationship of hospitals to each other, but also of the relationship of the hospitals to MCOs’ members. It is undisputed that there is an approximately thirteen-mile-long space between Evanston and Highland Park in which there are no other hospitals, and that no other hospitals are located within the geographic triangle formed by the ENH hospitals. Thus, it is likely that a significant number of MCO members who live in the triangle view Evanston and Highland Park as their preferred choices from a geographic perspective, and, therefore, that the financial cost to an MCO of removing the ENH hospitals from its network would exceed that of absorbing the price increase and spreading it over a larger membership base.

This conclusion is bolstered by ENH’s ability successfully to charge substantially higher-than-predicted price increases to the MCOs after the merger. The MCO testimony also partially supports supports this determination. United’s Jillian Foucre testified that Evanston and Highland Park would be the preferred choices of executives who lived in the triangle made up by the North Shore suburbs, and that executives who lived within the area made up by by the triangle would not want to travel greater distances north or south to go to hospitals. TR 901-02 (Foucre). Aetna’s Mendonsa testified that he thought that people who lived in the communities around the ENH hospitals would not want to travel to other hospitals, explaining that “[s]omeone that’s going to Evanston is not not going to drive all the way out to Park Ridge, which is where [Advocate] Lutheran General is, and . . . neither are they going to do do that with Northwest Community Hospital.” TR 541-43 (Mendonsa), in camera. Respondent’s contention that the two hospitals were “vastly different” from a geographic perspective, RRB 14, also conflicts with Spaeth’s testimony that Evanston was VOLUME 144 Opinion of the Commission Highland Park’s second overall closest competitor (after Lake Forest). TR 2163-64 (Spaeth).

We agree with respondent that not all of the MCO testimony is particularly precise and that it does not all support complaint counsel’s case. We do not agree, however, that the MCO testimony undermines complaint counsel’s case. Furthermore, while we likely would give less weight to customer testimony with such ambiguities in a challenge to an unconsummated merger, ambiguities are less concerning here, where our analysis is a retrospective inquiry based on empirical evidence and documents reflecting the parties’ postmerger assessments of the deal. Antitrust analysis depends fundamentally on market facts. As the ALJ and we have found, the facts here – the merging parties’ contemporaneous business assessment about the transaction’s competitive effects, complaint counsel’s and Respondent’s econometric analyses of ENH’s postmerger prices, and portions of the merging parties’ and the MCOs’ testimony – demonstrate on the whole that it is very likely that the merger enabled the combined firm to exercise market power. The section of the Merger Guidelines and the cases upon which respondent relies set forth conditions that typically are necessary for a transaction to enable the unilateral exercise of market power. These authorities do not mandate the use of a particular type of proof to establish those conditions. In particular, they do not require a court to enumerate the customers who view the merging parties as their first and second choices. As respondent acknowledges, the Merger Guidelines provide that a plaintiff may draw upon different types of evidence to establish unilateral effects. Merger Guidelines § 2.211 n.22.

One type of evidence that can be used to identify unilateral effects is “natural experiments,” by which economists use natural variations in the economy or other social phenomena to perform an an economic analysis. For example, in Staples, the FTC and the EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 495 Opinion of the Commission court relied, in part, on data that showed that “Staples and Office Depot both charge[d] higher prices where they face[d] no superstore superstore competition [than when they did face competition from other superstores, which] demonstrate[d] that an office superstore can raise prices above competitive levels.” See Staples, 970 F. Supp. Supp. at 1082; see generally Joseph Larson et al., The Role of Economics and Economists in Antitrust Law, 2004 Colum. Bus. L. Rev. 419, 453 (2004) (describing the use of natural experiments in merger analysis, including how “[c]omparisons of prices before and and after competitor entry and exit are good candidates for natural experiments”).

Here, complaint counsel relied on economic analysis of Respondent’s post-merger prices (a form of natural experiment), as well as Evanston’s and Highland Park’s business documents, to establish the relevant product and geographic market and to show that the transaction enabled the merged firm unilaterally to exercise market power. The documents do not need to state affirmatively that a sufficient number of MCOs (or their members) viewed the merging parties as next best substitutes. Seldom do business documents use the language of the Merger Guidelines and economists to describe competition in markets. Further, economic analysis of actual market events, combined with review of other evidence, is a sound methodology to determine whether customers accounting for a significant share of ENH’s business viewed Evanston and Highland Park as next-best substitutes for particular needs, and support our making such a determination in this case. e. Repositioning of Competitors Respondent also maintains that complaint counsel has failed to to show that repositioning by ENH’s competitors did not prevent or or eliminate any anticompetitive effects. RRB 20-22. We disagree. Following the Merger Guidelines, the courts generally hold that entry must be likely in a two-year period in order to conclude that it it will offset a transaction’s anticompetitive effects. See Cardinal VOLUME 144 Opinion of the Commission Health, 12 F. Supp. 2d at 55; Merger Guidelines § 3.2. As the ALJ found, new entry or repositioning did not reduce the market power power that ENH obtained from the merger during a two-year period. period. To the contrary, the econometric evidence, viewed in conjunction with the rest of the record, demonstrates that ENH was was able to increase its prices by above-market rates for at least two two years after the merger occurred.

The weight of the evidence shows that it is unlikely that new entry or expansion reduced ENH’s market power after the two-year period either. No new hospitals have been built in the relevant geographic market since the merger, which suggests that entry or expansion has not alleviated the market power created by the transaction. IDF ¶ 1021. Further, because it takes at least two and one-half years to build a new hospital, it is unlikely that new entry will occur in the geographic market in the near future. IDF ¶ 1024. In addition, the documents, MCO testimony, and econometrics do not indicate that ENH could exercise market power due to capacity constraints at hospitals outside of the geographic market. Rather, the likely cause of the market power created by the merger was the elimination of competition between hospitals that were the most geographically convenient for a significant number of MCO members who lived within the triangle formed by the three ENH hospitals. Thus, we agree with the ALJ that new entry or repositioning did not alleviate the transaction’s anticompetitive effects.

f. Elzinga-Hogarty Test Finally, respondent argues that patient flow data undermine the the ALJ’s conclusion that the triangle formed by the three ENH hospitals is a relevant geographic market and that the ALJ erred by by not considering such data. RB 32-33; ID 139. As the name suggests, patient flow data provide information about where patients EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 497 Opinion of the Commission travel to obtain hospital services. TR 2356, 2375 (Elzinga); TR 6203-04 (Noether). Respondent claims that in the context of “an 80% service area,” Evanston had more patient overlap with Northwestern Memorial, Rush North Shore, Advocate Lutheran General, St. Francis, and Weiss than with Highland Park. RB 32, in camera. In addition, respondent maintains that there was at least as great an overlap before the merger between Highland Park and Advocate Lutheran General or Lake Forest as between Evanston and Highland Park. Id.

A number of courts have considered patient flow data when they have defined the geographic market. In particular, they have applied the Elzinga-Hogarty (“E-H”) test to patient flow information as a proxy test to determine whether a firm could exercise market power in a potential geographic market. See California v. Sutter Health Sys., 84 F. Supp. 2d 1057, 1072 (N.D. Cal. 2002); FTC v. Tenet Healthcare Corp., 17 F. Supp. 2d 937 (E.D. Mo. 1998), rev’d on other grounds, 186 F.3d 1045 (8th Cir. 1999); FTC v. Freeman Hosp., 911 F. Supp. 1213, aff’d, 69 F.3d 260, 264-65 (8th Cir. 1995); United States v. Mercy Health Servs., 902 F. Supp. 968, 978 (N.D. Iowa 1995), vacated as moot, 107 F.3d 632 (8th Cir. 1997). The E-H test was devised by professors Kenneth G. Elzinga and and Thomas F. Hogarty to help to delineate geographic markets, specifically in the coal and beer industries. TR 2374-76 (Elzinga); see Kenneth G. Elzinga & Thomas F. Hogarty, The Problem of Geographic Market Delineation in Antimerger Suits, 18 Antitrust Bull. 45 (1973); Kenneth G. Elzinga & Thomas F. Hogarty, The Problem of Geographic Market Delineation Revisited: The Case of of Coal, 23 Antitrust Bull. 1 (1978). The objective of the E-H test is is to “measure[] the accuracy of a [potential] market delineation by by determining the amount of either imports into or exports from a a tentative market.” United States v. Country Lake Foods, Inc., 754 754 F. Supp. 669, 672 n.2 (D. Minn. 1990). The test’s underlying assumption is that if an area has significant exports outside of the area or imports into the area, then that area is not a relevant VOLUME 144 Opinion of the Commission geographic market because it is unlikely that a dominant firm within the area could exercise market power. See id.; TR 2372-73 (Elzinga). At trial, Professor Elzinga testified that the E-H test was not an appropriate method to define geographic markets in the hospital sector because of two related problems, which he termed the “silent majority fallacy” and the “payor problem.” TR 2369 (Elzinga). The silent majority fallacy is the false assumption that patients who travel to a distant hospital to obtain care significantly constrain the prices that the closer hospital charges to patients who will not travel to other hospitals. TR 2356, 2384-87, 2391 (Elzinga). Elzinga testified that for the most part, patient decisions do not have such a constraining effect because their choices of hospitals largely are based not on price but on other factors, such as location and the preferences of their physician. TR 2387-88 (Elzinga); see TR 2463- 65 (Haas-Wilson). He explained that:

[p]eople who travel outside their home turf for hospital services usually do so [because] . . . [t]here’s some particular service or amenity that they associate with that distant hospital that’s important to them, or they may have family who lives some distance away and they travel to that hospital. People who consume . . . hospital services close to home typically are there either because their doctor places them at that hospital or, for purposes of their own convenience or the convenience of their family, it is very important for them to be hospitalized close to home. So, unlike products like coal and beer that will move about in response to the market signals . . . prices change and beer gets shipped to a different location – here, the prices of hospital services do not drive most people to change the location of where they consume hospital services.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 499 Opinion of the Commission TR 2387-88 (Elzinga).89 Further reducing the effect of prices on patients’ hospital choices choices is that patients rarely pay directly the full cost of hospital services. Insurance companies pay the large majority of hospital costs in most instances from revenues obtained through a broad base base of employer and employee-paid premiums and deductibles. Consequently, when a hospital raises its prices, the increase often is is spread out over a broad number of employers and members, many many of whom will never use the hospital. Even if an MCO tries to to steer patients toward less costly hospitals through “tiering” of coco-payments, the price effect often is diluted because the coco-payments often do not cover the difference between the total costs of the expensive hospital and those of other, less costly hospitals.90 Consequently, there is little reason to infer from some residents’ choice of a more distant hospital that others would do likewise in response to a price increase from a closer hospital.91 Put more formally, the workings of the third-party payor system system in the United States are such that rarely do patients fully internalize the benefits and costs of their decision to purchase a 89 See also Capps et al., supra note 76, at 739 (“Given the propensity of some patients to travel substantial distances for care, [the Elzinga-Hogarty] standard has led to large market boundaries and, consequently, permissive merger rulings. Our results indicate that this may be a serious error. . . . Many patients, especially those with conditions that are relatively straightforward to treat, have a strong preference to go to a convenient, nearby hospital. These preferences give hospitals with no nearby competitors a strong bargaining position.”). 90 Respondent maintains that MCOs can in fact force patients at least partially to internalize the price of hospital services through various steering techniques, such as hospital-specific co-payments and tiered networks. RB 33 n.6. The bulk of the evidence, however, is that, at least in the Chicago area, MCOs largely do not engage in such steering. E.g., TR 594-95 (Neary); TR 1760- 61 (Hillebrand).

91 See FED. TRADE COMM’N, IMPROVING HEALTH CARE: A DOSE OF COMPETITION, Ch. 4, at 8-10 (July 2004), available at http://www.ftc.gov/reports/healthcare/040723healthcarerpt.pdf (describing the silent majority fallacy).

VOLUME 144 Opinion of the Commission medical product or service. This lack of internalization is what Elzinga termed the “payor problem”:

[T]here’s a wedge between the consumption of the service and the person who decides where the service will be consumed and then some other party actually paying for the service, and consequently, the usual market analysis of goods and services . . . in response to price incentives really doesn’t fit. And so it follows in my view that looking at the flow of patients really doesn’t help you define the contours of a relevant geographic market area[] because the patients who are moving are not necessarily moving in response to price incentives.

TR 2395-96 (Elzinga).

Elzinga concluded that because “the ability of particular hospitals to raise prices is not disciplined or thwarted by the travel patterns” of patients, TR 2388 (Elzinga), using patient flow data is uninformative about whether it would be profitable for merging hospitals to raise prices, and that the application of the E-H test to patient flow data would identify overly broad geographic markets. TR 2393 (Elzinga).

We find Elzinga’s testimony to be persuasive. Respondent did not directly dispute Elzinga’s views about the general lack of validity of using the E-H test in hospital markets to define geographic markets, including the propensity of the test to define improperly large markets. Moreover, Noether agreed that “the use of of the [E-H] test is not appropriate for this case.” DX 7126 at 6. Nonetheless, there is some merit to Respondent’s argument that the the ALJ erred in holding that patient flow data are always irrelevant to determining the relevant geographic market. RB 32-33; ID 139. MCO demand for hospital services is partially a derived demand EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 501 Opinion of the Commission based on patient preferences, and the percentage of patients in a given area who use a hospital can, in certain circumstances, provide some rough indication of MCO preferences when they form a network. Ultimately, however, we believe that we should view patient flow data with a high degree of caution because of the silent majority fallacy and payor problem and, at best, we should use it as one potentially very rough benchmark in the context of evaluating other types of evidence. A robust application of the hypothetical monopolist methodology is almost certain to produce a more reliable determination of the geographic market than is analysis of patient flow data.

In this case, even assuming that Respondent’s description of the patient flow information is correct, it provides no sound basis to alter our conclusion that the merger resulted in ENH’s ability to exercise market power or that the triangle formed by the ENH hospitals is a relevant geographic market. For the reasons that Professor Elzinga explained, that Evanston and Highland Park may have had a greater patient flow overlap with certain other hospitals than they did with each other is not inconsistent with the conclusion that the combination of Evanston and Highland Park enabled the merged entity to exercise market power. To the contrary, here the record reflects that the merger did just that, and, consequently, that the relevant geographic market is narrower than the patient flow data might suggest.

4. Summary of Competitive Effects Analysis In summary, we find that the merger enabled ENH to exercise market power, and that ENH used this market power to increase its its average net prices to MCOs for acute inpatient hospital services services by a substantial amount – at least the 9% or 10% calculated calculated by Baker. No one type of evidence is dispositive. Instead, Instead, the econometric evidence, viewed in conjunction with Respondent’s pre- and post-merger documents and the MCO and executive testimony, demonstrate that ENH’s substantially higherhigher-than-predicted merger-coincident price increases were due to VOLUME 144 Opinion of the Commission to market power, rather than competitively-benign factors. Respondent’s alternative explanations for these price increases are not supported by the weight of the record evidence. We also find that that because the merger enabled ENH to raise prices by a substantial substantial amount (at least equal to a SSNIP) through the unilateral unilateral exercise of market power, the geographic triangle in which which the three ENH hospitals are located constitutes a well-defined well-defined antitrust geographic market under Section 7. See IV Areeda, Hovenkamp & Solow, supra note 69, ¶ 913b, at 64. VI. EFFICIENCIES AND JUSTIFICATIONS Having found that the transaction reduced competition substantially, we now address Respondent’s efficiency claims and other justifications for the transaction. Respondent argues that the merger produced competitive benefits that outweigh the harm to competition alleged to have resulted from this merger. First, respondent argues that the merger increased the financial strength of Highland Park, transforming it from a weak to a strong competitor. Second, respondent argues that the merger produced significant quality improvements at Highland Park, enhancing that hospital’s ability to compete with other hospitals in the Chicago area. RB at 62. Finally, respondent argues that its not-for-profit status reduces the merger’s potential to cause competitive harm. We address these arguments in turn.

A. The “Weakened Company” Justification ENH argues that, prior to the merger, Highland Park was on a financial “downward spiral” that limited its competitive viability in in the future, and that the evidence of Highland Park’s weakened financial condition rebuts or mitigates complaint counsel’s showing showing regarding the merger’s anticompetitive effects. ENH implicitly concedes that Highland Park’s alleged financial difficulties fall short of the criteria required to establish a “failing EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 503 Opinion of the Commission firm” defense under the Merger Guidelines.92 Instead, it relies on United States v. General Dynamics Corp., 415 U.S. 486 (1974), and and cases that have followed it, for the proposition that, even if the the acquired firm is not “failing,” evidence that it has “severely limited” resources is relevant to the assessment of whether the challenged transaction is likely to cause competitive harm. RB at 63. 63.

In General Dynamics, the Supreme Court held that the market share statistics used by the government to challenge the merger of two coal companies were insufficient to sustain its case because, by by failing to take into account the fact that the acquired firm’s coal coal reserves were depleted or committed under long-term contracts, contracts, those statistics overestimated the acquired firm’s ability to to compete in the future. 415 U.S. at 500-04. Several courts have applied the General Dynamics rationale in ruling that evidence of the acquired firm’s weakened financial condition, among other factors, may rebut the government’s statistical showing of anticompetitive market concentration. See Kaiser Aluminum & Chem. Corp. v. FTC, 652 F.2d 1324, 1337-41 (7th Cir. 1981); FTC v. FTC v. National Tea Co., 603 F.2d 694, 698-700 (8th Cir. 1979); FTC v. Arch Coal, Inc., 329 F. Supp. 2d 109, 153-54 (D.D.C. 1998). 1998). These courts have generally cautioned, however, that 92 As the Merger Guidelines state:

A merger is not likely to create or enhance market power or facilitate its exercise if the following circumstances are met: 1) the allegedly failing firm would be unable to meet its financial obligations in the near future; 2) it would not be able to reorganize successfully under Chapter 11 of the Bankruptcy Act; 3) it has made unsuccessful good-faith efforts to elicit reasonable alternative offers of acquisition of the assets of the failing firm that would both keep its tangible and intangible assets in the relevant market and pose a less severe danger to competition than does the proposed merger; and 4) absent the acquisition, the assets of the failing firm would exit the relevant market.

Merger Guidelines § 5.1 (footnotes omitted). VOLUME 144 Opinion of the Commission “[f]inancial weakness, while perhaps relevant in some cases, is probably the weakest ground of all for justifying a merger,” and “certainly cannot be the primary justification” for permitting one. Kaiser Aluminum, 652 F.2d at 1339, 1341; accord Arch Coal, 329 F. F. Supp. 2d at 154.93 Notably, “while a merger is a relatively ‘permanent’ arrangement having long-lasting competitive effects, financial difficulties not raising a significant threat of failure are typically remedied in a moderate length of time.” IVA Areeda, Hovenkamp & Solow, supra note 69, ¶ 963, at 14. As the Eleventh Eleventh Circuit held in University Health: [W]e will credit such a defense only in rare cases, when the defendant makes a substantial showing that the acquired firm’s weakness, which cannot be resolved by any competitive means, would cause that firm’s market share to reduce to a level that would undermine the government’s prima facie case. 938 F.2d at 1221; accord Tenet Healthcare, 17 F. Supp. 2d at 947.94 The precise standard for evaluating a weakened company justification is not material here because the record evidence does 93 As the Seventh Circuit observed in rejecting a weakened company defense, even the acquisition of a weak company can have anticompetitive consequences. Kaiser Aluminum, 652 F.2d at 1339 (“The acquisition of a financially weak company in effect hands over its customers to the financially strong, thereby deterring competition by preventing others from acquiring those customers, making entry into the market more difficult.”); id. at 1341 (“History records and common sense indicates that the creation of monopoly and the loss of competition involve the acquisition of the small and the weak by the big and the strong.”).

94 See also Rockford Mem’l, 717 F. Supp. at 1289 (rejecting defendants’ argument that the merger should be allowed “on the basis of its prediction of future financial calamity,” finding that “this 'failing market’ or 'writing on the wall’ defense [is] too broad and ungainly to ward off a Section 7 violation”), aff’d, 898 F.2d 1278 (7th Cir. 1990).

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 505 Opinion of the Commission not substantiate ENH’s contention that Highland Park’s pre-merger financial condition prevented it from competing effectively. Instead, the evidence shows that Highland Park’s financial condition was essentially sound. Highland Park had a strong balance sheet, with more than sufficient cash and assets to cover its long-term debt, continue operations, and – as Highland Park’s strategic and financial plans indicated it intended to do – make substantial capital expenditures to improve its services and facilities. IDF ¶¶ 1028-51. Before the merger Highland Park had “historically achieved strong financial results compared to the median of not-for-profit hospitals.” CX 545 at 3. At the end of 1998, Highland Park and its affiliated corporations had cash and unrestricted investments of approximately $218 million and long-term debt of $120.5 million. By the end of 1999, cash and unrestricted investments had increased to approximately $260 million, while long-term debt had diminished to $116.7 million. CX 693 at 16-17. At the end of 1998, Highland Park had enough cash on hand to run a fully functional hospital for 444 days without any additional revenue (2.4 times the national average for “A” rated hospitals) – and this amount did not even include assets of the pre-merger Highland Park Foundation, whose funds went to support Highland Park and backed up its long-term debt. CX 1912 at 2; TR 5846, 5859-60 (Kaufman); IDF ¶¶ 1052-55. Indeed, Highland Park was sufficiently well-capitalized that, during the 1999 1999 merger negotiations with Evanston, it insisted on contributing contributing $100 million to establish an independent community foundation. TR 5843 (Kaufman); CX 1912 at 3. Although Highland Park experienced operating losses in 1999, 1999, its management believed that Highland Park would “remain financially strong over the foreseeable future.” CX 1055 at 3. The vast majority of the operating loss reported by Highland Park in 1999 was for merger-related costs. CX 1732 at 4; TR 412-13 (Newton). Highland Park’s 1999-2003 financial plan – which assumed that Highland Park would not merge with another institution – set forth a long-range capital budget that included over over $100 million for various strategic initiatives and capital investments. CX 545 at 3. Highland Park anticipated that, based on VOLUME 144 Opinion of the Commission on growth through new clinical services and existing cash and investments, the hospital could “generate sufficient cash” to “restore “restore the profitability” of the hospital and fund its numerous planned strategic initiatives and improvements. CX 1903 at 1; CX 545 at 3-4. Highland Park also had the support of a very wealthy community, which contributed millions of dollars to capital campaigns to fund various hospital projects. For example, one such such campaign in the early or mid-1990s raised more than $10 million for new surgical suites; another in 1998 raised money for Highland Park’s dialysis center. TR 319-21 (Newton); TR 4954, 4959-60 (Styer).

Even as Highland Park contemplated merging with Evanston or another hospital, its management believed that continuing operations as an independent hospital was a viable alternative. IDF ¶¶ 1056-57, 1060-61. Highland Park’s Chairman of the Board testified that, if the merger with Evanston had fallen through, “[t]here was no urgency to have an alternative immediately available” and that Highland Park had the “financial wherewithal to sustain [itself]” for at least ten more years. CX 6305 at 11 (Stearns); IDF ¶¶ 1058-59. ENH argues that Highland Park’s financial health was far worse worse than its reporting of positive operational income for all years years except 1999 would suggest, because Highland Park was “subsidizing” its operations with investment income. RB 64.95 However, financial statements prepared by Highland Park’s transaction counsel show that, even excluding investment income, Highland Park had positive operating income in 1997 and 1998. RX RX 514 at 12. ENH’s due diligence report also indicates that Highland Park had positive operating income (not including the prepre-merger Highland Park Foundation, investment income, or 95 The record shows that ENH itself reported certain investment income as part of its operational income, in both the pre- and post-merger periods. RX 1194 at ENHLTH 1407; CX 2068 at 6.

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 507 Opinion of the Commission financing and interest payments) in 1996, 1997, and 1998. RX 609 at EY000256-57. Furthermore, the fact that Highland Park had additional sources of funds available to it, including income from its investments and funds from Highland Park’s pre-merger foundation, supports a finding that Highland Park had the financial wherewithal to make necessary capital investments and enhance its facilities and services – investments to which Highland Park was committed, even without a merger, to improve the hospital’s future performance. In sum, the record does not support a conclusion that Highland Park’s pre-merger financial health precluded Highland Park from being a meaningful competitive force, or that there was no economically reasonable strategy that Highland Park could follow, either as a standalone entity or in partnership with another, to improve its prospects. Whatever challenges Highland Park faced prior to the merger, it had considerably greater financial resources and competitive options available to it than anything courts have found to satisfy a weakened company justification. B. ENH’s Quality Improvements Justification ENH also argues that any adverse competitive effects resulting from the merger are outweighed by significant quality improvements at Highland Park that the merger has produced. ENH presented evidence that it has spent over $120 million post-merger to make improvements and expand services at Highland Park in 16 areas: (1) OB/GYN, (2) quality assurance, (3) quality improvements, (4) nursing, (5) physical plant, (6) oncology, (7) radiology and radiation medicine, (8) emergency care, (9) laboratory medicine, (10) pharmacy, (11) cardiac surgery, (12) interventional cardiology, (13) intensive care, (14) psychiatry, (15) electronic medical records, and (16) medical staff integration and academic affiliation. ENH’s improved quality argument raises interesting questions about how quality of care fits within a competitive effects analysis. analysis. Quality is one dimension on which firms compete, and differences in prices may reflect differences in quality. Improved VOLUME 144 Opinion of the Commission quality also may factor into analysis of efficiencies. As the Merger Guidelines recognize, “mergers have the potential to generate significant efficiencies by permitting a better utilization of existing assets, enabling the combined firm to achieve lower costs in producing a given quantity and quality than either firm could have achieved without the proposed transaction,” and efficiencies “can enhance the merged firm’s ability and incentive to compete, which may result in lower prices, improved quality, enhanced service, or new products.” Merger Guidelines § 4. However, ENH does not argue that the claimed quality improvements at Highland Park have come about as a result of cost-saving efficiencies produced by the merger.96 Instead, ENH characterizes quality improvements at Highland Park as benefits distinct from cost-savings that offset any adverse competitive effects produced by the merger. The case law provides no clear answers regarding how, or whether, such claimed qualitative benefits ought to fit into a competitive effects analysis. ENH’s quality improvements argument argument here is similar to one made by the defendants, and rejected rejected by the court, in Rockford Memorial Corp. In that case, the defendants argued that, even if the merger had anticompetitive effects, any adverse effects for consumers were outweighed by qualitative benefits to consumers from expanded and improved services that the merging hospitals intended to undertake. 717 F. Supp. at 1287-88. Although the court acknowledged that “the improvement in services would have a positive effect for consumers consumers of healthcare in the relevant market,” it held that such improvements were “irrelevant for the present § 7 inquiry” because because “the court’s exclusive role is to evaluate the merger’s effect 96 Although ENH asserts, in passing, that some of ENH’s improvements to Highland Park enhance cost efficiency, e.g., RB 75, it has made no effort to quantify any such cost savings or otherwise substantiate this claim. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 509 Opinion of the Commission effect on competition for the relevant market and no more.” Id. at 1288-89.97 Other courts have been more receptive to quality-of-care arguments, but those decisions shed little light on how qualitative benefits are to be weighed against the competitive harm shown to result from a merger. See, e.g., Tenet Healthcare, 186 F.3d at 1053- 54 (mentioning improved quality as a benefit of merger, but basing reversal of district court’s preliminary injunction on failure to prove relevant market).

But whatever uncertainties there may be, it is clear that claims of quality improvements must be subject to the same “rigorous analysis” that applies to all claims of procompetitive efficiencies to ensure that they “represent more than mere speculation and promises.” Heinz, 246 F.3d at 721. ENH must show that the claimed benefits are (1) verifiable; (2) merger-specific, i.e., ones that could not practicably be achieved without the proposed merger; and (3) greater than the transaction’s substantial anticompetitive effects. See Merger Guidelines § 4; see also Heinz, 246 F.3d at 721-22 (finding that, among other things, asserted efficiencies must be “mergerspecific”); University Health, 938 F.2d at 1223 (“speculative, selfserving assertions” will not suffice); Staples, 970 F. Supp. at 1089- 90 (rejecting claimed efficiencies that were “unverified” and not supported by “credible evidence”).

ENH argues that the first of these requirements is satisfied here here because – unlike the typical case in which the merger has not yet been consummated – the claimed improvements here already have been implemented and therefore can be “verified,” and the natural inference is that they resulted from the merger. RB 76. We disagree. The fact that we can verify that ENH actually made the claimed improvements at Highland Park following the merger tells 97 In the court’s view, weighing the claimed quality improvements against the merger’s anticompetitive effects would require a “value choice . . . beyond the ordinary limits of judicial competence.” Rockford Memorial, 717 F. Supp. at 1288 (citing Philadelphia Natl Bank, 374 U.S. at 371) (internal quotation marks omitted).

VOLUME 144 Opinion of the Commission tells us little about whether these changes improved quality of care, or whether these improvements could have been achieved by Highland Park without the merger and “without the concomitant loss loss of a competitor.” Heinz, 246 F.3d at 722. The ALJ found that there were several problems with ENH’s quality improvement claims. First, ENH did not present any quantifiable evidence that improvements at Highland Park enhanced competition. Second, ENH failed to show that quality improved across the combined ENH system (not just at Highland Park) and relative to other hospitals. Third, the ALJ found that the vast majority of the claimed improvements at Highland Park were not merger-specific. ID 177-78. As to the last point, the ALJ found that, before the merger, Highland Park had already committed (and had the financial ability) to invest over $100 million to improve and expand its facilities and services, including in many of the same areas identified by ENH as merger-related improvements. ID 182. The ALJ also found that, even apart from Highland Park’s actual plans, the types of improvements claimed by ENH – improved facilities, staffing changes, and new procedures – did not require a merger. The ALJ did find that two of the claimed quality improvements – installation of the EPIC electronic medical records management system, and integration and affiliation with an academic teaching hospital – were merger-specific, but he concluded that these improvements did not outweigh the anticompetitive effects of the merger. ID 190-92.

Although our analysis differs in some respects from that of the ALJ, we agree that the evidence presented by ENH fails to rebut complaint counsel’s showing of anticompetitive effects.98 As we 98 For example, we do not agree with the ALJ that ENH must show, as part of its initial burden of production, that quality improved across the ENH system. If ENH showed that the merger improved quality at Highland Park, complaint counsel could certainly counter ENH’s evidence by showing a decline in quality elsewhere in the ENH system (and, indeed, it presented expert testimony to this EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 511 Opinion of the Commission explained in our findings of fact, we find that the quality improvements asserted by ENH are not properly credited as benefits benefits of the merger because Highland Park could, and likely would, have made similar improvements without a merger. Our core core findings that support these conclusions are the following: (1) Highland Park had plans in place to improve its quality and expand expand its services without a merger, including undertaking many of of the same improvements that ENH credits to the merger, such as developing a cardiac surgery program in affiliation with Evanston or or another hospital, IDF ¶¶ 952-58; (2) before the merger Highland Highland Park already had begun to make a number of the improvements that ENH contends the merger produced; and (3) a number of the changes that ENH made at Highland Park after the merger reflect emerging trends in the industry, rather than benefits unique to the merger. IDF ¶ 895 (quality assurance program); IDF ¶¶ ¶¶ 901-02 (quality improvement program); IDF ¶ 950 (decentralized (decentralized dispensation of medication); IDF ¶ 973 (use of intensivists); IDF ¶ 983 (electronic medical records systems); TR 3840-41 (Silver) (in-house physician coverage in obstetrics departments).

ENH contends that Highland Park could not have achieved any of these improvements without the merger because they required ENH’s superior leadership and “collaborative and multidisciplinary culture,” which Highland Park supposedly lacked. RB 77-78. This argument is without merit. As Areeda, Hovenkamp, and Solow have observed:

Differences in management efficiency among competing firms are well-nigh universal. The usual effect), but it should not be part of ENH’s initial burden to show otherwise. We also think that the ALJ spoke too broadly in stating that ENH must show that quality at Highland Park improved relative to other hospitals. As we discuss below, whether a claimed improvement merely reflects a general trend in the industry (and thus might be deemed not merger-specific) is a relevant factor to be considered with regard to certain types of improvements (e.g., certain process changes), but it is not necessarily an appropriate inquiry across the board. VOLUME 144 Opinion of the Commission cure for inefficient management is to replace it, as is frequently and easily done, sometimes by the board of directors, sometimes by disgruntled shareholders. As a result, management replacement is not a “merger-specific” economy. To be sure, a merger may be a quicker way of achieving this goal, particularly where the board is indecisive or the shareholders are divided. But most firms have relatively inefficient management from time to time. time. To permit all such firms to solve their problems problems by substantial horizontal merger could eviscerate § 7 of the Clayton Act.

IVA Areeda, Hovenkamp & Solow, supra note 69, ¶ 974, at 74 (footnote omitted). Moreover, the record shows that, when the need arose, Highland Park could readily institute new leadership to effect changes in its operations and improve its quality of care. TR 3746- 49 (Krasner); TR 5479-80 (Chassin).

As noted above, the only claimed improvement that we think is properly deemed merger-specific is medical staff integration and affiliation with a teaching hospital. ENH’s health care quality expert testified that the integration of medical staff and academic affiliation provides Highland Park physicians with greater opportunities to upgrade their skills and keeps them “on their toes.” TR 5373-78 (Chassin). But this does not constitute verifiable evidence that ENH has improved quality at Highland Park, much less that any such improvement is of sufficient magnitude to offset the competitive harm that demonstrably has resulted from the merger. In addition, in many instances, ENH produced little verifiable evidence that the changes it made at Highland Park improved quality quality of care. ENH’s quality claims are based to a large extent on on the testimony of its administrators, physicians, and nurse leaders, leaders, who offered their observations about the quality of care at EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 513 Opinion of the Commission Highland Park before and after ENH made changes. But, for the most part, ENH did not produce data to substantiate its assessments assessments of quality at Highland Park, even though the record shows that ENH routinely tracks numerous quality indicators as part part of its quality improvement program. CX 2052; CX 2436; RX 1326, in camera.99 Although ENH’s quality expert, Dr. Chassin, included some quantitative data in his analysis (e.g., comparing Highland Park’s pre- and post-merger rates of administration of aspirin and beta blockers to heart attack patients, TR 5281-83 (Chassin)), his analysis was principally qualitative, and was itself based in large part on anecdotal information provided by ENH’s current administrative and medical leadership. TR 3011-12 (Romano); TR 5161-66 (Chassin).100 We recognize that assessing the impact on quality of ENH’s changes at Highland Park is not a simple matter and that, as Dr. Chassin testified, outcome measures are not always valid measures measures of quality. TR 5143-45, 5148 (Chassin).101 But, as is the case with claimed economic efficiencies, difficulties of proof do not not relieve ENH of its burden to produce verifiable evidence. Given Given the particular circumstances of this case – the fact that the merger has already been consummated, many of the claimed improvements were implemented years ago, and ENH routinely tracks numerous quality indicators – ENH could have produced 99 For example, ENH’s witnesses testified that changes implemented by ENH in radiology and emergency care improved turn-around times in those departments, but ENH did not produce data to substantiate these statements. TR 3632-34, 3643 (Victor); TR 4283-84, 4296 (Harris). 100 Complaint counsel’s quality expert, Dr. Romano, testified – and Dr. Chassin himself acknowledged, TR 5473 (Chassin) – that Dr. Chassin’s methods for gathering information did not meet accepted standards of qualitative research. Among other things, Dr. Chassin made no effort to obtain the views of individuals who might have contradictory views or a perspective different from that of ENH’s leadership. TR 3013-18 (Romano). 101 On the other hand, Dr. Romano testified that structural measures are insufficient by themselves, “because they tell us very little, if anything, about the care that’s actually provided to patients.” TR 2988 (Romano). VOLUME 144 Opinion of the Commission more concrete evidence than it did to substantiate its claims that the changes it made at Highland Park improved the quality of care. As the court emphasized in Heinz, “a rigorous analysis” is required to ensure that defendant’s claims of offsetting procompetitive benefits “represent more than mere speculation.” Heinz, 246 F.3d at 721. The dearth of verifiable evidence here is all the more reason for us to find that ENH has failed to satisfy its burden to prove “extraordinary” procompetitive benefits, id., offsetting complaint counsel’s showing of competitive harm.

C. ENH’s Not-For-Profit Status ENH also contends that the ALJ erred in rejecting its argument that its status as a not-for-profit hospital system greatly reduces the potential for competitive harm. RB 83 n.27. ENH does not devote much time to this argument, and we need not either. As the Seventh Circuit has observed in rejecting this defense, “[t]he adoption of the nonprofit form does not change human nature . . . , as the courts have recognized in rejecting an implicit antitrust exemption for nonprofit enterprises.” Hospital Corp. of Am. v. FTC, 807 F.2d 1381, 1390 (7th Cir. 1986) (citation omitted); see also Rockford Mem’l, 898 F.2d at 1285. Neaman also testified that there was no relationship between ENH’s non-profit status and the prices that ENH set. TR 1032-33 (Neaman). More broadly, the totality of the record shows that ENH’s non-profit status did not affect its efforts to raise prices after the merger, and we readily agree with the ALJ that ENH’s status as a nonprofit entity does not suffice to rebut complaint counsel’s evidence of anticompetitive effects. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 515 Opinion of the Commission VII. COUNT II Complaint counsel has appealed the ALJ’s decision not to issue an order against respondent under Count II. Complaint counsel alleges in Count II that the transaction violated Section 7 because the evidence shows that the transaction allowed ENH to exercise market power. Complaint Counsel did not allege a relevant product or geographic market in Count II, stating that it is not necessary to do so. CB 72-74.

Having found that the evidence is sufficient to define the product and geographic markets, and that complaint counsel has prevailed under Count I, we consider it unnecessary to decide whether the law permits establishing a violation of Section 7 without defining a relevant market. Several observations are warranted, however. First, we are obviously aware that the Supreme Court has held repeatedly that “[d]etermination of the relevant market is a necessary predicate to a finding of a violation of the Clayton Act” and that the Court has linked this requirement to the language of Section 7, which states that the plaintiff must establish that “in any line of commerce . . . in any section of the country, the effect of such [transaction] may be substantially to lessen competition.” See Brown Shoe, 370 U.S. at 324 (citations omitted); see also Marine Bancorp., 418 U.S. at 618.

More recently, however, courts have focused on the integral link link between market definition and the direct analysis of whether a a transaction will produce market power. See, e.g., Swedish Match, Match, 131 F. Supp. 2d at 156 (finding that market definition is “the “the key to the ultimate resolution of this type of case because of the the relative implications of market power”); Staples, 970 F. Supp. at at 1082 (“Much of the evidence already discussed with respect to defining the relevant product market also indicates that the merger would likely have an anti-competitive effect.”). In addition, while the courts appropriately have continued to rely on structural VOLUME 144 Opinion of the Commission presumptions derived from market definition, they also have placed much greater emphasis on the use of direct effects evidence. Thus, the D.C. Circuit noted in Baker Hughes that “[m]arket share is just a way of estimating market power, which is the ultimate consideration . . . [w]hen there are better ways to estimate market power, the court should use them.” 908 F.2d at 992 (quoting Ball Mem’l Hosp., Inc. v. Mutual Hosp., Inc., 784 F.2d 1325, 1336 (7th Cir. 1986)). Implicit in these decisions is the well-established principle that market definition is not an end in itself but rather an indirect means to assist in determining the presence or the likelihood of the exercise of market power. See Baker Hughes, 908 F.2d at 992; Toys “R” Us, Inc. v. FTC, 221 F.3d 928, 937 (7th Cir. 2000). As Professor Hovenkamp has explained in his treatise, “[m]arket structure evidence is the surrogate for bad performance, not the other way around.” Herbert Hovenkamp, Federal Antitrust Policy § 12.8, at 550 (3d ed. 2005).

Plainly, the enforcement agencies and courts need predictive tools and other inferential mechanisms to analyze market power in many merger cases. Market definition is one such type of tool. See Toys “R” Us, 221 F.3d at 937; see also Hovenkamp, supra, § 12.4c, at 524 (“Both concentration measures and estimates of market share are generalized attempts to predict the likelihood of anticompetitive behavior in the market.”). The role of the market definition tool, however, is potentially much less important in merger cases in which the availability of natural experiments allows for direct observation of the effects of competition between the merging parties, as well as the absence of such competition.

A line of modern cases brought under Section 1 of the Sherman Sherman Act is instructive. These courts have analyzed whether it is is appropriate to determine the lawfulness of ongoing or completed completed conduct through direct effects evidence, in lieu of market market definition. In FTC v. Indiana Federation of Dentists, 476 EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 517 Opinion of the Commission U.S. 447 (1986) (“IFD”), the Supreme Court reviewed an FTC decision that a dental association violated the antitrust laws by promulgating and enforcing a rule to withhold x-rays requested by dental insurers for use in claims evaluations. The association argued argued on appeal to the Supreme Court that the FTC’s decision was was wrong as a matter of law because the FTC had not specifically specifically defined the relevant market. Id. at 460. The Supreme Court disagreed, holding that product market analysis “is but a surrogate for detrimental effects.” Id. The Court further stated that “proof of actual detrimental effects, such as a reduction of output, can obviate the need for an inquiry into market power” through product market analysis. Id. at 460-61 (quoting VII Phillip Areeda, Areeda, Antitrust Law ¶ 1511, at 429 (1st ed. 1986)). A number of lower courts in Section 1 cases, relying on IFD, have held that it is appropriate to prove anticompetitive effects through direct evidence in place of market definition. In Toys “R” Us, the Seventh Circuit reviewed an FTC decision that held unlawful agreements between Toys “R” Us and a group of toy manufacturers in which each manufacturer promised to restrict distribution of its products to low-priced warehouse stores. Toys “R” Us argued that the Commission’s decision was deficient because the Commission had not established that the company had a large share of a relevant market. 221 F.3d at 937. The court of appeals rejected this claim, holding that the Commission’s direct evidence of anticompetitive effects was sufficient to establish an antitrust violation: [Toys “R” Us] seems to think that anticompetitive effects in a market cannot be shown unless the plaintiff, or here the Commission, first proves that it it has a large market share. This, however, has things things backwards. As we have explained elsewhere, elsewhere, the share a firm has in a properly defined defined relevant market is only a way of estimating market power, which is the ultimate consideration. The Supreme Court has made it clear that there are VOLUME 144 Opinion of the Commission two ways of proving market power. One is through direct evidence of anticompetitive effects. . . . The other, more conventional way, is by proving relevant relevant product and geographic markets and by showing that the defendant’s share exceeds whatever whatever threshold is important for the practice in the case.

Id. (citations omitted).102 While IFD and Toys “R” Us involved horizontal conduct that arguably was subject only to a “quick look,” courts have held that it is equally appropriate to use direct effects evidence in lieu of formal market definition in cases subject to a full rule of reason analysis. See, e.g., Todd v. Exxon Corp., 275 F.3d 191, 207 (2d Cir. 2001) (“[U]se of anticompetitive effects to demonstrate market power . . . is not limited to 'quick look’ or 'truncated’ rule of reason cases.”). We recognize that IFD and its progeny did not make a complete complete break from the market definition process. In each of these these cases, the courts also found that there was sufficient evidence evidence to identify at least the “rough contours” of the relevant product and geographic markets. See Republic Tobacco Co. v. North North Atlantic Trading Co., 381 F.3d 717, 736 (7th Cir. 2004). We also recognize that these cases did not involve Section 7. But this 102 See also Tops Markets, Inc. v. Quality Markets, Inc., 142 F.3d 90, 98 (2d Cir. 1998)(finding that market power “may be proven directly by evidence of the control of prices or the exclusion of competition, or it may be inferred from one firm’s large percentage share of the relevant market”); K.M.B. Warehouse Distributors, Inc. v. Walker Mfg. Co., 61 F.3d 123, 129 (2d Cir. 1995) (“If a plaintiff can show an actual adverse effect on competition, such as reduced output[,] . . . we do not require a further showing of market power.”) (citation omitted); Capital Imaging Assocs. v. Mohawk Valley Med. Assocs., 996 F.2d 537, 546 (2d Cir. 1993) (explaining that plaintiff may avoid a “'detailed market analysis’ by offering 'proof of actual detrimental effects, such as a reduction of output’”) (citation omitted).

EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 519 Opinion of the Commission does not negate the conceptual force of these decisions. None of these courts held that market definition was a necessary supplement supplement to the direct effects evidence. Rather, they endorsed the the use of direct effects evidence to determine, even absent a market market definition, whether ongoing conduct has facilitated the exercise of market power.

Antitrust doctrine is not static. See, e.g., State Oil Co. v. Khan, 522 U.S. 3 (1997) (overruling early decision that held that vertical maximum price fixing was per se violation of the Sherman Act). It is important that the antitrust laws be able to “adapt[] to changed circumstances and the lessons of accumulated experience.” Id. at 20. Consequently, we do not rule out the possibility that a future merger case may lead us to consider whether complaint counsel must always prove a relevant market.

VIII. REMEDY Having found that Evanston’s acquisition of Highland Park violated Section 7 of the Clayton Act, we turn to fashioning the appropriate remedy. The ALJ determined that ENH should divest Highland Park. ID 202-06. The ALJ also proposed a variety of other requirements intended to ensure that Highland Park would remain a viable hospital after divestiture and retain certain improvements that were implemented after the merger. ID 206-08. Complaint counsel argues that the Commission should affirm the ALJ’s order, but also cross-appeals and urges the Commission to add provisions that would require ENH to assist Highland Park in the continuation of its cardiac surgery program, provide incentives for ENH’s employees to accept job offers from Highland Park, and indemnify any monitor or trustee charged with overseeing the divestiture.

Respondent argues that, if we find liability, we should forgo ordering divestiture and instead should restore competition by requiring ENH to negotiate and maintain separate MCO contracts on VOLUME 144 Opinion of the Commission on behalf of Evanston on the one hand and Highland Park on the other. In conjunction, or in the alternative, respondent also suggests suggests that we could require ENH to give the Commission advance notification of any future acquisition or joint venture that ENH proposes to undertake.

The goal of a remedy for a Section 7 violation is to impose relief that is “necessary and appropriate in the public interest to eliminate the effects of the acquisition offensive to the statute.” United States v. E.I. du Pont de Nemours & Co., 353 U.S. 586, 607 (1957). Thus, we attempt to craft a remedy that will create a competitive environment that would have existed in the absence of the violations. In re RSR Corp., 88 F.T.C. 800, 893 (1976), aff’d, RSR Corp. v. FTC, 602 F.2d 1317 (9th Cir. 1979). “The antitrust laws would deserve little respect if they permitted those who violated them to escape with the fruits of their misconduct on the grounds that imposition of an effective remedy would incidentally result in even a substantial monetary loss.” RSR, 88 F.T.C. at 895. Structural remedies are preferred for Section 7 violations. See United States v. E.I. du Pont de Nemours & Co., 366 U.S. 316, 329 (1961) (calling divestiture “a natural remedy” when a merger violates the antitrust laws). As we recently said, “[m]uch of the case law has . . . found divestiture the most appropriate means for restoring competition lost as a consequence of a merger or acquisition.” In re Chicago Bridge & Iron Co., No. 9300, 2005 WL 120878, at 93 (FTC Jan. 6, 2005). Divestiture is desirable because, in general, a remedy is more likely to restore competition if the firms that engaged in pre-merger competition are not under common ownership. There are also usually greater long-term costs associated with monitoring the efficacy of a conduct remedy than with imposing a structural solution.

In this case, the transaction eliminated the pre-merger price competition between Evanston and Highland Park, as well as the EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 521 Opinion of the Commission MCOs’ option of contracting with one hospital but not the other. We can seek to remedy this competitive harm by requiring ENH to divest Highland Park or through injunctive restraints. After careful review of the record, we have determined that this is the highly unusual case in which a conduct remedy, rather than divestiture, is more appropriate.

A long time has elapsed between the closing of the merger and the conclusion of the litigation. This does not preclude the Commission from ordering divestiture, but it would make a divestiture much more difficult, with a greater risk of unforeseen costs and failure. ENH has integrated the operations of Evanston, Glenbrook, and Highland Park Hospitals, and has made improvements at Highland Park since the merger. The large majority of these improvements could have occurred without the merger, and therefore do not bear on whether the transaction violated Section 7. Nonetheless, while the improvements do not vindicate the merger under the antitrust laws, they are relevant to determining whether divestiture is appropriate because divestiture may reduce or eliminate the resulting benefits for a material period of time. Thus, we need to consider whether certain improvements would not survive the divestiture and would take Highland Park a significant time to implement on its own after a divestiture. Two significant improvements meet these conditions – the development and implementation of the cardiac surgery program and the implementation at Highland Park of EPIC, the state-of-the-art medical record computer system.

The record reflects that a divestiture may have a substantial negative effect on Highland Park’s cardiac surgery programs. Complaint counsel’s expert, Dr. Romano, testified that it was not clear whether, without Evanston, Highland Park would have the volume that it needed to maintain the cardiac surgery program. TR 3193 (Romano), in camera. If Highland Park lost its cardiac surgery program, or if the quality of its surgical program diminished, then VOLUME 144 Opinion of the Commission the quality of patient care to the community would suffer. Highland Park would need to transport some or all of its patients needing emergency cardiac surgery to other hospitals, potentially creating life-threatening risks. TR 5612-13 (Chassin); TR 4457 (Rosengart). The possibility of a delay in reestablishing cardiac surgery services at Highland Park is a significant factor that we must weigh in considering a remedy.

A delay in reestablishing Highland Park’s cardiac surgery program also could put at risk Highland Park’s interventional cardiology services. An interventional cardiology program involves procedures that may be scheduled in advance. To provide interventional cardiology services, however, it is necessary to have a cardiac surgery program as a back-up for the interventional program if complications occur. TR 5306-07 (Chassin). We are also concerned about the effect of divestiture on Highland Park’s ability to use EPIC. Although the implementation of the EPIC system at Highland Park was not a merger-specific efficiency, it likely would take Highland Park significant time to install EPIC (or a comparable record keeping system) independently, at a cost of millions of dollars if we ordered divestiture. ENH spent approximately $14 million on EPIC and took more than one year to deploy the system fully. TR 1984 (Hillebrand); TR 1251, 1355 (Neaman); TR 3523 (O’Brien); TR 3976, 3987-88 (Wagner). We could order ENH to continue to make EPIC available to Highland Park for some time, but we are concerned about the potential effects on patient care from the inevitable glitches involved in Highland Park’s swapping out complex software systems. Accordingly, we reject divestiture as a remedy and will impose impose an injunctive remedy that requires respondent to establish separate and independent negotiating teams – one for Evanston and and Glenbrook Hospitals (“E&G”), and another for Highland Park. While not ideal, this remedy will allow MCOs to negotiate EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 523 Opinion of the Commission separately again for these competing hospitals, thus re-injecting competition between them for the business of MCOs. Further, ENH ENH should be able to implement the required modifications to its its contract negotiating procedures in a very short time. In contrast, contrast, divesting Highland Park after seven years of integration would be a complex, lengthy, and expensive process. We note that our rationale for not requiring a divestiture in this case is likely to have little applicability to our consideration of the proper remedy in a future challenge to an unconsummated merger, including a hospital merger. For example, had we challenged this transaction prior to consummation, Evanston’s intention to implement a cardiac surgery program and install EPIC at Highland Park likely would not have carried much weight in our analysis of the proper remedy because, at that time, Highland Park probably could have produced both improvements on its own in a comparable period, and thus neither improvement would have been mergerspecific.

Nor will our reasoning here necessarily apply to consideration of the appropriate remedy in a future challenge to a consummated merger, including a consummated hospital merger. Divestiture is the preferred remedy for challenges to unlawful mergers, regardless of whether the challenge occurs before or after consummation. Thus, where it is relatively clear that the unwinding of a hospital merger would be unlikely to involve substantial costs, all else being equal, the Commission likely would select divestiture as the remedy. Although we have decided on the nature of the relief that is appropriate for this case, we lack sufficiently detailed information about the personnel involved in ENH’s contract negotiation operations, or ENH’s overall business operations, to craft the remedial order with the necessary precision. Accordingly, we order order that, within thirty (30) calendar days, respondent must submit a detailed proposal to the Commission for implementing the type of injunctive relief that we have selected. Specifically, the proposal VOLUME 144 Opinion of the Commission must identify and describe the mechanisms that respondent will use, and the steps that respondent will take, to implement the following requirements:

1. Respondent must allow all payors to negotiate separate contracts for E&G on the one hand and for Highland Park on the other hand;

2. Respondent must establish separate negotiating teams (and other relevant personnel) for E&G and Highland Park that will compete with each other, and other hospitals, for payors’ business;

3. Respondent must establish a firewall-type mechanism that prevents the E&G and Highland Park contract negotiating teams (and other relevant personnel) from sharing any information that would inhibit them from competing with each other and with other hospitals;

4. Respondent may not make any contract for E&G or Highland Park contingent on entering into a contract for the other, and may not make the availability of any price or term for a contract for E&G contingent on entering into a contract for Highland Park, or vice-versa; and 5. Respondent shall promptly offer all payors with which it currently has contracts the option of reopening and renegotiating their contracts under the terms of this order. Respondent’s proposal should also describe, where appropriate, appropriate, mechanisms for the Commission to monitor the establishment of the organizational structure needed to implement the terms of the order, as well as Respondent’s compliance with the the order throughout its term. Respondent’s proposal shall also recommend mechanisms for resolving disputes between payors and and respondent with respect to Respondent’s compliance with the EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 525 Concurring Statement terms of the order, including a discussion of the potential value of some form of dispute resolution mechanism. Complaint counsel must submit any objections to or comments on Respondent’s proposal within thirty (30) calendar days after respondent submits its proposal. Respondent may, if it chooses, respond to complaint counsel’s filing within ten (10) calendar days. Concurring Opinion of Commissioner J. Thomas Rosch I concur with the Commission opinion’s conclusion that Evanston Northwestern Healthcare Corp.’s acquisition of Highland Park Hospital violated Section 7 of the Clayton Act. There is much to be admired in the Commission opinion. However, particularly in light of Count II of the complaint, I believe the Commission opinion makes this case more difficult than necessary. I write separately to explain why that is so.

I depart from the Commission opinion in two fundamental respects. First, I believe the law and the facts in this case squarely support complaint counsel’s theory of anticompetitive effects. That theory is based on the unique competitive dynamics of hospital markets, stemming from the bargaining between hospitals and managed care organizations (“MCOs”) over inclusion in MCO networks that is described by the Commission opinion. See Comm. Op. 62-63 (describing the “bargaining model” involved here). Reflecting those dynamics, complaint counsel’s theory of anticompetitive effects is multi-dimensional. VOLUME 144 Concurring Statement At one level, notwithstanding the physical and geographical differences between Evanston and Highland Park,1 the two hospitals competed with each other in the sense that MCOs wanting to compete effectively for insureds located within the geographic triangle bounded by the three ENH hospitals viewed Evanston and Highland Park as each other’s “next best substitute” in forming networks for that purpose. Complaint counsel’s theory of anticompetitive effects was initially based on the merger’s elimination of this dimension of competition, which enabled ENH to include Highland Park in its system and engage in system (all-ornothing) supra-competitive pricing. See CPTB 4, 24; CB 14, 19-21. At a second level, premerger, Evanston and Highland Park were also constrained in their pricing to MCOs by localized competition – i.e., by the hospitals located closest to each of the two hospitals. Complaint counsel contends that the merger eliminated (or at least crippled) this localized competition because after the merger an MCO had to contract with all ENH hospitals in order to include one of them in its network, and that inhibited MCOs from playing Evanston and Highland Park off against their nearby competitors, as they could do pre-merger.

This multi-dimensional theory of liability, while unusual, is by no means unique. The Commission relied on similar theories when it challenged Rite-Aid’s attempt to acquire Revco and Time Warner Inc.’s proposed acquisition of Turner Broadcasting System, Inc. See Jonathan B. Baker, Unilateral Competitive Effects Theories in Merger Analysis, 11 Antitrust 21, 24-25 (1997). 1 Evanston Hospital and Highland Park are located 13.7 miles apart and each has other hospitals located closer to it than they are to each other (though there are no other hospitals located within the geographic triangle formed between the two Evanston hospitals and Highland Park), and Evanston is a teaching tertiary care hospital, while Highland Park is a non-teaching, community primarysecondary care hospital. However, as the Commission opinion notes, the two Evanston hospitals and Highland Park each provided primary-secondary care services. Comm. Op. 12, 26, 72-73 EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 527 Concurring Statement Second, I believe the evidence that these unilateral anticompetitive effects have actually occurred has a significant impact on market definition. Specifically, the fact that this is a consummated merger means that ours is a retrospective analysis. We can look to see if there is any probative post-merger evidence that demonstrates whether or not the merger has been anticompetitive. We do not need to try to predict the future as would be necessary to analyze an unconsummated merger proposal. Where, as here, the post-transaction record establishes that the transaction has produced unilateral anticompetitive effects, it is not essential to define the relevant market upfront using the methodology described in the Horizontal Merger Guidelines. At least the “rough contours” of the relevant market can be identified on the basis of those effects, and that is sufficient as a matter of law.

I. Anticompetitive Effects There is no dispute that immediately after the merger ENH increased prices for services at the ENH hospitals to a number of MCOs by many times the five percent increases described in the Horizontal Merger Guidelines as “significant” and that those price increases were “non transitory.” See U.S. Dept of Justice & Federal Trade Commu, Horizontal Merger Guidelines §1.11 (1992, revised 1997) (“Merger Guidelines”); Comm. Op. 16-17, 26-27, 64-65, 78. There is, however, a dispute over the cause of those price increases. Complaint counsel argued that this transaction created a hospital system that could and did obtain supra-competitive prices. See CPTB 2-4, 33-34; CB 14, 19-21. Respondent argued that its price increases at Evanston Hospital simply reflected its efforts to raise Evanston’s prices to a competitive level, and did not reflect supracompetitive pricing. RB 3. It also maintained that the post-merger price increases at Highland Park reflected an increase in the quality of services provided at the hospital. RB 3-5. VOLUME 144 Concurring Statement I conclude that the merger materially changed the competitive dynamics that had theretofore existed in a fashion that violated Section 7.

A. The Law Complaint counsel’s theory of anticompetitive effects is viable as a matter of law. Respondent asserts that the Merger Guidelines description of unilateral anticompetitive effects in mergers involving differentiated products (Sections 2.21 and 2.211) does not squarely adopt that theory. RB 39-43; RRB 11, 22 (discussing Sections 2.21 and 2.211 and In re R.R. Donnelly & Sons Co., 120 F.T.C. 36 (1995), embracing these provisions). I agree with respondent that the unilateral effects of a merger between producers of differentiated products (or services, like hospital services) cannot be considered illegal in all circumstances. I also agree that Sections 2.21 and 2.211 provide valuable guidance respecting the appropriate parameters. However, I disagree with respondent that complaint counsel’s theory is outside those parameters.

The fundamental teaching of Sections 2.21 and 2.211 is that anticompetitive effects are likely when differentiated products of the merging parties are each other’s next best substitute.2 To be sure, those provisions might not apply if the merger eliminated only premerger localized competition, considered in isolation. In that dimension of competition, Evanston and Highland Park were arguably not each other’s best alternative within the meaning of Sections 2.21 and 2.211 of the Merger Guidelines. Advocate Lutheran General was arguably Evanston’s closest local competitor and Lake Forest was arguably the closest alternative to Highland Park.

2 The provisions also establish a safe harbor when the merger could not result in substantial market power. But under complaint counsel’s theory, after the merger the merging hospitals here enjoyed substantial market power. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 529 Concurring Statement However, under complaint counsel’s theory the merger’s impact on localized competition cannot be considered in isolation. It was the consequence of the merger’s primary effect, which was to eliminate competition between Evanston and Highland Park for inclusion in MCO hospital networks. To be specific, under complaint counsel’s theory, before the merger MCOs who wanted to compete effectively for insureds located within the triangle considered Evanston and Highland Park to be each other’s “next best substitute” in forming a network for that purpose, and the merger eliminated the competition between those next best substitutes. The lessening of the localized dimension of competition is an ancillary anticompetitive effect of the merger because the elimination of that dimension of competition resulted from the merger’s elimination of competition between those next best substitutes. Thus, the unilateral effects provisions of the Merger Guidelines apply if the record sufficiently demonstrates that the transaction has had those anticompetitive effects. This application of Sections 2.21 and 2.211 is not blunted by the language in Section 2.21 stating that “[t]he price rise will be greater the closer substitutes are the products of the merging firms.” As the rest of that sentence makes clear, even products that are highly differentiated in terms of their physical and locational differences can be considered to be close substitutes with each other if “buyers of one product consider the other product to be their next choice.” Thus, the elimination of the first dimension of competition – the competition between Evanston and Highland Park resulting from MCOs’ desire to include one or the other of them in their networks – would represent an elimination of “close substitutes” within the meaning of Sections 2.21 and 2.211. And, since under complaint counsel’s theory the injury to the second dimension of competition – the localized competition between each of the merging hospitals and its geographically more proximate rivals – was a consequence of the elimination of competition between those “close substitutes,” those provisions of the Merger Guidelines would apply to that injury as well.

VOLUME 144 Concurring Statement Conceptually, the effect of the elimination of the competition between Evanston and Highland Park is the same as if Evanston and Highland Park had entered into an agreement with each other as to the prices they would charge MCOs (or to be more blunt if they had entered into a price-fixing agreement). To be sure, a marketing joint venture could produce a similar result. We tolerate a marketing joint venture when it is shown to produce a new product that would not otherwise exist, absent the collaboration, and if it is shown that the joint venture will produce efficiencies that outweigh any pricing effects. See Federal Trade Commu & U.S. Dept of Justice, Antitrust Guidelines for Collaborations Among Competitors §§ 3.3, 3.36 (2000). However, under complaint counsel’s theory, neither can be said about this transaction. Before the transaction, MCOs could, if they chose to do so, create a hospital network like the post-merger system of ENH hospitals by bargaining with Evanston and Highland Park individually (and playing each off against the hospitals proximate to each – for example, Lake Forest in the case of Highland Park). Thus, the merger was not necessary to produce that kind of hospital network. Moreover, according to complaint counsel (and the Commission opinion), respondent failed to prove that any claimed efficiencies outweighed the pricing effects of the merger. Most significantly, complaint counsel’s theory fits snugly within the language of Section 7 of the Clayton Act. That provision prohibits any merger that has the effect of substantially lessening competition. Under complaint counsel’s theory, this merger had the effect of substantially lessening both dimensions of the pre-merger hospital competition that MCOs could take advantage of in fashioning a network that would be attractive to insureds located within the triangle.

B. The Facts As a factual matter, complaint counsel’s view of the anticompetitive effects in this case is supported by MCO testimony, testimony of Respondent’s own principal economic expert, Professor EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 531 Concurring Statement Jonathan Baker, and the documents and testimony of the merging parties. MCO representatives described the pre-merger dynamics of competition among hospitals for inclusion in MCO networks. They testified that prior to the merger, MCOs wanting to compete effectively for insureds located within the geographic triangle formed by Evanston and Highland Park viewed those hospitals as “close substitutes” for each other when forming networks for that purpose. See Comm. Op. 18-25, 78.

For example, Jane Ballengee, PHCS’ Regional Vice President for Network Development, testified that before the merger PHCS “could have one or the other hospital in their network.” CB 21 (citing TR 166-67 (Ballengee)). Robert Mendonsa, a formal general manger at Aetna, testified that before the merger Evanston was “extremely desirable” and that Aetna’s “walk-away point would have been pretty high . . . [but that Aetna] would have walked away because we still had Highland Park and we had Northwestern in the city and we had that coverage.” TR 530 (Mendonsa), in camera. United’s Jillian Foucre testified that Evanston and Highland Park would be the preferred choices of executives who lived in the triangle made up by the North Shore suburbs, and that executives who lived “within that area” made up by the triangle would not want to travel greater distances north or south to go to hospitals. TR 901-02 (Foucre). Foucre managed a team who negotiated with United’s network providers in the Chicago area. TR 879 (Foucre); CX 5174, in camera.3 Additionally, MCO representatives testified that prior to the merger there was another dimension of competition: Evanston and Highland Park were also constrained in their pricing to MCOs by localized competition – i.e., the hospitals located close to each (and not each other). For example, Foucre also testified that, prior to the merger, she viewed Condell and Lake Forest as competitors to 3 Patrick Neary, who at the time of the merger was One Health’s Director of Network Development, also testified that he thought that ENH had purchased “their main competitor,” although he did not specify why this was the case. TR 600-01 (Neary).

VOLUME 144 Concurring Statement Highland Park, and that Evanston competed with Advocate Lutheran General, Rush North Shore, and St. Francis. TR 941-44 (Foucre). Lenore Holt-Darcy, Unicare’s Regional Vice President at the time of the merger, testified that Highland Park competed with Lake Forest and Condell hospitals, and that Evanston competed with a significant number of tertiary hospitals in the Chicago area, including Rush North Shore and St. Francis. TR 1595-97 (Holt- Darcy), in camera.

The changes in these competitive dynamics are directly reflected in the post-merger pricing applicable to ENH’s hospitals. It is undisputed that after the merger ENH negotiated a system contract for all three of its hospitals; MCOs were not given the option of entering into separate contracts for the hospitals, to decline to use one or more of the hospitals, or to pay different prices for the care at any one of them. See IDF ¶ 449; TR 1528 (Holt-Darcy), in camera (Post-merger, ENH offered an “all-or-nothing deal” to Unicare in which there would be one rate for all three hospitals, regardless of the level of service at each facility, like the “Three Musketeers, all for one and one for all.”); see also Comm. Op. 16. Furthermore, as the Commission opinion says, economic evidence proffered by Professor Baker shows that immediately after the merger, the system prices that ENH charged a number of MCOs increased by many times more than the five percent described in the Merger Guidelines as “significant.” Comm. Op. 17, 27, 64-65. Beyond that, the record refutes Respondent’s efforts to explain those price increases by factors divorced from the merger itself. First, while respondent claims that the pricing at Highland Park was attributable to the improvements that ENH made there, the record shows that price increases were imposed before these improvements were made. IDF ¶¶ 179, 457. Second, the econometric evidence presented by Professor Baker itself contradicted Respondent’s claim that the price increases only brought Evanston’s prices up to competitive levels: as the Commission opinion says, the “control group” of hospitals against whose price increases Professor Baker EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 533 Concurring Statement compared ENH’s post-transaction price increases contained only high-end very expensive hospitals that were not comparable to Evanston or Highland Park, Comm. Op. 39, 43-44, 69; IDF ¶¶ 817- 19, 821, 824, and indeed, for several MCOs ENH’s price increases exceeded even that “control group’s” price increases. Comm. Op. 44.

The evidence of post-merger supra-competitive pricing at Highland Park is especially compelling. The evidence is undisputed that tertiary care teaching hospitals command substantially higher prices than do primary-secondary care community hospitals like Highland Park. RB 17-18, RRB 36-37; TR 156-59 (Ballengee). Evanston was (and is) indisputably a tertiary care teaching hospital, but Highland Park was (and is) indisputably a community hospital, not a tertiary care teaching hospital. CB 54, n.57; ID 191.4 Indeed, respondent repeatedly emphasized how different Evanston and Highland Park were from each other, RB 2, 7, 9, 10; RRB 28 n.6, 36, and also admitted that tertiary care teaching hospitals like Evanston command higher prices than primary-secondary care community hospitals like Highland Park. RB 17-18, 51; RRB 36-37. These admissions by respondent and its expert were not gratuitous. They were amply supported by MCO testimony. See TR 158-59 (Ballengee); TR 622 (Neary); TR 935, 1112 (Foucre), in camera; TR 565 (Mendonsa), in camera; TR 1289 (Neaman), in camera; TR 1590 (Holt-Darcy), in camera; see also RRB 36-37; RB 51; TR 6065 (Noether), in camera. Thus, to borrow an economics term, the demand curves for teaching and community hospitals are materially different from one another, and as a consequence teaching hospitals can and do charge more for their services. 4 While Evanston is not at the highest end of the teaching hospital spectrum (see supra p. 5), it is undeniably a teaching hospital. See Comm. Op. 43-44; CB 45. Evanston is affiliated with the Northwestern University School of Medicine, and this relationship was strengthened between 1992 and 1996. TR 1282 (Neaman); RX 584 at ENH JH 2951-52.

VOLUME 144 Concurring Statement Yet respondent has essentially admitted that, post-transaction, MCOs were charged the same prices for Highland Park’s services that they were charged for Evanston’s services, and Respondent’s expert testified that Evanston’s prices were “at the middle of the pack” of Chicago area academic hospitals. TR 6065-66 (Noether), in camera; RB 91-92. That ENH could and did charge teaching (and tertiary care) hospital prices at Highland Park is direct evidence that, as a result of the merger, it enjoyed and exercised market power sufficient to impose supra-competitive prices. See ID 171-72 (“[E]ven if the evidence demonstrates that Evanston deserved higher prices because of its teaching status, this does not provide any justification for charging the same higher rates for Highland Park, a non-teaching community hospital.”); see also CB 47 n.49. Finally, the record establishes that ENH did not suffer a “critical loss” – or indeed any loss – of sales to competing hospitals as a result of its price increases. Notwithstanding ENH’s system-wide pricing at significantly increased prices, only one MCO (OneHealth) initially did not contract with ENH, and OneHealth ended up contracting with ENH after it concluded it could not afford to refrain from doing so. IDF ¶¶ 420-33. Thus, the record establishes that the price increases were the result of post-merger market power rather than of exogenous factors.

Indeed, respondent itself has said that ENH experienced no loss of business to competitors after the merger, citing the absence of any such output reduction as a reason why its price increases cannot be considered to be the product of an exercise of market power, as a matter of economics and law. RB 56; RRB 23-25. To borrow (respectfully) from Judge Diane Wood, this claim “has things backward.” Toys “R” Us, Inc. v. FTC, 221 F.3d 928, 937 (7th Cir. 2000). Where, as here, there is evidence that the defendant has increased its prices significantly and the defendant’s output does not decline, this in and of itself is evidence that the defendant enjoys market power. See R.J. Reynolds Tobacco Company v. Cigarettes Cheaper!, 462 F.3d 690, 695 (7th Cir. 2006) (Easterbrook, J.) EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 535 Concurring Statement (rejecting the district court’s ruling on summary judgment that the defendant lacked market power because inter alia, the summary judgment record did not demonstrate that the defendant “lacks power to make significant price increases without substantial loss of sales”); Olympia Equip. Leasing Co. v. Western Union Tel. Co., 797 F.2d 370, 373 (7th Cir. 1986) (Posner, J.) (market power is “the power to raise prices without losing so much business that the price increase is unprofitable”); see also William M. Landes & Richard A. Posner, Market Power in Antitrust Cases, 94 Harv. L. Rev. 937, 939 (1981).

The fact that ENH did not suffer any loss of business to other competitive hospitals in the face of its post-merger pricing also rebuts Respondent’s assertion that MCOs could easily assemble a hospital network excluding the ENH hospitals by contracting with a system of hospitals like the Advocate Lutheran system to serve insureds located within the triangle and thus constrain the postmerger pricing of the ENH system. RB 46. Respondent has relied for this assertion on the ALJ’s conclusion that Advocate Lutheran could constrain ENH’s pricing. ID 144, 149. However, that conclusion is not inconsistent with the ALJ’s ultimate conclusion that the merger violated Section 7. The hospitals in a network excluding the ENH hospitals would be more distant from the triangle than the ENH hospitals, and, as such, that network would be an imperfect substitute at best.

The existence of an imperfect substitute might constrain ENH’s pricing somewhat. However, the case law recognizes that even firms enjoying monopoly power may be somewhat constrained in their pricing by other products; that constraint does not mean that the firm lacks monopoly power. United States v. Aluminum Co. of Am., 148 F.2d 416, 426 (2d Cir. 1945); see also IIA Phillip E. Areeda, Herbert Hovenkamp & John Solow, Antitrust Law, ¶ 506a, at 104-05 (2d ed. 2002) (“IIA Areeda, Hovenkamp & Solow”). Indeed, Professor Baker has written that the imperfect substitutes in the proposed RiteAid-Revco and TimeWarner-Turner Broadcasting mergers discussed above were not considered to constrain post-merger VOLUME 144 Concurring Statement market power (and hence pricing) sufficiently to avoid Section 7 liability. See Baker, supra, at 24-25. Similarly here, the record establishes that the ability of MCOs to assemble a network of non- ENH hospital systems did not prevent ENH from pricing the hospitals in its system at supra-competitive levels. Nor does the direct evidence of post-transaction supracompetitive pricing stand alone. It is supported by the evidence described by the Commission opinion that senior officials at Evanston and Highland Park anticipated that the merger would enable them to raise their prices, that the merged firm did in fact implement an extraordinary price increase immediately after completion of the transaction, and that the same senior officials then attributed their success at raising prices to increased bargaining leverage produced by the merger. Comm. Op. 14-18, 65-67. II. Market Definition The Commission opinion also makes the market definition question more difficult than it needs to be in this case. As the Commission opinion says, Count II of the complaint in this case raised the question whether it is always necessary to define the relevant market in a Section 7 challenge at the time and in the fashion described in the Merger Guidelines. Comm. Op. 86. In proceeding under Count II complaint counsel did not define a market upfront using the Merger Guidelines methodology. Rather, it relied instead primarily on the direct evidence of the transaction’s anticompetitive effects, in accordance with Count II. CB 5. I agree with complaint counsel that especially when a merger has been consummated and the evidence shows it has had actual anticompetitive unilateral effects, the law allows liability to be established by direct evidence of those effects, without initially defining a relevant market using Merger Guidelines methodology, at least where, as here, the evidence of anticompetitive effects identifies the “rough contours” of the market. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 537 Concurring Statement A. The Law The Commission opinion articulately describes the trend in the courts towards greater reliance on direct evidence in defining markets. Comm. Op. 86-88. In cases brought under Section 1 of the Sherman Act, the courts have analyzed the analogous issue of whether it is appropriate to determine the lawfulness of completed or ongoing conduct through direct effects evidence, in lieu of market definition. See id. (discussing FTC v. Indiana Fed’n of Dentists, 476 U.S. 447 (1986) (“IFD”)); Toys “R” Us, 221 F.3d 928; Todd v. Exxon Corp., 275 F.3d 191, 207 (2d Cir. 2001); Ball Mem’l Hosp. v. Mutual Hosp. Ins., 784 F.2d 1325, 1336 (7th Cir. 1986)). The purpose of market definition and the direct analysis of anticompetitive effects are consistent – both techniques seek to determine whether a planned agreement by competitors is likely to facilitate the exercise of market power, or whether a completed one has enabled the exercise of market power. See Toys “R” Us, 221 F.3d at 937. As the Commission opinion observes, for more than a decade the courts and scholars have recognized repeatedly that market definition is not an end in itself but rather an indirect means to assist in determining the presence or the likelihood of market power. Comm. Op. 86-88; see also United States v. Baker Hughes, Inc., 908 F.2d 981, 992 (D.C. Cir. 1990); IIA Areeda, Hovenkamp & Solow, supra, ¶ 532a, at 190-91; Herbert Hovenkamp, Federal Antitrust Policy § 12.8, at 550 (3d ed. 2005) (“Hovenkamp”). Market definition is a tool for analyzing market power, but it is not the only tool, either as a matter of law or economics. Toys “R” Us, 221 F.3d at 937.

As the Commission opinion also says, enforcement agencies and courts often need predictive tools like market definition in order to analyze market power in unconsummated merger cases because the transaction has not yet occurred. See Hovenkamp, supra, § 12.4c, at 524-25. However, challenges to consummated mergers do not necessarily require predictive or inferential mechanisms because there may be a rich amount of empirical evidence that shows a VOLUME 144 Concurring Statement transaction’s actual anticompetitive effects. To the contrary, it would make no sense to adopt a rule providing that even when there is clear direct evidence that a consummated transaction has enabled the merged party to engage in supra-competitive pricing, the enforcement agency must nonetheless define with precision the relevant market upfront in order to establish liability under Section 7.

Like the Commission opinion, I recognize that IFD and its progeny did not “make a complete break from the market definition process.” Comm. Op. at 88. In each of those cases, the courts also found there was sufficient evidence to identify at least the “rough contours” of the relevant markets. Id. The Section 1 cases discussed by the Commission opinion permitted the use of direct effects evidence in order to determine whether ongoing conduct has facilitated the exercise of market power so long as the rough contours of the relevant market are identified. There is no principled reason why the same analysis cannot be applied in Section 7 cases. Indeed, a decade and a half ago, Judge Posner observed that judicial interpretation of Section 1 of the Sherman Act and Section 7 of the Clayton Act had converged. United States v. Rockford Memorial Corp., 898 F.2d 1278, 1281-83 (7th Cir. 1990); see also IV Phillip E. Areeda, Herbert Hovenkamp & John L. Solow, Antitrust Law, ¶ 913b, at 64 (2nd ed. 2006) (“In cases where a merger facilitates a significant 'unilateral’ price increase for a grouping of sales that was not a distinctive-looking market prior to the merger, the appropriate conclusion is that the merger has facilitated the emergence of a new grouping of sales capable of being classified as a relevant market. This formulation meets the statutory requirement [in Section 7] that the ‘effect’ of a merger is anticompetitive in some 'line of commerce’ in some ‘section of the country.’”); Comm. Op. 60-62 (citing authorities). To be sure, a number of merger decisions – including several involving hospitals – have required that the relevant market be EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 539 Concurring Statement defined upfront and with precision. Indeed, several courts have rejected challenges to hospital mergers on the ground that the plaintiff failed to properly define the relevant market this way. See, e.g., FTC v. Freeman Hospital, 69 F.3d 260 (8th Cir. 1995); California v. Sutter Health Sys., 84 F. Supp. 2d 1057 (N.D. Cal. 2000). However, the mergers in cases post-dating enactment of the Hart-Scott-Rodino (“HSR”) Act three decades ago have been unconsummated mergers. Consequently, the analysis had to be prospective. The agencies and the courts had to predict what the consequences of the transaction would be. That is a different task than the task in a consummated merger case like this one. As previously discussed, predictive tools, such as market definition, are less necessary in a consummated case when we can determine by direct evidence whether the merger enabled the combined firm to engage in anticompetitive conduct.

This is not to say that the post-transaction behavior in this case lacks significance in future unconsummated hospital merger cases. To the contrary, it may be that the experience in this case will be important in predicting the likely effects in certain of those cases. Moreover, evidence of likely post-transaction anticompetitive effects may sufficiently identify the contours of the relevant market in other unconsummated merger cases. See Comm. Op. at 61-62, discussing, inter alia, FTC v. Staples, Inc., 970 F. Supp. 1066 (D. D.C. 1997), an unconsummated merger case in which the relevant product market was defined principally on the basis of the evidence of likely anticompetitive effects. At all events, the evidence of actual anticompetitive effects that exists in this case distinguishes it from all of the cases requiring upfront and precise market definition, including consummated merger cases pre-dating enactment of the HSR Act.5 5 Also, in most of the pre-HSR Act merger cases requiring upfront and precise market definition the theory of anticompetitive effects has been a “coordinated effects” theory – i.e., that the merger threatened to facilitate coordination in a highly concentrated market. See, e.g., United States v. Philadelphia Natl Bank & Trust Co., 374 U.S. 321 (1963); see also FTC v. Elders Grain, Inc., 868 F.2d 901, 906 (7th Cir. 1989) (Posner, J.) (describing VOLUME 144 Concurring Statement In short, I believe that as a matter of law, it was not necessary that anything more than the “rough contours” of the relevant market be defined in order to establish the existence of a Section 7 violation in this case, where complaint counsel’s theory of anticompetitive effects could be tested because the merger had been consummated. The evidence shows that this consummated merger enabled the merged firm unilaterally to engage in supra-competitive pricing, and that fact supports the propriety of relying on direct evidence in defining the rough contours of the relevant market.6 B. The Facts In this case, Respondent’s documents and economic evidence described above, as well as the testimony of MCOs previously described, not only established the existence of anticompetitive effects resulting from the merger, but also identified at least the “rough contours” of the product and geographic markets alleged by complaint counsel. More specifically, complaint counsel asserted that the relevant product market is “general acute care hospital services, including primary, secondary, and tertiary services, sold to MCOs.” CB 37. Complaint counsel contended that the relevant geographic market was the triangle bounded by the three hospitals in the ENH system. CB 38; ID 137.

As Areeda and Hovenkamp explain, a relevant market is “a market relevant to the particular legal issue being litigated.” IIA Areeda, Hovenkamp & Solow, supra, ¶ 533c. Here the issue is coordinated effects as the prevailing theory of anticompetitive effects in merger cases). As the Commission opinion points out, when that is the theory, it is important that all the competitors in the market be identified. Comm. Op. 59. 6 Of course, if anticompetitive effects have not yet occurred because the merged party is aware of the antitrust risks of engaging in post-transaction anticompetitive conduct, or for some other reason, the upfront market definition methodology described in the Merger Guidelines may be useful to predict whether or not they are likely to occur in the future. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 541 Concurring Statement whether the merger enabled ENH to impose supra-competitive prices on MCOs who wished to compete effectively for insureds located within the geographic triangle bounded by the three ENH hospitals. I agree with the Commission opinion that the relevant product market in this case is acute inpatient services, which hospitals alone can provide. As the Commission opinion points out, the record in this respect is consistent with the long line of cases that have reached the same conclusion. Comm. Op. 56. I also conclude that complaint counsel demonstrated that the relevant geographic market consisted of the triangle bounded by the three ENH hospitals. That conclusion is based on the evidence previously described that MCOs considered Evanston or Highland Park to be next best substitutes in forming networks in order to compete effectively for insureds located within that triangle. See supra p. 4. That conclusion is also based on the evidence previously described that after the merger, ENH gained the power to control the price of all three ENH hospitals, and ENH enjoyed and exercised this market power to impose extraordinarily high system prices on MCOs as the price for their effective competition in that geographic area. See supra pp. 5-7; CB 14, 19-21. And it is based on the evidence that, despite ENH’s post-transaction system pricing and despite the extraordinarily high pricing that occurred at all three ENH system hospitals, none of the MCOs competing in that triangle ultimately declined to deal with ENH.

Again, respondent did not contest that the three ENH hospitals were uniquely located with respect to that triangle, or that ENH could and did engage in system pricing after the merger. Respondent instead argued that the triangle did not constitute the relevant geographic market because each of the ENH hospitals was located closer to other hospitals than to each other and that the pricing at these other hospitals would constrain the pricing at each. RB 2, 10. That is a non sequitur. It is correct that at one level of competition, prior to the transaction the pricing at Evanston and Highland Park was constrained by other hospitals that were located proximate to each. But that does not mean that same competitive constraint VOLUME 144 Concurring Statement existed after the merger, when MCOs were forced to contract with all three ENH hospitals on ENH’s terms, instead of confronting each constituent hospital with the local competition each faced, as MCOs could do before the merger. Indeed, Respondent’s argument simply underscores that injury to that localized pre-merger competition is another consequence of the merger, which strengthens the conclusion that the competitive forces affecting pricing vis-à-vis the triangle were lessened as a result of the merger. In short, what the record demonstrates is that, as complaint counsel has claimed, the merger had the effect of lessening competition in a relevant market consisting of primary, secondary, and/or tertiary inpatient hospital care services in the triangular area bounded by the ENH hospitals. ENH’s control of all three hospitals in the triangle enabled it to impose supra-competitive prices for inpatient hospital care services that could not have been charged prior to the merger when the hospitals forming the triangle bargained separately.

I would affirm for these reasons, and I agree with the Commission opinion’s relief order.

CONCURRING OPINION OF COMMISSIONER JON LEIBOWITZ I join the opinion of the Commission with respect to its findings of fact, its conclusion that the merger violated the Clayton Act as identified in Count 1 of the Complaint, and the remedy identified in that opinion. However, I believe that the weight of the evidence clearly supports a finding that the merger violated the Clayton Act in the manner identified in Count 2 of the Complaint as well. EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 543 Final Order Consequently, I join in Section II of Commissioner Rosch’s concurrence.

FINAL ORDER This matter having been heard by the Commission upon the appeal of Respondents and the cross-appeal of Complaint Counsel, and upon the respective briefs and oral arguments in support of such positions, and the Commission having determined that Respondent Evanston Northwestern Healthcare Corporation (Respondent ENH) has violated Section 7 of the Clayton Act – for the reasons stated in the accompanying Opinion – the Commission has determined to require Respondent ENH to cease and desist from certain enumerated practices, and to require Respondent ENH to propose, for issuance by the Commission, a Final Order that conforms to the prescriptions of this Order. Accordingly, IT IS ORDERED THAT the determination in the Initial Decision in this matter that the transaction at issue violated Section 7 of the Clayton Act is AFFIRMED;

IT IS FURTHER ORDERED THAT the Order issued as part of the Initial Decision in this matter, be, and it hereby is, VACATED;

IT IS FURTHER ORDERED THAT on or before September 10, 2007, Respondent ENH shall file with the Commission a detailed proposal for implementing the type of injunctive relief that the Commission has selected. Specifically, as prescribed in the Opinion of the Commission, the proposal shall identify and describe the mechanisms that Respondent ENH will use, and the steps that VOLUME 144 Final Order Respondent ENH will take, to implement the following requirements:

1. Respondent ENH must allow all payors to negotiate separate contracts for Evanston & Glenbrook Hospitals (E&G) on the one hand and for Highland Park on the other hand; 2. Respondent ENH must establish separate negotiating teams (and other relevant personnel) for E&G and Highland Park that will compete with each other, and other hospitals, for payors’ business;

3. Respondent ENH must establish a firewall-type mechanism that prevents the E&G and Highland Park contract negotiating teams (and other relevant personnel) from sharing any information that would inhibit them from competing with each other and with other hospitals; 4. Respondent ENH may not make any contract for E&G or Highland Park contingent on entering into a contract for the other, and may not make the availability of any price or term for a contract for E&G contingent on entering into a contract for Highland Park, or vice-versa; and 5. Respondent ENH shall promptly offer all payors with which it currently has contracts the option of reopening and renegotiating their contracts under the terms of this order.; IT IS FURTHER ORDERED THAT Respondent ENH’s proposal shall also describe, where appropriate, mechanisms for the Commission to monitor the establishment of the organizational structure needed to implement the terms of the order, as well as Respondent ENH’s compliance with the order throughout its term; IT IS FURTHER ORDERED THAT Respondent ENH’s proposal shall recommend mechanisms for resolving disputes EVANSTON NORTHWESTERN HEALTHCARE CORPORATION 545 Final Order between payors and Respondent ENH with respect to Respondent ENH’s compliance with the terms of the order, including a discussion of the potential value of some form of dispute resolution mechanism;

IT IS FURTHER ORDERED THAT Complaint Counsel shall file with the Commission any objections to or comments on Respondent ENH’s proposal within thirty (30) calendar days after Respondent ENH files its proposal; and IT IS FURTHER ORDERED THAT Respondents shall file any response to Complaint Counsel’s filing within ten (10) calendar days after Complaint Counsel file their objections or comments. By the Commission.

VOLUME 144 Complaint