In re: Northshore University HealthSystem Antitrust Litigation

• Court: District Court, N.D. Illinois
• Decided: February 20, 2023
• Docket: 1:07-cv-04446
• Status: Unknown

Opinion

UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF ILLINOIS EASTERN DIVISION

IN RE NORTHSHORE UNIVERSITY ) HEALTHSYSTEM ANTITRUST ) No. 07 CV 04446 LITIGATION ) ) ) Judge Edmond E. Chang ) )

MEMORANDUM OPINION & ORDER

In this antitrust class action,1 the parties are now on their fifth set of certifica- tion-related briefings. Most recently, Defendant NorthShore University HealthSys- tem filed a motion for decertification, this time challenging, among other things, the adequacy of the class representatives. R. 896, Def.’s Mot. Decert; see also Def.’s De- cert. Br.2 Shortly after, but before the class-certification issue was resolved, the par- ties filed cross-motions for summary judgment, along with motions challenging ex- pert evidence. R. 898, Def.’s Mot. SJ/Daubert; R. 911, Pls.’s Mot. SJ/Daubert. The Court provisionally granted NorthShore’s decertification motion on adequacy grounds, but allowed the class’s counsel to find a new class representative; mean- while, the Court put a hold on deciding the rest of the Rule 23(b)(3) issues. R. 989. Eventually, the class’s counsel proposed a new class representative, David Freedman.

1The Court has subject matter jurisdiction under 28 U.S.C. § 1331. 2Citations to the record are noted as “R.” followed by the docket number and the page or paragraph number. Sealed and unredacted records will be labeled as such, but the Opinion does not tie any of the figures to specific entities or to current information, so there are no redactions. The Court then held that Freedman is a proper and adequate representative. R. 1072. The Court now considers the other Rule 23(b)(3) decertification arguments presented by NorthShore, R. 896, as well as the parties’ renewed cross-motions for summary

judgment, R. 898; R. 911. I. Background This Opinion assumes familiarity with the underlying facts, which are set out in greater detail in two district-court opinions and a Seventh Circuit opinion: In re Evanston Nw. Corp. Antitrust Litig., 2013 WL 6490152 (N.D. Ill. Dec. 10, 2013), Mess- ner v. NorthShore Univ. HealthSys., 669 F.3d 802 (7th Cir. 2012), and In re Evanston Nw. Healthcare Corp. Antitrust Litig., 268 F.R.D. 56 (N.D. Ill. 2010).3

A. The Merger and FTC Proceedings Back in early 2000, Northshore—then doing business as Evanston Northwest- ern Healthcare Corporation—merged with Highland Park Hospital, located in High- land Park, Illinois. Before the merger, Northshore owned Evanston Hospital in Ev- anston, Illinois, as well as Glenbrook Hospital in nearby Glenview, Illinois. Since then, NorthShore has operated the three hospitals as a single, integrated entity. Just

before consummating the merger, NorthShore hired Bain & Company, Inc., an inde- pendent consulting firm to evaluate its commercial payor contracts.4 See R. 962,

3In deciding the decertification motion, the Court engages in fact-finding. But in de- ciding the motion for summary judgment, the Court must view the evidence in the light most favorable to the non-moving party. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587 (1986). Because NorthShore and Plaintiff have both moved for summary judg- ment, the Court will consider the evidence in the light most favorable to each party to see if the opposing party is entitled to summary judgment. 4The parties refer to Bain & Company, Inc.’s written work product as the “Bain docu- ments.” PSOF ¶¶ 40–42; R. 949, DSOAF ¶¶ 10–11; see also R. 925-9, Pls.’s Exh. 48.5 In a draft memorandum of their findings, Bain concluded that many of NorthShore’s contracted rates at the time should undergo a one-time corrective adjustment to make the rates

more profitable. Pls.’s Exh. 48. For example, Bain advised that NorthShore’s con- tracts with United (a health insurance company) were underpriced and would require an adjustment to make them economically viable for NorthShore. Id. Then, after the merger, Northshore raised some of its prices. PSOF ¶ 15; R. 939 (SEALED), Def.’s Resp. PSOF ¶ 15. Later, the Federal Trade Commission (widely known by its acronym, FTC) filed an administrative complaint against NorthShore, alleging that the merger substan-

tially lessened competition and enabled NorthShore to raise its prices of inpatient services to private payers6 above the price that the hospitals would have charged ab- sent the merger. During the FTC proceedings, NorthShore admitted that it raised its prices, but argued that it did so because the pre-merger prices were below market. PSOF ¶ 4; Def.’s Resp. PSOF ¶ 4. This was referred to as NorthShore’s “learning about demand” defense. An FTC Administrative Law Judge rejected this defense and

5Citations to the parties’ Local Rule 56.1 Statements of Fact are “DSOF” for NorthShore’s Statement of Facts [R. 906]; “PSOF” for Plaintiff’s Statement of Facts [R. 962]; “Pl.’s Resp. DSOF” for Plaintiff’s Response to NorthShore’s Statement of Facts [R. 960]; “Def. Resp. PSOF” for NorthShore's Response to Plaintiff’s Statement of Facts [R. 947]; “DSOAF” for NorthShore’s Statement of Additional Facts [R. 949]; “Pl.’s Resp. DSOAF” for Plaintiff’s Response to NorthShore’s Statement of Additional Facts [R. 976]; “PSOAF” for Plaintiff’s Statement of Additional Facts [R. 914]; and “Def.’s Resp. PSOAF” for NorthShore’s Response to Plaintiff’s Statement of Additional Facts [R. 948]. 6In the FTC proceeding, “private payers” was defined as MCOs that contracted with NorthShore during the relevant time period. found that NorthShore’s price increases were in violation of Section 7 of the Clayton Act, 15 U.S.C. § 18. On appeal, the FTC Commission issued a Final Order affirming the ALJ’s find-

ing of liability and ordering Northshore to allow payors and MCOs with pre-existing contracts to “re-open and renegotiate their contracts.” R. 906, DSOF ¶¶ 60–63. After that, at the end of July 2008, NorthShore sent letters to MCOs and informed them of the FTC’s Final Order. DSOF ¶¶ 66–69; R. 960, Pl.’s Resp. DSOF ¶¶ 66–69.7 Specifi- cally, the letters informed MCOs that they could respond to the letter and choose to renegotiate their contract within 60 days or, alternatively, choose not to respond at all. DSOF ¶¶ 67–69; Pls.’s Resp. DSOF ¶¶ 67–69. The letter stated that if an MCO

chose not to respond, then its existing contract with NorthShore would remain in effect until the contract’s expiration or termination. R. 901-10 (SEALED), Def.’s Exh. 59-67. As it turned out, no MCO chose to renegotiate its contracts; the MCOs either expressly declined to renegotiate or did not respond to the letter at all. DSOF ¶¶ 70– 71; Pl.’s Resp. DSOF ¶¶ 70–71. B. Procedural Background

After the FTC issued its Final Order, the then-Plaintiffs (at the time, there was more than one named Plaintiff) filed this case as a proposed class action on behalf of all end payors who purchased inpatient or outpatient healthcare services directly

7Defendant’s Exhibits 59–67 demonstrate that NorthShore addressed letters dated July 30, 2008, to a handful of MCOs. But there is no other record suggesting that each of those MCOs actually received them. from NorthShore.8 The Plaintiffs alleged that NorthShore illegally monopolized the healthcare services market and used its leverage to artificially inflate prices paid by the Plaintiffs and the proposed class in violation of Section 2 of the Sherman Act and

Section 7 of the Clayton Act. R. 240, Am. Consolidated Class Action Compl. ¶¶ 1–3. After discovery, an interlocutory appeal, more discovery, certification of the class, still more discovery, and a dispute over a change in class representative, the case has reached the stage of cross-motions for summary judgment and a motion by NorthShore to decertify the class. Both parties also filed Rule 702 motions against the other side’s experts. On decertification, NorthShore argues that the Plaintiff has not satisfied Rule

23(b)(3)’s predominance and superiority requirements. And as for summary judg- ment, NorthShore argues two things: (1) that the Plaintiff has not properly defined the relevant market, and (2) that the Plaintiff is barred from pursuing any damages arising after the 2008 FTC-imposed remedy. For the class, the Plaintiff argues that the class is entitled to summary judgment on liability, and that NorthShore’s affirm- ative defense fails. The Court addresses each motion in turn, as well as the relevant

Rule 702 motions. II. Legal Standard A. Decertification With regard to NorthShore’s motion for decertification, there is no difference between evaluating a class-certification motion and a motion asking to decertify an

8Since then, the class has been limited to payors who purchased inpatient healthcare services. See R. 989 at 7-10. already-certified class. Courts should typically decide the question of class certifica- tion before evaluating the merits of a given action. See Weismueller v. Kosobucki, 513 F.3d 784, 786-87 (7th Cir. 2008). Ultimately, “Plaintiffs bear the burden of producing

a record demonstrating the continued propriety of maintaining the class action.” Har- per v. Yale Int’l Ins. Agency, Inc., 2004 WL 1080193, at *2 (N.D. Ill. May 12, 2004); see also Binion v. Metro. Pier and Exposition Auth., 163 F.R.D. 517, 520 (N.D. Ill. 1995) (citing Gen. Tel. Co. of the Sw. v. Falcon, 457 U.S. 147, 160 (1982) (“[A] court remains free to modify or vacate a certification order if it should prove necessary”). A plaintiff obtains (or maintains) class certification by satisfying each requirement of Federal Rule of Civil Procedure 23(a): numerosity, commonality, typicality, and ade-

quacy of representation—as well as one subsection of Rule 23(b). See Harper v. Sheriff of Cook Cty., 581 F.3d 511, 513 (7th Cir. 2009); see also Oshana v. Coca–Cola Co., 472 F.3d 506, 513 (7th Cir. 2006). The plaintiff bears the burden of showing (based on a preponderance of the evidence) that each requirement is satisfied. See Retired Chi- cago Police Ass’n v. City of Chi., 7 F.3d 584, 596 (7th Cir. 1993). “Failure to meet any of the Rule's requirements precludes class certification.” Harper, 581 F.3d at 513;

Creative Montessori Learning Ctrs. v. Ashford Gear LLC, 662 F.3d 913, 916 (7th Cir. 2011) (quoting Wal–Mart Stores, Inc. v. Dukes, 564 U.S. 338, 351-51 (2011)) (“A class may be certified only if the trial court is satisfied, after a rigorous analysis, that the prerequisites of Rule 23(a) have been satisfied.”) (cleaned up).9

9This opinion uses (cleaned up) to indicate that internal quotation marks, alterations, and citations have been omitted from quotations. See Jack Metzler, Cleaning Up Quotations, 18 Journal of Appellate Practice and Process 143 (2017). The Court “must make whatever factual and legal inquiries are necessary to ensure that requirements for class certification are satisfied before deciding whether a class should be certified, even if those considerations overlap the merits of the case.”

Am. Honda Motor Co. v. Allen, 600 F.3d 813, 815 (7th Cir. 2010); see also Schleicher v. Wendt, 618 F.3d 679, 685 (7th Cir. 2010) (“a court may take a peek at the merits before certifying a class,” but that peek is “limited to those aspects of the merits that affect the decisions essential under Rule 23”). B. Summary Judgment Summary judgment must be granted “if the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a

matter of law.” Fed. R. Civ. P. 56(a). A genuine issue of material fact exists if “the evidence is such that a reasonable jury could return a verdict for the nonmoving party.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986). In evaluating sum- mary judgment motions, courts must “view the facts and draw reasonable inferences in the light most favorable to the” non-moving party. Scott v. Harris, 550 U.S. 372, 378 (2007) (cleaned up). The Court “may not weigh conflicting evidence or make cred-

ibility determinations,” Omnicare, Inc. v. UnitedHealth Grp., Inc., 629 F.3d 697, 704 (7th Cir. 2011) (cleaned up), and must consider only evidence that can “be presented in a form that would be admissible in evidence.” Fed. R. Civ. P. 56(c)(2). The party seeking summary judgment has the initial burden of showing that there is no genuine dispute and that they are entitled to judgment as a matter of law. Carmichael v. Vil- lage of Palatine, 605 F.3d 451, 460 (7th Cir. 2010); see also Celotex Corp. v. Catrett, 477 U.S. 317, 323 (1986); Wheeler v. Lawson, 539 F.3d 629, 634 (7th Cir. 2008). If this burden is met, the adverse party must then “set forth specific facts showing that there is a genuine issue for trial.” Anderson, 477 U.S. at 256.

III. Analysis A. Decertification Continued certification of the class under Rule 23(b)(3) is proper if the Plaintiff can show that two things remain intact: predominance and superiority. Fed. R. Civ. P. 23(b)(3); Messner v. NorthShore Univ. HealthSystem, 669 F.3d 802, 811 (7th Cir. 2012). NorthShore contends that Plaintiff no longer can show either element. To meet these class-certification requirements the first time around, the Plaintiff relied on the

opinions and testimony of Dr. David Dranove. Eventually, however, Dranove became unavailable to serve as an expert for the Plaintiff, R. 959, Pls.’s Decert. Resp. Br. at 9 n.3, so now the Plaintiff primarily relies on Dr. William Vogt. Dr. Vogt’s proposed method to show antitrust impact is largely the same as Dr. Dranove’s, but with one exception: Vogt was only able to use one of the MCO-level data sets, that of BCBS PPO, as opposed to all of them. R. 901-2 (SEALED), Def.’s

Exh. 12, Vogt Report ¶¶ 7, 10 n 1. But otherwise, the Plaintiff’s proof remains almost the same. Like Dranove’s analysis, the primary aspect of Vogt’s methodology is a dif- ference-in-differences analysis (which the parties have been calling a “DID” analysis). Id. ¶ 10. As a refresher from past opinions in this case, the DID method consists of identifying a control group of comparable hospitals and then running regressions to see if prices increased at a faster rate at the subject hospital (here, Northshore) than at a group of control hospitals. Id. ¶¶ 76–86; 109. Vogt’s control group consisted of 21 hospitals that had similar characteristics to Northshore but that were not involved in a merger during the period of measurement. Id. ¶¶ 89–96. Vogt conducted his DID

analyses by employing a statistical technique called linear-regression modelling. Id. ¶ 109. This linear-regression analysis took into account the complexities of the mar- ket for hospital services.10 Id. ¶ 109. Essentially, Vogt’s DID analysis compares (on the one hand) the change in Northshore’s prices from before to after the merger against (on the other) the comparable change in prices for the control group. Id. ¶ 55. The effect of the merger on Northshore’s prices is then estimated by the difference in these two changes (Northshore’s price increase minus the control-group price in-

crease). Id. Dr. Vogt used two data sets to conduct his analysis: one using average price increases based on the Medicare Cost Report (MCR) data, and the other using MCO- level data (the BCBS PPO data). Def.s’ Exh. 12, Vogt Rep. ¶ 97. The MCR data set is reflected in a publicly available report that hospitals participating in the Medicare program are required to provide. Id. ¶ 98; R. 901-3 (SEALED), Def.’s Exh. 13, 9/14/16

Vogt Dep. Tr. at 162:6–12. The reports include inpatient service prices from hospitals in a given year, averaging over different payors. Def.’s Exh. 12, Vogt. Rep. ¶ 101; R. 901–4 (SEALED), Def.’s Exh. 17, 5/8/09 Dranove Dep. Tr. at 130:8–11, 131:3–6; R. 901-10 (SEALED), Def.’s Exh. 87, Willig Rep. ¶ 122. That means that unlike the MCO-level data, the MCR data does not “delineate specific private-payors, it

10See Messner, 669 F.3d at 816-17 for a discussion on the complexities of the market for hospital services. aggregates them all into a total number.” Def.’s Exh. 13, 9/14/16 Vogt. Dep. Tr. at 88:17–20. In addition, MCR data includes information from commercial payors as well as government payors, like Medicare and Medicaid. Exh. 12, Vogt Rep. ¶ 101.

Because government payors are not part of the class, Vogt subtracted the Medicare (but not the Medicaid)11 payments for payors who received inpatient services.12 Def.’s Exh. 12, Vogt Rep. ¶ 101; Exh. 13, 9/14/16 Vogt Dep. Tr. at 79:23–80:7. Vogt then plugged the MCR inpatient data and the BCBS PPO data into DID regressions. Def.’s Exh. 12, Vogt Rep. ¶ 101. 1. Admissibility of MCR and BCBS PPO Data As a preliminary matter, NorthShore argues that the Plaintiff cannot show

predominance because Dr. Vogt’s methodology is unreliable and thus must be ex- cluded under Daubert and Evidence Rule 702. Def.’s Decert. Br. at 17-19. When an expert’s report or testimony is “critical to class certification,” the district court “must make a conclusive ruling on any challenge to that expert’s qualifications or submis- sions before it may rule on a motion for class certification.” See Messner, 669 F.3d at 812; see also Wal–Mart Stores, Inc. v. Dukes, 564 U.S. 338, 354–55 (2011) (expressing

doubts about the district court’s conclusion that “Daubert did not apply to expert tes- timony at the certification stage of class-action proceedings”). The term “critical” is

11Dr. Vogt does not appear to have explained why he did not exclude the Medicaid payments from the data set. 12It is worth noting that the class originally included commercial payors who received both inpatient and outpatient services. R. 240 at 1. But the Court later limited the class to only those who received inpatient services. R. 989. So there is no need to address Dr. Vogt’s DID analysis using outpatient data. The arguments on outpatient services in these motions are now moot, given the limitation on the class. used “broadly to describe expert testimony important to an issue decisive for the mo- tion for class certification.” Messner, 669 F.3d at 812. Vogt’s opinions are indeed “crit- ical” to the issue of decertification here—the opinions are an important part of the

Plaintiff’s argument against decertification. So NorthShore’s Rule 702 and Daubert challenges against Vogt’s submissions must be addressed before tackling the decerti- fication issues. The admissibility of expert testimony is governed by Federal Rule of Evidence 702 and the Supreme Court’s decision in Daubert. C.W. ex rel. Wood v. Textron, Inc., 807 F.3d 827, 834 (7th Cir. 2015). “In performing its gatekeeper role under Rule 702 and Daubert, the district court must engage in a three-step analysis before admitting

expert testimony.” Gopalratnam v. Hewlett–Packard Co., 877 F.3d 771, 779 (7th Cir. 2017) (cleaned up). “In other words, the district court must evaluate: (1) the proffered expert’s qualifications; (2) the reliability of the expert’s methodology; and (3) the rel- evance of the expert’s testimony.” Id. (emphases in original). In this case, Dr. Vogt’s qualifications are not the problem. He earned a Ph.D. in Economics from Stanford University and served as an Associate Professor of Eco-

nomics at the Terry College of Business at the University of Georgia. Def.’s Exh. 12, Vogt Rep. ¶ 1. Vogt focuses his research on the industrial organization of health care markets, with a particular focus on the hospital industry. Id. at ¶2. NorthShore does not cast doubt on his qualifications. Instead, NorthShore challenges the reliability and relevance of Dr. Vogt’s opin- ions. Def.’s SJ/Daubert Br. at 37-41. Under Rule 702, the three requirements for reliability are: “(1) the testimony is based upon sufficient facts or data, (2) the testi- mony is the product of reliable principles and methods, and (3) the witness has ap- plied the principles and methods reliably to the facts of the case.” Mihailovich v.

Laatsch, 359 F.3d 892, 918 (7th Cir. 2004) (quoting Fed. R. Evid. 702). To determine whether these requirements have been met, the district court should consider, among other things: “(1) whether the proffered theory can be and has been tested; (2) whether the theory has been subjected to peer review; (3) whether the theory has been evaluated in light of potential rates of error; and (4) whether the theory has been accepted in the relevant scientific community.” Gopalratnam, 877 F.3d at 779. Whether to allow expert testimony rests within the sound discretion of the district

court. Id. at 780. NorthShore does not directly challenge Dr. Vogt’s methodology in the sense that NorthShore does not attack the soundness—generally speaking—of DID anal- yses. According to Vogt, and corroborated by the record, DID analysis is the standard and preferred method in analyzing the effects of hospital mergers on prices. Def.’s Exh. 12, Vogt Rep. ¶ 55. Indeed, NorthShore points out that the Seventh Circuit, in

the prior appeal in this case, expected that the Plaintiff would employ (on remand) Dr. Dranove’s methodology, Def.’s Decert. Br. at 14–15, which itself also was a DID analysis. So, at the very least, Northshore does not take direct issue with the notion that DID analyses can be used to assess the effects of a merger. Instead, NorthShore takes issue with Vogt’s data sources—MCR and BCBS PPO data—on which to apply that methodology. NorthShore argues that MCR data is both unreliable and irrelevant in analyz- ing this merger. Typically, “reliability is primarily a question of the validity of the methodology employed by an expert, not the quality of the data used in applying the

methodology or the conclusions produced.” Manpower, Inc. v. Ins. Co. of Pa., 732 F.3d 796, 806 (7th Cir. 2013) (emphasis added). In performing its gatekeeper role, “the district court usurps the role of the jury, and therefore abuses its discretion, if it un- duly scrutinizes the quality of the expert’s data and conclusions rather than the reli- ability of the methodology the expert employed.” Gopalratnam, 877 F.3d at 781. But that is not to say that an expert may rely on data that has no quantitative or quali- tative connection to the methodology. Id. Indeed, the text of Rule 702 itself requires

that expert testimony be based on “sufficient facts or data.” Id. (emphasis added). Here, there is not enough reason to doubt the reliability of the MCR data all the way to the point of excluding Dr. Vogt’s opinions. In the qualitative sense, “suffi- cient data” simply means that the expert employed the “kinds of facts or data on which experts in the filed would reasonably rely.” Gopalratnam, 877 F.3d at 781. Vogt credibly explains how MCR data has been commonly used by economists employing

DID analyses to study the effects of hospital mergers on prices. See Def.’s Exh. 12, Vogt Rep. ¶ 113 n.44; see also Pls.’s Resp. DSOF ¶ 113 (citing various studies on methods used to identify and estimate hospital merger effects). NorthShore attempts to undermine these studies by arguing that although MCR data may be appropriate to use when studying large samples of mergers, it is inappropriate for the study of a particular merger. Def.’s SJ/Daubert Br. at 39. In support of this contention, NorthShore points to Christopher Garmon’s working paper, The Accuracy of Hospital Merger Screening Methods, Fed. Trade Comm’n Bureau of Econ Working Paper No. 326 (2016) (hereinafter Garmon Working Paper).13 See Def.’s SJ/Daubert Br. at 39–

40. But Garmon merely suggested that MCR data14 “may not be appropriate for the study of a particular paper;” he did not outright conclude that MCR data could not be used at all. Garmon Working Paper at 17 (emphasis added). In fact, Garmon noted that in retrospective studies of mergers, like the study here, “post-merger price changes estimated using [MCR] data are consistent with price changes using detailed claims data.” Id. NorthShore also argues that no other expert analyzing this merger has used

MCR data as the core of their analysis. Def.’s SJ/Daubert Br. at 39. But the relevant inquiry when analyzing the adequacy of data is not whether other experts in this particular case have relied on the same data. Instead it is whether experts in the relevant field would reasonably rely on the data. See Gopalratnam, 877 F.3d at 781. As noted earlier, Dr. Vogt has persuasively explained (at least enough to get to fact- finder) that that much is true here.

NorthShore presents two other arguments to challenge the reliability of Dr. Vogt’s use of MCR data. First, NorthShore points to discrepancies between the MCR data estimates and NorthShore’s actual prices. Def. SJ/Daubert Br. at 40-41. But

13NorthShore refers to this document as Exhibit 88 but it does not seem to appear on the docket, so the Court refers to the online version. 14Garmon refers to MCR data as “HCRIS data,” which stands for Healthcare Cost Re- port Information System. The two terms are interchangeable and refer to data taken from the Centers for Medicare and Medicaid Services (which the health care industry calls CMS). See Def.’s Exh.12, Vogt Rep. ¶ 98-100. NorthShore does not explain how these differences affect Vogt’s pricing analysis, or why this makes the MCR data insufficient. Neither does NorthShore’s expert, Dr. Robert Willig. According to Willig, the average reimbursement prices used in Vogt’s

analysis are always higher than commercial reimbursement prices in NorthShore’s billing data. Def.’s Exh. 87, Willig Rep. ¶ 124; see id. Figure 6. But as the Plaintiff points out, because Vogt’s DID analysis compares the rates of change in prices be- tween the subject and the control, what is important for purposes of his analysis is the trend of price changes, not the absolute numerical prices themselves. Pls.’s SJ/Daubert Br. at 63; Def.’s Exh. 12, Vogt Rep. ¶ 78. Second, NorthShore argues that the MCR data is “tainted” by Medicaid pay-

ments. Def.’s SJ/Daubert Br. at 39. In response, the Plaintiff contends that the pres- ence of that data did not significantly change the inpatient results, and if anything actually made the results more conservative. Pls.’s Resp. DSOF ¶ 74; Def.’s Exh. 12, Vogt Rep. ¶ 126; R. 901-10 (SEALED), Exh. 92, 1/25/17 Vogt Reply Rep. ¶¶ 53–55. NorthShore does not really attempt to reply to this argument. Assuming the prices paid by Medicaid are not more than the private-payor amount, Dr. Vogt’s explanation

seems correct. But neither side has presented much evidence on this point, so the Court makes no definitive finding on it, leaving this for the trial factfinder if the par- ties raise it at trial. For now, the important point is that neither NorthShore in its briefing nor Dr. Willig in his report explain how the presence of the Medicaid data makes Vogt’s pricing analysis unreliable. Moving on from reliability (but actually not moving on much from it), NorthShore also challenges the relevance of Dr. Vogt’s opinions, again criticizing the MCR and the BCBS data sets. To be relevant for purposes of Rule 702, the proposed

expert testimony must assist the trier of fact in understanding the evidence or in determining a fact in issue. Owens v. Auxilium Pharm., Inc., 895 F.3d 971, 972 (7th Cir. 2018). To meet this standard, the expert opinion must “fit the issue to which the expert is testifying and be tied to the facts of the case.” Id. at 973. Here, as explained earlier, Vogt’s DID analysis does bear on the intensely contested issue of the antitrust impact of the merger, which in turn will help the trier of fact make a decision on NorthShore’s liability. To combat the relevancy of Vogt’s opinions based on MCR data,

NorthShore relies on the same argument underlying its decertification motion: that averages cannot demonstrate antitrust impact on individual class members. Def.’s SJ/Daubert Br. at 41. There are two problems with this argument. First, NorthShore fails to present any legal support that stands for the proposition that averages cannot demonstrate antitrust impact. Second, even if it turns out that the Plaintiff ultimately cannot prove antitrust impact with the MCR data, that does not mean that the MCR

data is not relevant—the test for relevancy is not whether the testimony ultimately is credited by the factfinder. With respect to the relevancy of the BCBS PPO data, NorthShore contends that because BCBS is not a member of the class, the data cannot be used as repre- sentative evidence to establish antitrust impact. Def.’s SJ/Daubert Br. at 35. It is true that BCPS PPO was compelled to arbitrate for its own claims. R. 742. But the self- insured and the patients who paid under the BCBS PPO contracts remain in the class. R. 960, Pl.’s Resp. DSOF ¶ 92. What’s more, even with BCBS PPO itself outside the class, the BCBS PPO data is indeed relevant to the DID analysis: because BCBS

was the biggest MCO and retained the greatest bargaining power in the area,15 it is reasonable to conclude that if the BCBS PPO data shows antirust impact on BCBS, then purchasers of inpatient hospital services plans with less bargaining power would also have been impacted. Dr. Vogt’s analysis based on BCBS PPO data thus is rele- vant to the factfinder’s determination of class-wide antitrust impact. In sum, NorthShore’s Rule 702 and Daubert motion against Dr. Vogt’s opin- ions, at least as to his opinions based on MCR and BCBS PPO data, is denied.16 The

Court will address the remainder of NorthShore’s challenges to Vogt’s proposed tes- timony later on in this Opinion. Having decided on the admissibility of Vogt’s opin- ions, it is time to turn to the decertification motion. 2. Predominance Under Civil Rule 23(b)(3), predominance is satisfied when “common questions represent a significant aspect of a case and can be resolved for all members of a class

in a single adjudication.” Kleen Prod. LLC v. Int’l Paper Co., 831 F.3d 919, 925 (7th Cir. 2016) (cleaned up); see also Wal–Mart Stores, Inc. v. Dukes, 564 U.S. at 350. Even when applied rigorously, Rule 23 of course at times will authorize certification in

15The parties agree with this characterization of BCBS’s size and bargaining power. See Def.’s SJ/Daubert Br. at 47; Pl.’s Decert. Resp. Br. at 18. 16NorthShore also challenges the admissibility of Dr. Russell Lamb’s opinions based on MCR and BCBS PPO data. Lamb is the Plaintiff’s damages expert, but at the summary judgment stage, the Plaintiff does not need to prove damages, so the admissibility of Lamb’s opinions does not need to be decided now. antitrust cases. See Messner, 669 F.3d at 814 (citing Amchem Products, Inc. v. Wind- sor, 521 U.S. 591, 625 (1997)). And here it does. The starting point of predominance analysis is the elements of the underlying

cause of action. See Kleen, 831 F.3d at 925; Messner, 669 F.3d at 814. In antitrust cases, a plaintiff must prove: “(1) that defendants violated federal antitrust law; and (2) that the antitrust violation caused them some injury.” Messner, 669 F.3d at 815.17 Only the second element, referred to as “antitrust impact,” is at issue here. To defeat decertification, the Plaintiff must show (not merely allege) that the factfinder will be able to rely on commonly applicable evidence to prove that NorthShore’s merger in- jured the members of the class. Id. at 816. But that does not mean that the Plaintiff

must actually prove—at the certification (or decertification) stage—the element of antitrust impact. Id. at 818. Instead, the Plaintiff need only “demonstrate that the element of antitrust impact is capable of proof at trial through evidence that is com- mon to the class rather than individual members.” Id. (emphases in original). The focus, then, is on the evidence necessary to establish antitrust impact, not on whether the Plaintiff has adequately proven it before the trial even starts.

NorthShore again jabs at Dr. Vogt’s use of MCR and BCBS data. At the heart of NorthShore’s opposition to the MCR data is that averages cannot show predomi- nance because averages mask class members who were not in fact impacted. And as for the BCBS PPO data, NorthShore repeats the argument against that data as the

17Section 4 of the Clayton Act also requires a plaintiff to show damages, but individual proof of this element is not necessarily an obstacle to a showing of predominance. See Mess- ner, 669 F.3d at 815 (citing Wal–Mart Stores, 564 U.S. at 362; see also Kleen, 831 F.3d at 925. defense did in the Daubert motion challenging Vogt’s opinions: BCBS PPO is not a member of the class, so the data is not representative of the class. Def.’s Decert. Br. at 16-17. Taking that last critique first, for the same reasons that the attack on BCBS

PPO data was unpersuasive in the attempt to knock out Vogt’s testimony, the argu- ment is unpersuasive now. The biggest problem for NorthShore’s challenge to Dr. Vogt’s methodology is that the methodology is, with one exception, essentially the same methodology Dr. Dranove proposed to use. And the Seventh Circuit has already held that that meth- odology allows the Plaintiff to satisfy the predominance requirement. See Messner, 669 F.3d at 818. According to NorthShore, Dranove’s methodology was a DID analysis

using only MCO-level data. Def.’s Decert. Br. at 2–3; Def.’s Decert. Br. at 10–11. But NorthShore’s description of Dranove’s methodology, however, is only half right. Both Dranove and (just as importantly) the Seventh Circuit explained that this was only one of Dranove’s methodologies. R. 901-12 (SEALED), Def.’s Exh. 100, Dranove Rep. ¶ 15; Messner, 669 F.3d at 819-20. As a reminder, Dr. Dranove proposed to use two alternative methodologies to

show antitrust impact: “one in which uniformity of merger-related price increases was presumed, and another in which uniformity was absent.” Messner, 669 F.3d at 819. Like Dranove, that is what Dr. Vogt proposed to do here. If the price increases were entirely or largely uniform, then Dranove proposed to use his first methodology to show the merger’s impact on individual class members by “plugging the average price increase imposed by any given contract into his DID analysis.” Messner, 669 F.3d at 819-820 (emphasis added). Dranove explained that he can use a number of different data sets for this first methodology, one of which was the very same data Vogt now proposes to use: MCR data. R. 901–4, Def.’s Exh. 15, Dranove Reply Rep.

¶ 82, Figure 5; Def.’s Exh. 100, Dranove Rep. ¶ 65; see also Messner, 669 F.3d at 819– 20. But if price increases were non-uniform, then Dranove proposed to use the second methodology: he would run DID analysis on the differences in price across time on a plan-by-plan basis for each MCO. Def.’s Exh. 100, Dranove Rep. ¶ 15. This analysis would require using the MCO data. As explained earlier, Vogt can only use one of the MCO-level data sets, that is, the BCBS PPO data. So, essentially, except for this dis- tinction, Vogt’s methodology is the same as Dr. Dranove’s. And the Seventh Circuit

already held that either methodology was sufficient to show predominance in this case, regardless of whether uniformity of prices was present. See Messner, 669 F.3d at 818–19. The parties go back and forth on whether NorthShore uniformly exercised its market power. In line with its argument that the Plaintiff must use the MCO-level data to show predominance, Northshore presents evidence to suggest that it did not

exercise its power uniformly. Def.’s Decert. Br. at 12. But the Plaintiff presents evi- dence of the opposite. Pls.’s Decert. Resp. Br. at 11 (citing Exh. 12, Vogt Rep. ¶ 48). The Court need not decide right now, however, which of the parties’ evidence of uni- formity or lack thereof to believe. Requiring a definitive finding on that point would mean applying the too-high burden made earlier in this case. See Messner, 669 F.3d at 817–18 (holding that the district court applied “too stringent a standard” in evaluating predominance when it made uniformity of price increases a condition for class certification). All that the Plaintiff has to show at this stage is that the class is capable of proving antitrust impact. The Plaintiff has done that based on Dr. Vogt’s

proffered testimony. Again, Dr. Vogt’s DID analysis compares the change in NorthShore’s average prices from before to after the merger to the change in average prices for a control group. Def.’s Exh. 12, Vogt Rep. ¶ 55. Vogt then estimates the effect of the merger on NorthShore’s prices by using the difference in those two changes (NorthShore’s price increase minus the control group price increase). Id. In other words, the Plaintiff has demonstrated that the class can use common evidence—the post-merger price in-

creases NorthShore negotiated with insurers—to show that all or most of the insurers and individuals who received coverage through those insurers suffered antitrust in- jury as a result of the merger. Ss the Seventh Circuit made clear in Messner, “that is all that is necessary to show predominance for purposes of Rule 23(b)(3).” 669 F.3d at 818. The remainder of NorthShore’s arguments to the contrary are unpersuasive.

NorthShore relies on a laundry list of non-binding cases to generally argue that “courts reject the use of average price differentials to show evidence of antirust im- pact that is common to the class.” Def.’s Decert. Br. at 8–9. But none of these cases stand for the general proposition that averages can never be used to show predomi- nance. Just because averages could not be used to show predominance in those cases, does not mean they cannot be used here. Indeed, “[t]here is no mathematical or mechanical test for evaluating predominance.” Messner, 669 F.3d at 814. Whether the Plaintiff should have been able to use the rest of the MCO data is missing the point: the Plaintiff has shown that the class is at the very least capable of proving

antitrust impact employing a DID analysis on MCR and BCBS data. 3. Superiority NorthShore’s only argument as to why the Plaintiff cannot establish superior- ity is that the class cannot establish predominance. Def.’s Decert. Br. at 18. But as discussed above, the Plaintiff has satisfied the predominance requirement. With NorthShore’s arguments on decertification rejected, the motion to decertify is denied. B. Summary Judgment

1. NorthShore’s Summary Judgment Motion Moving on to NorthShore’s summary judgment motions, NorthShore presents two main arguments. First, NorthShore argues that the Plaintiff has not established a proper relevant geographic market. In response, the Plaintiff contends that not only has the class successfully identified the market, no trier of fact could find for NorthShore on the issue, so it is the class that is entitled to summary judgment on

this issue. Second, NorthShore argues that the class is barred from pursuing damages after the issuance of the 2008 FTC Final Order. a. Relevant Geographic Market

For both theories of liability advanced by the Plaintiff—Section 2 of the Sher- man Act and Section 7 of the Clayton Act—monopoly power in a defined market must be proven in order for the Plaintiff to ultimately prevail. Specifically, Section 2 of the Sherman Act makes it unlawful for anyone to “monopolize, or attempt to monopolize, or combine or conspire with any other person or persons, to monopolize any part of the trade or commerce ....” 15 U.S.C. § 2. This section of the Sherman Act prohibits

“the employment of unjustifiable means to gain that power” and requires “two ele- ments: (1) the possession of monopoly power in the relevant market and (2) the willful acquisition or maintenance of that power ....” United States v. Grinnell Corp., 384 U.S. 563, 570–71 (1966) (emphasis added). “For purposes of § 2 of the Sherman Act, a market is defined by the reasonable interchangeability of the products and the cross-elasticity of demand for those products.” In re Dairy Farmers of Am., Inc. Cheese Antitrust Litig., 767 F.Supp.2d 880, 901 (N.D.Ill.2011) (citing United States v. E.I. du

Pont de Nemours & Co., 351 U.S. 377, 394—95 (1956)). Essentially the same elements are required for a claim under Section 7 of the Clayton Act, which generally speaking bans any corporate acquisition that substan- tially reduces competition. More specifically, that section makes it unlawful to “ac- quire ... the assets of another person ... where in any line of commerce ... in any section of the country, the effect of such acquisition may be substantially to lessen competi-

tion, or tend to create a monopoly.” 15 U.S.C. § 18. The statutory text thus demands that the class show not only the relevant geographic market (the “section of the coun- try”) but also the relevant product market (the “line of commerce”). FTC v. Advocate Health Care Network, 841 F.3d 460, 467 (7th Cir. 2016) (citing United States v. E.I. du Pont de Nemours & Co., 353 U.S. 586, 593 (1957) (“Determination of the relevant market is a necessary predicate to a finding of a violation of the Clayton Act.”) and citing Brown Shoe Co. v. United States, 370 U.S. 294, 324 (1962) (“The ‘area of effec- tive competition’ must be determined by reference to a product market (the ‘line of commerce’) and a geographic market (the ‘section of the country’).”)).

That means the class must show both monopoly power and the relevant market to prevail on its claims. See Republic Tobacco Co. v. N. Atl. Trading Co., 381 F.3d 717, 736 (7th Cir. 2004). As a preliminary matter, the class argues that it does not need to show a relevant market for two reasons: (1) the merger already happened and (2) the Plaintiff can offer direct evidence to show market power. Pls.’s SJ/Daubert Br. at 17– 18. But neither of these reasons lets the Plaintiff off the hook of showing the relevant market. The Court rejected the first argument in a prior order. See R. 989 at 9. On

the second, a plaintiff can prove market power in two separate ways. The more con- ventional way “is by proving relevant product and geographic markets and by show- ing that the defendant's share exceeds whatever threshold is important for the prac- tice in the case.” Toys “R” Us, Inc. v. FTC, 221 F.3d 928, 937 (7th Cir.2000). Under this method, the plaintiff “must precisely establish” the relevant market. Id. (empha- sis added). Alternatively, a plaintiff may establish market power through “direct ev-

idence of anticompetitive effects.” Id. at 937. This method allows for a more relaxed relevant market definition, but it does not allow an antitrust plaintiff to dispense entirely with market definition. A plaintiff still must “show the rough contours of a relevant market, and show that the defendant commands a substantial share of the market.” Republic Tobacco Co., 381 F.3d at 737. And that means showing both a rel- evant products (or services) market and a geographic market. See id. After adamantly arguing otherwise, the Plaintiff eventually concedes that spe- cific point at the very end of the class’s reply brief, where for the first time the class says that the rough contour of the geographic market is the north-shore suburbs of

Chicago. See R. 977, Pls.’s SJ/Daubert Reply Br. at 24-25. But that is too little too late. The Plaintiff cannot for the first time in the reply brief argue that they are tak- ing the direct-evidence route to show the relevant market. And either way, the class’s one-sentence attempt to do so falls short. The Plaintiff neither engages in a meaning- ful discussion of the relevant case law nor applies the law to the facts of this case. So on summary judgment, the Plaintiff must go with the alternative route (at trial, the Plaintiff will not necessarily be limited in that way).

At this stage, though, the Plaintiffs do not have to actually establish the rele- vant market or exclude all other possible geographic markets. Generally speaking, defining the relevant market is a question for jury. See Republic Tobacco Co., 381 F.3d at 725 (“[T]he definition of a relevant market is a question of fact.” Fishman v. Estate of Wirtz, 807 F.2d 520, 531 (7th Cir. 1986) (“Claims of monopolization under section 2 of the Sherman Act … require the trier of fact to delineate the relevant

market.”) (cleaned up). So to defeat NorthShore’s summary judgment motion, the class only needs to show that a genuine dispute over the relevant market exists. And when it comes to defeating the Plaintiff’s cross-motion, NorthShore only need to do the same. The Plaintiff originally alleged that the relevant product market in this litiga- tion is “Healthcare Services,” which they define as “general inpatient and hospital- based outpatient services provided by” NorthShore. R. 224, Pls.’ Consolidated Class Action Compl. ¶¶ 8, 25. But later this Court removed outpatient services from the class definition because the Plaintiff presented no evidence at all to support defining

a market for such services. R. 989, Order at 7–10. That also means that the product market is limited to inpatient services. So only the geographic market is at issue here. There are several routes an antitrust plaintiff can take to show the relevant geographic market. But a common method, and the one the Plaintiff’s expert used, is the hypothetical monopolist test. It is used to determine whether a hypothetical mo- nopolist could profitably impose a “small but significant non-transitory increase in price” (in antitrust parlance, a “SSNIP”) in the proposed market. See Advocate, 841

F.3d at 468 (citing Gregory J. Werden, The 1982 Merger Guidelines and the Ascent of the Hypothetical Monopolist Paradigm, 71 Antitrust L.J. 253, 253 (2003)); see also Def.’s Exh. 12, Vogt Rep. ¶ 164. Or, in other words, whether the hypothetical monop- olist “could profitably raise prices above competitive levels.” Advocate, 841 F.3d at 646. A five-percent price increase is often large enough to be considered a SSNIP. Def.’s Exh. 12, Vogt Rep. ¶ 166 (citing U.S. Dep’t of Justice & FTC Horizontal Merger

Guidelines § 4.1.2 (2010)). If that hypothetical monopolist could profitably impose a SSNIP, then the region is a relevant geographic market. Id. But if enough consumers would respond to a SSNIP by purchasing the product from outside the proposed geo- graphic market—thus rendering the SSNIP unprofitable—then the proposed market definition is too narrow and the test should be repeated until the SSNIP is profitable. Id. Because NorthShore actually came to own these three hospitals here, Dr. Vogt worked backwards and started with Evanston Hospital, Glenbrook Hospital, and Highland Park Hospital as the geographic market. He then used NorthShore’s price

changes as a guide to what a hypothetical monopolist would have done. Def.’s Exh. 12, Vogt Report ¶ 178–80. To do this he looked at the results of his DID analysis. According to Dr. Vogt’s DID analysis of the MCR data, NorthShore raised its prices after the merger in a range of 16.2% to as much as 54.7%. Id. ¶ 117, Table 7. These price increases remained above those of the control group throughout the measure- ment period. Id. ¶ 118, Figure 3. And the DID analysis of the BCBS PPO data simi- larly showed annual increases in prices for inpatient services with the exception of

two years: 2001 and 2003.18 Id. ¶ 132, Table 11. Overall, the inpatient price increases rose above the prices at the control hospitals.19 Id. Vogt concluded that an actual hypothetical monopolist who came to control the three hospitals would have also im- posed at least a SSNIP. Id. ¶ 182. To counter Dr. Vogt’s conclusion, NorthShore relies on Dr. Margaret Guerin- Calvert’s “diversion ratio analysis.” Diversion ratio analysis is another tool used to

measure relevant geographic markets. This form of analysis relies on patient-dis- charge data, R. 943-1 (SEALED), Exh. 112, 11/22/16 Guerin-Calvert Dep. Tr. at 126:22-127:2; 130:2-17, and identifies where patients would go for care if their first-

18The price increases in these years were -1.5% and 0.1%, respectively. Def.’s Exh.12, Vogt. Rep. Table 11. NorthShore agrees that neither of these results are statistically signifi- cant. Def.’s Resp. PSOF ¶17. 19 Dr. Vogt noted that NorthShore’s inpatient services price rose in line with the con- trol hospitals until 2003. Def.’s Exh. 12, Vogt. Rep. ¶ 131. choice hospital was unavailable, R. 901-4 (SEALED), Def.’s Exh. 16, Guerin-Calvert Rep. ¶ 109. In other words, it measures the closeness of competition between hospi- tals. Id.; see also Advocate, 841 F.3d at 475. A diversion ratio is the proportion of

profits or patients that would be shifted between and among merging parties if one supplier raised its prices. Def.’s Exh. 16, Guerin-Calvert Rep. ¶ 109. For example, a high estimated diversion ratio between Hospital A and Hospital B means that MCOs could expect more patients to choose B if A were not available as an in-network alter- native. Id. On the other hand, a low diversion ratio between A and B suggests the opposite, and that A and B are not close substitutes. Id. According to Dr. Guerin-Calvert, Dr. Vogt’s geographic triangle is too narrow

and the relevant market is instead made up of at least 12 hospitals. Def.’s Exh. 16, Guerin-Calvert Rep. ¶ 112. The results of Guerin-Calvert’s diversion ratio analysis showed that 86% of Evanston’s patients, 73.6% of Highland Park’s patients, 92.2% of Lutheran General’s patients, and 88.9% of Lake Forest’s patients drove past at least one other hospital to receive care. Id. ¶ 114, Table 9. According to Guerin-Calvert, it is economically reasonable that any hospital with a diversion ratio greater than or

equal to the diversion from one merging party to another can be considered a hori- zontal competitor in the relevant geographic market. Id. ¶ 126. Based on this, Ms. Guerin-Calvert found that at least 12 hospitals should be considered competitors and should be included in the geographic market. Id. ¶ 127-28, Table 11. Dr. Guerin-Calvert cross-checked her findings with the business documents of other hospitals, the merging parties, and payors, and determined that these documents supported adding other competitors to the market definition. Def.’s Exh. 16, Guerin-Calvert Rep. ¶¶ 19, 110, 167-180. For example, NorthShore records from 1999 through 2016 referred to 10 hospitals as a “pricing peer group.” Id. ¶ 168; DSOF

¶ 41. Similarly, the records of several other nearby hospitals identified NorthShore as a competitor. Def.’s Exh. 16, Guerin-Calvert Rep. ¶ 173, DSOF ¶¶ 44, 46, 47, 48. Several MCO records also say that there were alternative providers within the same geographic area as Evanston, Highland Park, and Glenbrook. Def.’s Exh. 16, Guerin- Calvert Rep. ¶ 177; DSOF ¶ 43. But neither side’s evidence is dispositive on summary judgment. When viewing these facts in the light most favorable to the class, a reasonable jury could find that

Dr. Vogt’s analysis supports defining the relevant geographic market as the hospital triangle. NorthShore argues otherwise, but its arguments do not conclusively resolve the dispute given the summary judgment standard. First, NorthShore argues that the geographic market in FTC v. Advocate Health Care Network, an unrelated merger between NorthShore and Advocate Health Care, should be applied here. Def.’s SJ/Daubert Br. at 17–18. But that merger had nothing to do with the merger at issue

here, and NorthShore does not explain or offer legal support to show how the geo- graphic market in an unrelated merger—with a different factual record—dictates the geographic market here. Second, NorthShore argues that Dr. Vogt did not appropriately perform the hypothetical monopolist test because the results of his DID analysis did not show that NorthShore raised it prices to “supracompetitive levels.” Def.’s SJ/Daubert Br. at 23– 24.20 Remember that, under the hypothetical monopolist test, a region is a relevant geographic market if the hypothetical monopolist “could profitably raise its prices above competitive levels.” Advocate, 841 F.3d at 468 (emphasis added). According to

NorthShore, Vogt only demonstrated that NorthShore’s prices increased post-merger, but not that the prices increased above competitive levels as required by the test. But a reasonable jury could find that Vogt properly performed the test. Vogt defined the geographic market based on the results of his DID analysis, which showed that NorthShore actually imposed a SSNIP, Def.’s Exh. 12, Vogt Rep. ¶ 182, and that NorthShore’s price increases were above the control groups’ price increases. See Def.’s Vogt. Rep ¶ 118, Figure 3; ¶ 132, Table 11. So a reasonable jury could very well find

that NorthShore raised its prices above competitive levels. Third, NorthShore challenges the admissibility of Dr. Vogt’s opinions that rely on the conclusions of Dr. Deborah Haas-Wilson, the FTC’s expert during the agency proceedings.21 But under Evidence Rule 703, an expert may rely on information pro- vided by non-testifying experts if experts in the filed would reasonably rely on the

20NorthShore relies on this same “supracompetitive” argument to challenge the reliability of Dr. Vogt’s hypothetical monopolist test under Daubert. Def.’s SJ/Daubert Br. at 43–44. But as explained earlier, reliability is generally a question of the validity of the methodology em- ployed by the expert, not the results it produces. Gopalratnam, 877 F.3d at 780–81. The cor- rectness of the results of Vogt’s DID analysis are instead, “factual matters to be determined by the trier of fact, or where appropriate, on summary judgment.” Id. at 781. And as explained in this Opinion, a genuine dispute remains over this issue. So whether Vogt’s DID analysis shows that NorthShore price increases were above competitive levels is a question for the trial. 21The parties agree that Dr. Vogt relied on the Dr. Haas-Wilson’s findings as a basis for Vogt’s finding that NorthShore increased its prices post-merger relative to the control group. See Pls’s Resp. DSOF ¶ 94; see also Def.’s Exh. 12, Vogt Rep. ¶¶ 145–46; Def.’s Exh. 92, Vogt Reply Rep. ¶ 9. type of information at issue. See Fed. R. Evid. 703; Dura Auto. Sys. Of Ind., Inc. v. CTS Corp., 285 F.3d 609, 613 (7th Cir. 2002)). Indeed it “is common in technical fields for an expert to base an opinion on what a different expert believes on the basis of

expert knowledge not possessed by the first expert.” Dura, 285 F.3d at 613. The reli- ance-based testimony need only be excluded when an expert is “just parroting the opinion” of another expert. Id. Otherwise, an expert may rely on information provided by non-testifying experts, so long as experts in the particular field would reasonably rely on the information and the testifying expert does not serve as a mere mouthpiece for the absent expert. Id. NorthShore argues that Dr. Vogt’s opinions should be excluded because he

merely repeats Dr. Haas-Wilson’s opinions without conducting an independent re- view of her work. Def.’s SJ/Daubert Br. at 51. The Plaintiff of course contends that Vogt properly relied on Haas-Wilson’s opinions. Pls.’s SJ/Daubert Br. at 71. When viewed through the lens of summary judgment, a trial factfinder can reasonably find that experts in the particular field of analyzing hospital-merger price increases do rely on other experts. Indeed, it arguably would be cross-examination fodder for a

plaintiff’s expert to not consider the FTC’s expert analysis when a government-re- tained expert’s analysis is available. In any event, Vogt also conducted an independ- ent DID analysis to arrive at his own conclusions on NorthShore’s price increases. He then analyzed Haas-Wilson’s DID analysis to corroborate his own conclusions. Def.’s Exh. 13, 9/14/16 Vogt Dep. Tr. at 153:2-16, 154:19-156:3; Def.’s Exh. 12, Vogt Rep. ¶¶ 142-46. Vogt’s conclusions thus do not rise or fall on Dr. Haas-Wilson’s findings alone. Lastly, NorthShore points to its own geographic market definition as evidence that Dr. Vogt’s is wrong. But even if Dr. Guerin-Calvert’s analysis and the additional records that NorthShore relies on show that Vogt’s proposed geographic market ex-

cludes significant competitors, it does not necessarily follow that Vogt’s geographic market is defined too narrowly. Indeed, a relevant geographic market need not in- clude every hospital that competes for business in the vicinity; it need only include those competitors that would “substantially constrain” the merged firm’s “price-in- creasing ability.” Advocate, 841 F.3d at 469. NorthShore does not otherwise explain how or why the results of Vogt’s hypothetical monopolist test are inaccurate. Either way, when evaluated under the summary judgment standard, Dr.

Guerin-Calvert’s diversion ratio analysis does not fatally undermine the Plaintiff’s geographic market definition. The problem with diversion ratios is that they focus on the patients who leave a proposed market. Advocate, 841 F.3d at 475; Def.’s Exh. 12, Vogt Reply Rep. ¶ 32. But in the healthcare context, arguably it is insurers—not pa- tients—who are the most relevant buyers. Advocate, 841 F.3d at 475. As the buyers, insurers consider whether employers would offer their plans, and whether employees

would sign up for them. Id. “As a result, measures of patient substitution like diver- sion ratios do not translate neatly into options for insurers.” Id. So Guerin-Calvert’s diversion ratio analysis is not a slam dunk. Neither are the business records to which she points in support of the diversion ratio analysis. Def.’s SJ/Daubert Br. at 22-23. At best, these records show that NorthShore had other competitors in the market, but—again—the relevant geographic market focuses only on those competitors that would “substantially con- strain” the hospital’s price-increasing ability. Advocate, 841 F.3d at 469. Neither NorthShore nor Dr. Guerin-Calvert explain how these records make such a showing,

especially when viewed in the light most favorable to the class. In the same vein, however, neither is the class’s evidence dispositive. When the summary-judgment lens flip and the evidence is viewed in the light most favorable to NorthShore, a reasonable jury can alternatively decide that Dr. Vogt’s geographic market is in fact too narrow. Dr. Guerin-Calvert’s diversion ratio analysis shows that a majority of patients traveled past at least one hospital for inpatient services, Def.’s Exh. 16, Guerin-Calvert Rep. ¶ 114, Table 9, and that nearby hospitals were much

closer substitutes for Evanston and Highland Park than either hospital was for each other, id. ¶ 132. In other words, the analysis shows that patients would potentially buy a health plan outside of Vogt’s proposed geographic triangle, which means that a reasonable jury could find that the relevant geographic market is indeed broader than just the three hospitals that the Plaintiff proposes. The Plaintiff criticizes Dr. Guerin-Calvert’s analysis, but none are so persua-

sive that the Plaintiff wins summary judgment outright. First, the class argues that the Seventh Circuit in Advocate outright rejected diversion ratio analysis as unrelia- ble. Pls.’s SJ/Daubert BR. at 21. But the Seventh Circuit did no such thing; it merely criticized how the defendants in that particular case had interpreted the analysis. Advocate, 841 F.3d at 474-76. Because patients were not the relevant buyers in the healthcare market, the Seventh Circuit reasoned that diversion ratios were not as “compelling” as the district court considered them to be. Id. at 475. But that does not mean diversion ratios are, as a matter of law, totally irrelevant. For example, when MCOs are conducting negotiations and deciding which hospitals to include in their

network, the evidence reasonably supports a finding that MCOs are concerned about how employers will react to changes in their network. R. 901-4 (SEALED), Def.’s Exh. 14, 2/16/17 Vogt. Dep. Tr. at 48:6–50:12; R. 943-1 (SEALED), Def.’s Exh. 112, 11/22/16 Guerin-Calvert Dep. Tr. at 127:14–128:1. In turn, employers are concerned about how their employees—the patients—will react. Id. So even though MCOs may not directly rely on diversion ratios, the ratios do present material information that MCOs, either directly or indirectly, have in mind when they negotiate. Def.’s Exh. 14, Vogt. Dep.

Tr. at 85:12-86-18. The information that diversion ratios present thus can be useful when defining the geographic market. Def.’s Exh. 14, 2/16/17 Vogt Dep. Tr. at 86:13– 18. Second, the Plaintiff confuses Dr. Guerin-Calvert’s diversion ratio analysis with the Elzinga-Hogarty test, which has been criticized for often overestimating the size of hospital markets. Pls.’s SJ/Daubert Br. at 21. But the only similarity between

the two tests is that they use patient-discharge data. See Def.’s Exh. 112, Guerin- Calvert Dep. Tr. at 126: 22–127:2. Otherwise, the tests are different and ask different questions. Id. at 124:1–130:1. Guerin-Calvert explicitly stated (numerous times) that she did not perform the Elzinga-Hogarty test, and in fact rejected it as an accepted methodology. Id. at 126: 15–21, 127:14–128:1, 133:17–22 (“I have several times ex- pressly said, ‘I am not doing Elzinga-Hogarty.’ I do not accept it as a methodology.”).22 In sum, there is a genuine dispute over the relevant geographic market, so

neither party is entitled to summary judgment on the issue. b. 2008 FTC Remedy

Lastly, NorthShore argues that because the class failed to respond to Northshore’s letter (dated July 30, 2008) offering to reopen negotiations, the class consequently waived any claim to damages and failed to mitigate damages, so the class is estopped from pursuing damages now.23 Def.’s SJ/Daubert Br. at 26–34. The problem for NorthShore, though, is that even assuming NorthShore’s letter was re- ceived by the intended recipients, NorthShore only sent the letter to MCOs. DSOF ¶ 66. But not all class members are MCOs. So that means that some class members, like Plaintiff Freedman, were never even aware of the remedy proposed by the FTC’s Final Order. So NorthShore’s summary judgment as to waiver and failure to mitigate fails most certainly must fail as to the non-MCOs. See Delta Consulting Grp., Inc. v. R. Randle Const., Inc., 554 F.3d 1133, 1140 (7th Cir. 2009) (“In Illinois, waiver is the

22For these same reasons, the Plaintiff argues that Dr. Guerin-Calvert’s diversion ra- tio analysis is inadmissible under Daubert. But as explained in denying summary judgment for the class on this point, the analysis can form the basis for a reasonable factfinder against the Plaintiff, and the reliability and relevancy of the analysis readily satisfied Rule 702. 23In support of its estoppel defense, NorthShore initially argued that each MCO now pursuing damages misled, either in writing or through inaction, NorthShore into thinking that the particular MCO wished to continue operating under its existing contracts, and that this somehow was a concession that the prices were competitive. Def. SJ/Daubert Br. at 33. The class responded by arguing that the MCOs’ actions do not constitute a misleading repre- sentation, and NorthShore’s reply brief was silent on the issue. The failure to reply is telling, because there is no authority for the proposition that continuing under the contracts equates to a concession that the prices were competitive. voluntary and intentional relinquishment of a known right.”) (emphasis added); Straits Fin. LLC v. Ten Sleep Cattle Co., 900 F.3d 359, 375 (7th Cir. 2018) (“[U]nder Illinois law, the duty to mitigate damages does not arise until the party upon whom

the duty is impressed is aware of facts which make the duty to mitigate necessary.”) (emphasis added) (cleaned up). More importantly, even as to MCOs, NorthShore is not entitled to summary judgment. The Plaintiff correctly points out that nothing in the letter that NorthShore sent out suggests that MCOs would be waiving their right to damages in subsequent private-enforcement actions if the MCOs chose not to reopen and renego- tiate their contracts. All the Final Order did was allow MCOs to reopen or renegotiate

their contracts going forward. DSOF ¶¶ 67–69. NorthShore fails to explain how or why opting not to renegotiate their contracts, equates to a knowing and voluntary waiver of their right to the damages they now seek. NorthShore also argues that the class failed to mitigate damages after July 30, 2008, by choosing not to renegotiate the contracts. It is true that the mitigation doc- trine bars recovery to plaintiffs who idly sit by and knowingly allow their damages to

accumulate while doing nothing to avoid them. But at the same time, plaintiffs are not required to engage in futile efforts. See Karahodzic v. JBS Carriers, Inc., 881 F.3d 1009, 1017 (7th Cir. 2018) (citing Amalgamated Bank of Chicago v. Kalmus & Assocs., 741 N.E.2d 1078, 1086 (2000) (the duty to mitigate imposes a duty on the injured party to exercise reasonable diligence and ordinary care in attempting to minimize his damages after injury has been inflicted)). As argued by the Plaintiff, the FTC “remedy” might not have been a remedy with teeth—there was no guarantee that reopening or renegotiating contracts would have restored competition or lowered prices. Pls.’s SJ/Daubert Br. at 47–48. Based on the summary judgment record, and

viewed in the light most favorable to the class, a reasonable jury could find in favor of the class on this point. Even though the FTC imposed its remedy as a way to “re- store competition,” Dr. Vogt reasonably opined that the FTC remedy was “highly un- likely to restore damaged competition and that this was and is a broadly shared view among academic economists.” Def.’s Exh. 92, Vogt Reply Rep. ¶ 106. The FTC also acknowledged that a divesture of the merger would have been the most appropriate remedy for restoring competition, but the order was the best that the FTC could do

eight years after the merger. In the Matter of Evanston Northwestern Healthcare Corp., 2007 WL 2286195, at *77–78 (FTC 2007). A genuine dispute of material fact exists on the mitigate defense, so NorthShore is not entitled to summary judgment on it. 2. Plaintiff’s Summary Judgment Motion

For its part, the class argues that it is entitled to summary judgment because the evidence presented shows there is no dispute over NorthShore’s liability, and be- cause NorthShore’s quality-of-care affirmative defense fails. The motion is denied as to liability (as forecast by the earlier discussion) and granted on the affirmative de- fense. a. Liability As explained earlier, to prevail on its claims, the class must prove both market power and the relevant market. According to the Plaintiff, there is no dispute of ma-

terial fact over either. But as already discussed, a genuine dispute over the relevant geographic market exists, so the Plaintiff is not entitled to summary judgment on liability. And either way, the Plaintiff’s evidence as to market power is also not dis- positive. To support its motion, the Plaintiff first points to Dr. Vogt’s expert report. As discussed earlier, Vogt found that NorthShore raised its prices post-merger faster than the group of control hospitals. He concluded that these price increases were the

exercise of anticompetitive market power obtained in the merger. Def.’s Exh. 12, Vogt Rep. ¶¶ 10, 156–58. Vogt then supported this conclusion using the results of other experts’ analyses, and documentary and testimonial evidence. Id. ¶¶ 152–55. But at the same time, NorthShore counters this evidence with the analysis of their own expert, Dr. Robert Willig, who concluded the opposite. See Def.’s SJ/Daubert Reply Br. at 17–20; Def.’s Resp. PSOF ¶¶ 15–16; DSOAF ¶¶ 9–16. Willig examined

NorthShore’s pre-merger pricing under eight contracts with its four largest payers, and found that NorthShore’s prices before the merger were actually below competi- tive levels. DSOAF ¶ 14. And according to Willig’s analysis, NorthShore raised its prices post-merger but not above competitive levels. Id. ¶¶ 15-16. To corroborate his findings, Willig relied on the Bain consulting documents, which as mentioned earlier say that NorthShore was pricing below market prior to the merger. Def.’s Exh. 87, Willig Rep. ¶¶ 72–76. NorthShore likewise points to these documents to counter the Plaintiff’s motion. Def.’s SJ/Daubert Reply Br. at 18 (citing DSOAF ¶¶ 10–11). But at this stage, the Court cannot weigh the conflicting evidence. Omnicare, Inc., 629 F.3d

at 704. In any event, a genuine dispute remains over market power. The Plaintiff attempts to undermine Dr. Willig’s opinions in three ways, but none succeed definitively on summary judgment. First, the Plaintiff argues that the FTC’s rejection of NorthShore’s learning-about-demand defense is prima facie evi- dence here. Pls.’s SJ/Daubert Br. at 7. Remember that during the FTC proceedings, NorthShore made the same argument it makes now through Willig: NorthShore raised its prices post-merger but only to account for below-market prices pre-merger.

The FTC found that this defense did not account for the price increases, and now the Plaintiff argues that that finding should be admitted as prima facie evidence. Section 5 of the Clayton Act grants prima facie weight to a “final judgment or decree ... rendered in any civil or criminal proceeding brought by or on behalf of the United States under the antitrust laws.” See 15 U.S.C. § 16(a). But this prima facie weight has limits. Section 5 grants prima facie weight only to matters as to which

issue preclusion would apply had the government itself brought the suit. See id. In other words, evidentiary use of prior judgments is determined by reference to the general doctrine of issue preclusion. But the Plaintiff makes no attempt to demon- strate that issue preclusion applies here. Either way, even if the FTC’s finding were given prima facie weight, the trier of fact here (in federal court) would not be bound by the finding. And, as described earlier, NorthShore has presented enough evidence to counter Vogt’s findings so that a reasonable jury could find in favor of NorthShore. Second, the Plaintiff argues that there is no admissible evidence that

NorthShore’s prices were below market before the merger. Pls.’s SJ/Daubert Br at 10. According to the Plaintiff, the Bain consulting documents are inadmissible hearsay, so Dr. Willig’s opinions based on those records are also inadmissible. Id. at 11.24 But the Court need not decide the admissibility of the Bain documents at this time. Whether admissible or not, the class’s assumes that Willig relied only on those rec- ords to reach his conclusions. But the evidence shows that Willig also performed an independent pricing analysis of NorthShore’s price increases. Def.’s Resp. PSOF

¶ 40. And that alone is enough to dispute the evidence that the Plaintiff has pre- sented. The Plaintiff also argues that the entirety of Dr. Willig’s pricing analysis is inadmissible under Daubert. Pls.’s SJ/Daubert Br. at 16. The class criticizes Willig’s use of raw prices.25 The parties agree that raw prices cannot be directly compared to analyze the effects of a merger because hospital services are differentiated products.

PSOF ¶ 35; Def.’s Resp. PSOF ¶ 35. According to the Plaintiff, Willig uses raw prices to draw his conclusions, so his analysis is unreliable. Pls.’s SJ/Daubert Br. at 16. But

24In litigating another issue, the class attempts to also use the Bain documents to its advantage, despite challenging their admissibility. Pls.’s SJ/Daubert Br. at 12. 25The Plaintiff also argues that Dr. Willig’s synthetic control group analysis is inad- missible under Daubert. Willig criticized the control group that Dr. Vogt used; Willig offered a counter control group obtained using an approach called synthetic control group analysis. Def.’s Exh. 87, Willig Rep. ¶¶ 172–79. But Willig does not use this control group as the basis of his own pricing analysis, so the Court need not evaluate its admissibility on summary judgment. Willig’s price analysis does account for the differentiation in the hospital market: the prices Willig relied on in his analysis were derived from a multiple regression that included controls for characteristics of the patient, service provided, plan type, and

payor. See R. 949, DSOAF ¶ 31; see also Def.’s Ex. 87, Willig Rep. ¶ 93 n.144. The Plaintiff does not otherwise explain how that methodology fails to take into account the differentiation between hospitals. Lastly, the class points to various other documents and testimony in arguing that it should win summary judgment on market power. But when viewed in NorthShore’s favor, the evidence does not dictate a victory for the Plaintiff. According to the Plaintiff, NorthShore’s own documents show that Northshore raised its prices

using its market power. Pls.’s SJ/Daubert Br. at 9–10. For example, a memorandum issued by NorthShore’s President, Mark Neaman, says that NorthShore’s revenue enhancements “could [not] have been achieved by either Evanston or Highland Park alone.” PSOF ¶ 9. But these do not conclusively demonstrate that NorthShore used its market power to increase its prices. The Plaintiff also points to the testimony of NorthShore executives Jeffrey Hillebrand and Jack Sirabian, Pls.’s SJ/Daubert Br. at

9, but just because Sirabian thought that pre-merger NorthShore offered a “fair price” for its services, and Hillebrand trusted Sirabian’s judgment, does not mean that those prices could not have been below market. PSOF ¶ 44–49. There is a genuine dispute over whether NorthShore exercised market power, so the Plaintiff is not entitled to summary judgment on this issue. b. Quality-of-Care Defense

Lastly, for one of NorthShore’s affirmative defenses, NorthShore asserts that the merger resulted in numerous efficiencies, including significant improvements in quality of care. The class argues that it is entitled to summary judgment on this af- firmative defense for two reasons: (1) the defense fails as a matter of law; and (2) even if it were a viable defense as a legal issue, the opinions of NorthShore’s quality of care expert, Dr. Gregg Meyer, are inadmissible. The Court agrees, at least on the latter argument. On the pure legal issue, the Plaintiff contends that neither the Supreme Court nor the Seventh Circuit have formally adopted a quality-of-care defense. Pls.’s

SJ/Daubert Br. at 26-27. It is true that the Supreme Court has cast some doubt on the availability of this defense, but the high court has not outright rejected it. And neither has the Seventh Circuit nor any of the courts in the cases cited by the class. In fact, the district court in F.T.C. v. Advocate Health Care actually acknowledged that both the agency’s Horizontal Merger Guidelines and some lower courts recognize the defense. 2017 WL 1022015, at *12 (citing F.T.C. v. H.J. Heinz Co., 246 F.3d 708,

720 (D.C. Cir. 2001)). Either way, the Court need not decide the availability of such a defense, because NorthShore has failed to produce evidence from which a reasona- ble jury could conclude that NorthShore’s improvements in quality were caused by the merger. To succeed on this defense, NorthShore needs to not only prove that the quality of care improved after the merger but also that the improvements were a result of the merger. The problem for NorthShore is that is provides insufficient evidence, even when viewed in its favor, to connect the dots between the improvements and the mer- ger. NorthShore primarily relies on Dr. Gregg Meyer, who provided an expert opinion

on whether the quality of patient care improved at Highland Park Hospital as a result of the merger. R. 943-2, Def.’s Exh. 128, Meyer Rep. ¶¶ 1, 11.26 The class argues that Meyer committed the post hoc ergo propter hoc fallacy, that is, he simply found an earlier event caused a later event merely because the earlier happened first. Pls.’s SJ/Daubert Br. at 30. The class is right. Meyer collected qualitative and quantitative data from after the merger, analyzed it, found quality improvements, and then jumped to the conclusion that the merger led to the improvements in quality. See

Def.’s Exh. 128, Meyer Rep. ¶ 283; R. 943-1, Def.’s Exh. 127, Meyer Rep. ¶ 101. But Meyer did not explain how he reached that conclusion and admitted to not having performed any kind of but-for causation analysis. R. 901-3, Pls.’s Exh. 13, Meyer Dep. Tr. at 29:17-21, 59:21 (“I haven’t said anything about what HPH would or would not do after the merger; had the merger not happened. I can’t know that. What I do know is what happened, and that’s what I discuss in the report that I produced.”). Neither

does NorthShore’s briefing provide any kind of causation explanation beyond pointing to either the question Dr. Meyer was asked to answer or his bare conclusion. See

26The Plaintiff proffered the expert opinion of Dr. Patrick Romano to rebut Dr. Meyer, but the Court need not decide the admissibility of Romano’s opinions because Meyer’s opinion does not satisfy Evidence Rule 702. Def.’s SJ/Daubert Reply Br. at 35–36 (citing DSOAF ¶¶ 50–52). Meyer’s opinion on the quality of care does not answer a relevant question, and is thus inadmissible.27 The remainder of the evidence that NorthShore relies on, namely the Obstet-

rics & Gynecology Annual Report and the Evanston Northwestern Healthcare Criti- cal Care Dashboard, also does not stave off summary judgment. Neither document discusses—much less establishes, even when viewed in NorthShore’s favor—the req- uisite causation. See DSOAF ¶ 17 (citing R. 943-4 (SEALED), Def.’s Exh. 153, ENHCA-042-004109; R. 943-4 (SEALED), Def.’s Exh. 154, ENHCA-042-004149). And without any kind of but-for or causation evidence, NorthShore cannot support its quality-of-care defense. Shafer v. Kal Kan Foods, Inc., 417 F.3d 663, 664 (7th Cir.

2005) (“Post hoc ergo propter hoc is not a good way to prove causation.”). Summary judgment is entered against the affirmative defense. IV. Conclusion NorthShore’s motion to decertify the class and motion for summary judgment are denied. The Plaintiff’s summary judgment motion is denied in large part and granted in part against the quality-of-care affirmative defense.

In moving forward with the litigation, in light of this Opinion, the parties shall confer and file a status report on (1) the restart of settlement negotiations; (2) the estimated number of trial days (dividing the estimate into the number of trial days

27The Plaintiff argues that the quality-of-care discussion in Guerin-Calvert and Dr. Willig’s expert reports should be excluded because neither are qualified to give a quality-of- care opinion. Pls.’s SJ/Daubert Br. at 34. But NorthShore does not rely on either expert’s report for its quality-of-care defense, so the Court need not decide the admissibility of those opinions. for evidence versus the number of days for jury selection, openings, closings, and de- liberations); (3) a schedule to disclose the defense’s substitute expert disclosure, lim- ited to substantially similar opinions (the motion to do so, R. 1113, is granted) and to

depose the expert; and (4) whether NorthShore plans on (or has by the time of the filing of the status report) filing a petition in the Seventh Circuit for permission to interlocutorily appeal under Civil Rule 23(f). The status report is due on March 14, 2023.

ENTERED:

s/Edmond E. Chang Honorable Edmond E. Chang United States District Judge

DATE: February 20, 2023