Negron v. CIGNA Corporation et al, No. 3:2016cv01702 - Document 401 (D. Conn. 2021)

Court Description: ORDER DENYING MOTION FOR CLASS CERTIFICATION (Docs. # 205 , 278 , 311 ). For the reasons set forth in the attached opinion, the Court GRANTS in part and DENIES in part plaintiffs' motion to strike (Doc. # 311 ). The Court DENIES Cigna's m otion to preclude the declaration and testimony of plaintiffs' expert Launce B. Mustoe (Doc. # 278 ). The Court DENIES plaintiffs' motion for class certification. (Doc. # 205 ). It is so ordered. Signed by Judge Jeffrey A. Meyer on 05/20/21. (Martinez, C.)

Download PDF

Negron v. CIGNA Corporation et al Doc. 401 UNITED STATES DISTRICT COURT DISTRICT OF CONNECTICUT KIMBERLY A. NEGRON et al., Plaintiffs, v. No. 3:16-cv-01702 (JAM) CIGNA HEALTH AND LIFE INSURANCE COMPANY, Defendant. ORDER DENYING MOTION FOR CLASS CERTIFICATION This is a putative class action involving allegations that defendant Cigna Health and Life Insurance Company (“Cigna”) fraudulently schemed to overcharge millions of people for prescription drugs in violation of the terms of their health plans. We are now at the class certification stage, and plaintiffs seek to certify classes and sub-classes under the Employee Retirement Income Security Act (“ERISA”) and the Racketeer Influenced and Corrupt Organizations Act (“RICO”). I will deny plaintiffs’ motion for class certification. Because it is evident that there are material differences in language among the thousands of health plans at issue in this action that govern whether the plaintiffs have suffered the same injury or any injury at all, I conclude that plaintiffs cannot carry their burden to show that there are questions of law or fact that are common to the class (much less that common questions will predominate over questions that require individual-specific resolution). As to the related pending motions, I will grant in part and deny in part plaintiffs’ motion to strike, and I will deny Cigna’s motion to preclude plaintiffs’ expert for class certification purposes. 1 Dockets.Justia.com BACKGROUND The basic background of this case is set forth in my prior ruling on Cigna’s partial motion to dismiss. See Negron v. Cigna Health and Life Ins. Co., 2020 WL 5216518 (D. Conn. 2020). The prescription drug transactions at issue here implicate four contractual relationships between: (1) an employee and his or her employer that provides prescription drug benefits under a health plan; (2) the employer and a health insurance company that underwrites and/or administers those benefits; (3) the health insurance company and a pharmacy benefit manager (“PBM”) that assists in administering the benefits; and (4) the PBM and the pharmacy that fills prescriptions covered under the plan. Id. at *1. Plaintiffs’ health plans describe what they must pay for prescription drugs in copayments and deductibles, while the PBM-pharmacy contracts at issue in this case state what a pharmacy must charge patients, the fee that the PBM will pay the pharmacy for filling a prescription, and the difference or “spread” between the patient charge and the pharmacy fee that the PBM will “claw back” for remittance to the health insurance company. Ibid. According to plaintiffs, “all Cigna plans uniformly stated that Cigna would provide prescription drug coverage for ‘Covered Expenses,’ which are ‘expenses for charges made by a Pharmacy, for Medically Necessary Prescription Drugs.’” 1 Plaintiffs use the term “Pharmacy Rate” to refer to these “charges made by a Pharmacy.”2 Plaintiffs also allege that all Cigna plans uniformly state that members “may be required to pay a portion of the Covered Expenses,” which plaintiffs allege is expressly defined to include copayments and deductibles. 3 Plaintiffs interpret this language to mean that “copayment and 1 Doc. #209 at 8 (emphasis in original). 2 Id. at 7. 3 Id. at 8 (emphasis in original). 2 deductible payments were limited to a ‘portion’ of (i.e., no more than the total) ‘charges made by a Pharmacy,’” 4 regardless of other terms in the individual plans. Plaintiffs characterize the “clawbacks” of the difference or “spread” between the member charge and the pharmacy fee as illegal “overcharges,” because their pharmacies charged them drastically more for prescription drugs than they were required to pay under their health plans, which plaintiffs argue capped their copayments and deductibles at the pharmacies’ transaction fee. See Negron, 2020 WL 5216518, at *1. They say that Cigna and its PBMs conspired to leverage their market power to contractually require pharmacies to charge these exorbitant and unauthorized amounts, in part by threatening to cut them out of Cigna’s network if they refused. Ibid. For its part, Cigna argues that the design of these plans is the result of a “choice that each employer makes.” 5 In Cigna’s telling, employers usually pay for prescription drug benefit costs, including PBM services, through either “traditional pricing” or “pass-through pricing.” For “pass-through pricing,” the plan sponsor’s prescription drugs costs are “typically equal to the pharmacy reimbursement rates,” but the fees for the plan’s administrative services are paid on a separate recurring basis. 6 For “traditional pricing,” or “spread pricing,” employers “negotiate predictable drugs costs for the plan year and pay for PBM services through a differential or ‘spread’ between the employer’s negotiated cost and the amount of the PBM’s (or its vendor’s) network pharmacy reimbursement.”7 Cigna essentially argues that plaintiffs are seeking a “pass- 4 Id. at 9 (emphasis in original). 5 Doc. #274 at 11. 6 Ibid. 7 Ibid. 3 through” pricing arrangement for plans with terms designed for a “traditional” or “spread” pricing arrangement. Plaintiffs seek to certify two classes, an ERISA class and a RICO class, each with a subclass. 8 Under plaintiffs’ amended class definitions, the Classes all include individuals residing in the United States and its territories who were enrolled in a Cigna or Cigna-affiliateissued or -administered health benefit plan or policy that: provided that a member “may be required to pay a portion of the Covered Expenses”; and provided that “Covered Expenses” are where an individual “incurs expenses for charges made by a Pharmacy”; and with respect to deductible payments, did not provide that the “Deductible payment” “will be based on the plan’s Prescription Drug Charge.”9 The Subclasses all include individuals residing in the United States and its territories who were enrolled in a Cigna or Cigna-affiliate-issued or -administered health benefit plan or policy that further provided that “in no event will the Copayment . . . exceed the amount paid by the plan to the Pharmacy.” 10 According to plaintiffs, the ERISA Class and Subclass contain all plan members with ERISA-governed plans that include this language, and the RICO Class and Subclass contain all plan members with ERISA or non-ERISA plans that include this language.11 The Classes also require each class member to have: paid a copayment or deductible payment to purchase prescription drugs pursuant to such plan or policy where according to the transaction data produced by Cigna in this action: the copayment or deducible payment exceeds the amount the pharmacy agreed with Cigna or the pharmacy benefit manager to accept for such 8 Docs. #205, 253. Plaintiffs moved to amend their proposed class definitions. Doc. #253. Both parties stipulated that the proposed Class and Subclass definitions in that motion should be substituted for the corresponding proposed definitions in plaintiffs’ motion for class certification. Doc. #260 at 2. I subsequently approved the parties’ stipulation and denied plaintiffs’ motion to amend the proposed class definitions as moot. Doc. #261. I also granted a partial motion to dismiss plaintiffs’ state law claims, Doc. #306; see also Negron, 2020 WL 5216518, and plaintiffs are no longer seeking to certify those claims as a Class and Subclass. 9 Doc. #253 at 1, 3. 10 Id. at 2, 4. 11 Doc. #209 at 9. 4 drugs on a transaction-by-transaction basis; and the excess amount is credited or transferred to Cigna or the pharmacy benefit manager.12 The Subclasses require almost the same second condition as the Classes, except that they only include individuals who paid a copayment to purchase prescription drugs, not those who paid a copayment or a deductible payment as for the Classes. 13 Plaintiffs interpret the Subclass plans’ additional language to provide a “second uniform contractual agreement that Class Members would never pay a copayment more than the amount that Cigna paid to the pharmacy.”14 They put forth a purportedly simple way to identify which plans fall into the Classes or Subclasses: plans with this exact operative language are in the Class or Subclass, and any plans without this language are not. 15 Cigna opposes plaintiffs’ motion for class certification, 16 and also moves to preclude the declaration and testimony of plaintiffs’ expert, Launce B. Mustoe, offered in support of plaintiffs’ motion. 17 Plaintiffs in turn seek to strike certain arguments made by Cigna in its opposition to plaintiffs’ motion for class certification and its motion to preclude.18 DISCUSSION Before turning to plaintiffs’ motion for class certification, I will first address the subsidiary motions including plaintiffs’ motion to strike and Cigna’s motion to preclude the testimony of plaintiffs’ expert for class certification purposes. 12 Doc. #253 at 2-4. 13 Id. at 2, 4. 14 Doc. #209 at 9 (emphasis in original). 15 Doc. #307 at 13. 16 Doc. #274. 17 Doc. #278. 18 Doc. #311. 5 A. Plaintiffs’ motion to strike After the motion for class certification and the motion to preclude Mustoe’s declaration and testimony were filed, plaintiffs moved to strike all of Cigna’s arguments and evidence, “now and for all future proceedings,” related to the use of the Document Source Tool (“DST”) Reports and their supposed shortcomings and those related to Cigna’s position that plaintiffs should have considered deductibles and out-of-pocket maximums with regard to the accumulator data.19 In the alternative, plaintiffs ask me to order Cigna to answer the “previously propounded interrogatories requesting it to link claims to specific plans, rather than place that burden on Plaintiffs under Rule 33(d).”20 Plaintiffs seek to invoke the Court’s powers under Rules 26 and 37 of the Federal Rules of Civil Procedure, and they also argue that Cigna abused its option in lieu of answering an interrogatory to produce business records under Rule 33(d). 21 Rule 26(e) provides that a party who has: responded to an interrogatory, request for production, or request for admission— must supplement or correct its disclosure or response: (A) in a timely manner if the party learns that in some material respect the disclosure or response is incomplete or incorrect, and if the additional or corrective information has not otherwise been made known to the other parties during the discovery process or in writing; or (B) as ordered by the court. Fed. R. Civ. P. 26(e). Rule 37(c)(1) provides in turn that if a party “fails to provide information . . . as required by Rule 26(a) or (e), the party is not allowed to use that information . . . to supply evidence on a motion, at a hearing, or at a trial, unless the failure was substantially justified or is harmless.” Fed. R. Civ. P. 37(c)(1). Rule 37(c)(1) further allows a court to “impose other appropriate 19 Doc. #311 at 1-2. 20 Id. at 2. 21 Doc. #312 at 26. 6 sanctions, including any of the orders listed in Rule 37(b)(2)(A)(i)-(vi).” Fed. R. Civ. P. 37(c)(1)(C). The sanctions may include: (i) directing that the matters embraced in the order or other designated facts be taken as established for purposes of the action, as the prevailing party claims; (ii) prohibiting the disobedient party from supporting or opposing designated claims or defenses, or from introducing designated matters in evidence; (iii) striking pleadings in whole or in part; [and] (iv) staying further proceedings until the order is obeyed. Fed. R. Civ. P. 37(b)(2)(A)(i)-(vi). As the Second Circuit has recognized, the purpose of allowing a court to preclude evidence under Rule 37 is to “prevent the practice of ‘sandbagging’ an opposing party with new evidence.” Haas v. Del. and Hudson Ry. Co., 282 F. App’x 84, 86 (2d Cir. 2008).22 By its terms Rule 37 requires courts to consider whether the failure to disclose “was substantially justified or is harmless,” and the Second Circuit in turn has listed a broad range of non-exclusive factors for courts to consider when deciding whether to preclude evidence as a sanction for failure to disclose. In one case, for example, the Second Circuit has instructed courts to consider factors including “(1) the willfulness of the non-compliant party or the reason for noncompliance; (2) the efficacy of lesser sanctions; (3) the duration of the period of noncompliance; and (4) whether the non-compliant party had been warned of the consequences of noncompliance.” Funk v. Belneftekhim, 861 F.3d 354, 366 (2d Cir. 2017). In another case, the Second Circuit has instructed courts to consider “(1) the party’s explanation for the failure to comply with the disclosure requirement; (2) the importance of the testimony of the precluded [evidence]; (3) the prejudice suffered by the opposing party as a result of having to prepare to meet the new [evidence]; and (4) the possibility of a continuance.” Patterson v. Balsamico, 440 F.3d 104, 117 22 Unless otherwise indicated, this ruling omits internal quotation marks, alterations, citations, and footnotes in text quoted from court decisions. 7 (2d Cir. 2006). Any sanctions imposed under Rule 37 must be “just,” and their severity must be “commensurate with the non-compliance.” Shcherbakovskiy v. Da Capo Al Fine, Ltd., 490 F.3d 130, 140 (2d Cir. 2007). I will begin by addressing the issue with the DST reports, followed by the claims data. 1. The DST Reports In plaintiffs’ first set of Requests for Production of Documents (“RFPs”), dated February 6, 2017, plaintiffs requested “Documents sufficient to identify the amount of all Clawbacks, Including the amount of Spread income generated” (RFP7) and “Documents sufficient to show the monthly revenues, profits, and other consideration received or credited from prescription drug premiums, Clawbacks, and Spread.” (RFP8). 23 In response, Cigna produced data for hundreds of millions of prescription drug transactions, including ones that were not related to the relevant plans. 24 Cigna’s counsel states in his declaration filed in support of Cigna’s opposition that at an in-person meet and confer in mid-July 2018, Cigna’s counsel mentioned the capability of searching documents in Cigna’s DST system, a method that was previously used for other litigation. 25 According to Cigna’s counsel, this is not the way the DST system is used in the regular course of business, and Cigna would need to see whether it could run DST reports to search for plans with certain language.26 In a letter dated December 17, 2018, Cigna told plaintiffs that Cigna had “not identified any system capabilities that can identify or pinpoint which [Cigna] clients had particular benefit language in their plan booklets during particular time 23 Doc. #312 at 9-10; Doc. #315-2 at 22, 24 (filed under seal). 24 Doc. #312 at 10. 25 Doc. #348 at 2 (¶ 4). 26 Ibid. 8 periods. It appears that a manual review of plan booklets would be required to determine such information.”27 In an email on January 3, 2019, plaintiffs’ counsel re-raised Cigna’s prior use of the DST reports in other litigation to search for certain plan language and asked why this method “cannot be modified to suit our needs.” 28 Plaintiffs’ counsel asked about this capability again six days later on January 9, 2019.29 Cigna’s counsel wrote in an email on January 14, 2019, “We are continuing to investigate Cigna’s ability to use search terms to identify plans,” as this “functionality is not used in Cigna’s normal course of business.”30 Cigna’s counsel also wrote that “Cigna has developed the ability to run text searches of the systems that store plan-related documents,” for clients with claims handled on certain systems, a capability that “can generate a list of plan documents with responsive language to an exact string of text.”31 Cigna’s counsel asked to set up a time for the parties to speak about this capability and the “limitations inherent in this approach.”32 During this meeting, Cigna’s counsel states that they “explained that DST reports do not contain information reflecting the end date for when a particular plan or plan-related document contained the specific phrase Plaintiffs requested for the DST search.”33 After the meeting, in a letter dated January 23, 2019, plaintiffs asked Cigna to produce all claims data that would be “relevant to proving liability and calculating damages for the Plaintiffs 27 Doc. #348-17 at 3; Doc. #372-2 at 3 (corrected version). 28 Doc. #348-4 at 2. 29 Doc. #348-5 at 2. 30 Doc. #348-6 at 2. 31 Ibid. 32 Ibid. 33 Doc. #348 at 2 (¶ 6). 9 and the Classes,” to the “extent it can be used to . . . tie transactions to employers and plans.”34 In early February 2019, Cigna produced a sample report from the DST system that included “any plan or related document in the DST system created between 1/1/2010 and 12/4/2018” that contained a particular phrase.35 Cigna’s counsel also noted that while the report included an “Effective Date” field that reflected the effective date of that document, Cigna was unable to provide the end date or the date that the relevant phrase no longer appeared in that document for that client and plan. 36 Plaintiffs’ counsel requested additional DST reports for other phrases with particular plan language later that month via email. 37 Plaintiffs assert that in producing the initial and subsequent DST reports, Cigna only provided six data fields—the account number, the account name, the document type, the CN number, the funding arrangement, and the effective date, but did not tell plaintiffs “what other information was searchable and could be included [in] the DST Reports.” 38 According to plaintiffs, the “Account Number” of a given employer “does appear in both the DST Report and the claims data,” and it is therefore “possible to use this DST Report to determine if a particular member had an employer that had a plan with the relevant language,” but that because some employers offer multiple plans, the DST reports could not match a given prescription drug transaction to a plan where an employer sponsored multiple plans with potentially different relevant language. 39 But according to Cigna’s counsel, during a meet-and- 34 Doc. #312 at 22; Doc. #315-2 at 70 (filed under seal). 35 Doc. #348-8 at 2. 36 Ibid. 37 Doc. #348-9 at 2. 38 Doc. #312 at 11-12. 39 Id. at 11-12. 10 confer held in March 2019, plaintiffs did not ask any other questions about the DST reports and “did not say that they needed information to try to distinguish among different plans if a [Cigna] plan sponsor had more than one benefit plan in a particular year.”40 In a follow-up email on April 5, 2019, plaintiffs asked Cigna’s counsel to “confirm that the [DST] reports contain all data concerning the plans available on the DST system.”41 Cigna’s counsel responded on April 24, 2019, “We are not sure what you mean by all data concerning the plans. However, with regard to the DST reports, we have provided you with all of the relevant data fields available for the report and which are necessary for the parties to review the client information, the document type, the plan identifier, the funding arrangement, and the effective date of the document.”42 Cigna claims that it had interpreted plaintiffs’ request for confirmation to ask only “whether there were other data fields on the specific DST report generated and produced in February,” in addressing the parties’ attempts to use the DST reports to identify plans with particular plan language. 43 In May 2019, plaintiffs served their fourth set of interrogatories, the second interrogatory of which asked Cigna to “[i]dentify, by claim number . . . each transaction in the prescription drug transaction data produced by Cigna on January 20, 2019 that was pursuant to a plan containing” particular plan language. 44 Cigna answered in June 2019: After reasonable investigation, the information requested in this Interrogatory is not maintained by Cigna in the ordinary course of business in the form and format requested by Plaintiffs. Further, to the extent that the information sought by this Interrogatory may be able to be ascertained from the prescription drug transaction data, DST language search results, and plan-related documents that [Cigna] has 40 Doc. #348 at 3 (¶ 8). 41 Doc. #348-12 at 2-3. 42 Doc. #348-13 at 2. 43 Doc. #347 at 11. 44 Doc. #312 at 13; Doc. #314-1 at 10. 11 produced in this Action, the burden on locating and identifying such information is the same for Plaintiffs as [Cigna] and [Cigna] directs Plaintiffs to those data and documents pursuant to Fed. R. Civ. P. 33(d). 45 Plaintiffs claim Cigna’s response to the interrogatory was “a nonresponsive answer because the documents Cigna referenced could not be used alone to tie a member’s claims to his or her Plan” and that “Cigna knew it had information in its possession that would tie claims to plans.” 46 Cigna claims that throughout this time, while it continued to produce DST reports in response to plaintiffs’ requests, “[n]one of those requests indicated what other information Plaintiffs needed to try to tie specific transactions to specific plans identified on the DST reports.” 47 Plaintiffs’ fifth set of interrogatories, dated July 19, 2019, again asked Cigna to “[i]dentify, by claim number . . . each copayment and/or deductible transaction in the prescription drug transaction data produced by Cigna that was pursuant to a plan” containing a series of relevant provisions.48 And plaintiffs’ sixth set of interrogatories, dated January 13, 2020, asked Cigna to identify each plan that contained particular phrases and for each identified plan to provide certain data produced in prior DST reports as well as other data including the “Group ID (which should include the Benefit Option/Plan of Benefit).”49 In response, Cigna’s counsel cited Rule 33(d) and directed plaintiffs to previously-generated DST reports and other documents.50 45 Doc. #312 at 13-14; Doc. #314-1 at 10. 46 Doc. #312 at 14. 47 Doc. #347 at 14. 48 Doc. #348-19 at 3-4. 49 Doc. #348-25 at 2. 50 Doc. #314-1 at 147-48. 12 In plaintiffs’ eighth set of interrogatories, served on February 25, 2020, shortly before plaintiffs’ motion for class certification was due, plaintiffs requested that Cigna identify each plan that contained a particular phrase and for each identified plan provide certain data produced in prior DST reports as well as other data fields including the “Group ID (which should include the Benefit Option/Plan of Benefit), and the BRANCH_POLICY_CDES.” 51 Cigna responded on March 26, 2020 that while it would produce another DST report for plans with a specific phrase, that report would “not include the ‘Group ID’ and the ‘BRANCH_POLICY_CDES’ as those are not fields available in the DST system.” 52 Cigna now argues that if plaintiffs “truly needed the ‘Group ID’ and ‘Benefit Option/Plan of Benefit’ from DST for any of the reports for purposes of their class certification motion and expert declaration, they could have sought to push back that deadline” for their class certification motion but did not.53 On June 18, 2020, in supplemental objections and responses to plaintiffs’ fourth and fifth sets of interrogatories, Cigna claimed that while it did not maintain the information in the form and format requested by plaintiffs, it could “however, link specific prescription drug transactions to specific plan language. In other words, [Cigna] can identify which plan document governs a particular prescription drug transaction using various data sources and documents,” and directed plaintiffs to the transaction data and the plan documents already produced. 54 Cigna added that it could “determine whether each prescription drug transaction made by a participant in a plan insured or administered by [Cigna] was adjudicated pursuant to a particular benefit plan” and 51 Doc. #348-27 at 2-3. 52 Doc. #314-1 at 155. 53 Doc. #347 at 15. 54 Doc. #275-12 at 4-5 (second supplemental objections and responses to interrogatory no. 2 from plaintiffs’ fourth set of interrogatories) (filed partially under seal with redactions); Doc. #312 at 18; see also Doc. #315-1 at 146-47 (supplemental objections and responses to plaintiffs’ fifth set of interrogatories) (filed under seal). 13 that it could “trace a particular participant’s benefit plan selection to the specific benefit plan and, in turn, plan documents that govern the prescription drug benefit for any particular transaction,” using a data field called “PLAN_OF_BEN,”55 which is purportedly the same as the “Ben Opt Code” field. 56 Plaintiffs argue that Cigna’s prior answers were designed to obfuscate how to achieve plaintiffs’ goal: linking specific prescription drug transactions to plans with particular language. Plaintiffs also argue that Cigna did not advise plaintiffs that the “PLAN_OF_BEN” or “Ben Opt Code” fields could be used for this purpose.57 Plaintiffs further argue that Cigna’s reliance on Rule 33(d) was a discovery abuse.58 Cigna responds that plaintiffs’ drafting of their fourth and fifth sets of interrogatories was flawed, but that Cigna nevertheless explained in its supplemental responses to those sets how Cigna could “confirm whether a particular transaction in the prescription drug transaction data was adjudicated pursuant to a plan containing certain plan language.”59 Cigna also argues that plaintiffs did not seek formal discovery into what fields were available in the DST reports until their tenth set of interrogatories served on June 30, 2020 or serve a Rule 30(b)(6) deposition notice until July 2, 2020, seeking discovery related to the DST system and reports. 60 Cigna did, in May 2020, in response to the eighth set of interrogatories, produce a DST report with the 55 Doc. #275-12 at 5-6 (second supplemental objections and responses to interrogatory no. 2 from plaintiffs’ fourth set of interrogatories) (filed partially under seal with redactions); Doc. #312 at 18-19, 32; Doc. #315-1 at 147-48 (supplemental objections and responses to plaintiffs’ fifth set of interrogatories) (filed under seal). 56 Doc. #312 at 18. 57 Id. at 19. 58 Id. at 21. 59 Doc. #347 at 16. 60 Id. at 17. 14 fields, “Contract State” and “Ben Opt Code.”61 Cigna has since produced additional DST reports with the Ben Opt Code field. 62 From my review of the parties’ submissions on the motion to strike and the path discovery took in this case, I find that Cigna acted inappropriately in a way that was neither substantially justified nor harmless in responding to plaintiffs’ requests once the purpose of plaintiffs’ requests became clear. Plaintiffs’ initial requests in 2017 and 2018 mainly appear to have related to its attempts to identify the amount of clawbacks and its attempts to identify which plans contained specific language. But in early 2019, after the prospect of using the DST reports was raised again, the focus of the parties appears to have clearly broadened beyond solely the capability of the DST reports and text searches to generate lists of plans that contained specific language to the ability to tie specific prescription drug transactions to specific employers and plans. These initial DST reports contained six data fields, but did not include the Ben Opt Code field that is now at the heart of plaintiffs’ motion to strike. Nor does it appear Cigna timely informed plaintiffs that this field existed. Instead, Cigna’s argument is that because plaintiffs did not ask about something they did not know existed, it is plaintiffs’ fault that they did not receive this information until more than a year later. By early 2019, plaintiffs’ goal—tying specific transactions to plans with specific language—was clear. Cigna’s evasive responses in April 2019 (“We are not sure what you mean by all data concerning the plans”) are compounded by Cigna’s arguments now that it simply thought plaintiffs were asking whether there were data fields other than the six provided on the 61 Doc. #312 at 17; Doc. #347 at 15; Doc. #348-28; Doc. #372-3 (corrected version); see also Doc. #348-38 at 6 (letter of June 27, 2020 from plaintiffs’ counsel confirming that Cigna produced DST reports responsive to the eighth set of interrogatories that included the “Contract State” and “Ben Opt Code” fields). 62 Doc. #347 at 18; Doc. #348-33 (letter of September 22, 2020 from Cigna’s counsel); Doc. #372-4 (corrected version) see also Doc. # 348-34 (letter of September 25, 2020 from Cigna’s counsel); Doc. #372-5 (corrected version). 15 specific DST report generated and produced in February 2019. This self-serving interpretation is compounded by Cigna’s responses to plaintiffs’ interrogatories served in 2019 and 2020 that redirected plaintiffs to the already-produced data and reports and relied on Rule 33(d). It should not have taken until plaintiffs’ sixth and eighth sets of interrogatories served in January and February 2020 that specifically referenced a Group ID or Benefit Option for Cigna to either produce to plaintiffs what they were seeking or tell plaintiffs that other data fields existed for the DST reports. Cigna certainly should not have waited until June 2020 to finally reveal that it could, in fact, link specific prescription drug transactions to specific plan language or that it could determine whether each prescription drug transaction made by a participant in a plan was adjudicated pursuant to a particular benefit plan using a specific data field. This is plainly what plaintiffs sought almost a year and a half before Cigna finally admitted it had this capability. Cigna’s assertions that it either did not understand what plaintiffs were seeking due to the drafting of their interrogatories or that the fault lies with plaintiffs because they did not seek discovery into what other data fields were available in the DST reports strains belief when it was clear what plaintiffs were seeking and that Cigna had the capability to make those connections. Plaintiffs could not know what they did not know. To answer interrogatories by resort to document production under Rule 33(d), the responding party needs to “specify[] the records that must be reviewed, in sufficient detail to enable the interrogating party to locate and identify them as readily as the responding party could.” Fed. R. Civ. P. 33(d). Cigna very well could have still relied on Rule 33(d) if it produced the data and DST reports containing the information it admits can be used to link specific prescription drug transactions to specific plan language. All that said, I am not convinced that the appropriate remedy is to strike from all consideration any of Cigna’s arguments related to plaintiffs’ use of the DST reports and their 16 alleged shortcomings. The fault I find with Cigna’s responses to plaintiffs’ discovery requests relates to Cigna’s failure to inform plaintiffs that it had the capability to link specific prescription drug transactions to specific plan language. Cigna’s arguments related to the DST reports in its motion to preclude and its opposition to the motion for class certification go well beyond that capability. For instance, Cigna’s motion for class certification focuses on three particular flaws with plaintiffs’ expert’s use of the DST reports, namely, that the phrases used to generate the reports do not match the Class language, that they do not reliably identify when a plan was operative, and that the expert’s methodology both overcounts and undercounts the number of Class transactions. 63 And while Cigna’s motion to preclude does argue that plaintiffs’ expert only proposes to use the DST reports to identify transactions connected to a particular “sponsor/employer,” not to connect specific transactions to specific plans with the relevant Class and Subclass terms, 64 not all of Cigna’s arguments regarding the DST reports relate to this issue. Accordingly, while I will not consider Cigna’s arguments to the extent that those arguments are related to the Ben Opt Code data field or the ability to link specific prescription drug transactions to specific plan language, I will also decline to strike all arguments related to those issues for the duration of the case. As Cigna has represented that it has produced DST reports with the Ben Opt Code field, including re-running previously provided DST reports, 65 I also decline to order plaintiffs’ alternative request for relief. 2. The claims data In addition to the issues with the DST reports, plaintiffs argue that Cigna did not produce all relevant claims data, contrary to Cigna’s assertions. On January 23, 2019, plaintiffs asked 63 Doc. #274 at 51. 64 Doc. #278-1 at 11; Doc. #279 at 11 (filed under seal). 65 Doc. #347 at 16. 17 Cigna to produce “all claims data that is relevant to proving liability and calculating damages for the Plaintiffs and the Classes,” to the extent that it could be used to, among other things, “identify and calculate overcharges [and] calculate deductible payment balances.”66 Cigna responded on February 7, 2019 that the prescription drug transaction data it produced “included all available fields that are relevant for purposes of this case,” and that for the deductible information, “Cigna’s data only contains data fields regarding the dollar amount applied toward a customer’s deductible for a particular transaction,” as the data was “transactionbased and thus does not track the accumulation toward a particular customer’s overall plan deductible.” 67 Cigna asserts that it “pointed out to Plaintiffs their failure to consider accumulation of deductibles and out-of-pocket maximums” in July 2019 and again in November 2019, in explaining Cigna’s issues with plaintiffs’ damages model.68 Nevertheless, Cigna asserts, plaintiffs never sought discovery requesting Cigna to produce the data necessary to “reprocess” the prescription drug transactions to account for the accumulation of deductibles and out-ofpocket maximums. 69 And plaintiffs’ theory of liability is indeed premised on the argument that Cigna violated the Class members’ plan terms every time they were overcharged for a prescription on a transaction-by-transaction basis.70 Given that plaintiffs have repeatedly asserted that their theory of the case is based on a transaction-by-transaction or claim-by-claim basis and that calculations determining the impact 66 Doc. #314-1 at 164. 67 Id. at 172. 68 Doc. #347 at 20. 69 Ibid. 70 Doc. #309 at 28-29. 18 of the accumulation of deductibles and out-of-pocket maximums are “not required, or, if they are, they are better performed at some later point in the litigation after class certification has been granted,” 71 and given that plaintiffs were informed of what Cigna views as flaws in their damages model months before seeking class certification, I will decline to strike Cigna’s arguments related to this issue from the briefing on the motion to preclude and the motion for class certification. To the extent that plaintiffs seek to explore this issue, they may seek additional discovery. B. Cigna’s motion to preclude expert testimony of Launce B. Mustoe The Federal Rules of Evidence allow for the admission of expert testimony if the following conditions are met: (a) the expert’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue; (b) the testimony is based on sufficient facts or data; (c) the testimony is the product of reliable principles and methods; and (d) the expert has reliably applied the principles and methods to the facts of the case. Fed. R. Evid. 702; see also Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579 (1993). The Court’s role is to act as a “gatekeeper” to ensure that the expert’s testimony is relevant and rests on a reliable foundation. See In re Vivendi, S.A. Sec. Litig., 838 F.3d 223, 253 (2d Cir. 2016). The Second Circuit has further “distilled Rule 702’s requirements into three broad criteria: (1) qualifications, (2) reliability, and (3) relevance and assistance to the trier of fact.” In re LIBOR-Based Fin. Instruments Antitrust Litig., 299 F. Supp. 3d 430, 466 (S.D.N.Y. 2018) (citing Nimely v. City of New York, 414 F.3d 381, 396-97 (2d Cir. 2005)). “[W]hen an expert opinion is based on data, a methodology, or studies that are simply inadequate to support the conclusions reached, Daubert and Rule 702 mandate the exclusion of that unreliable opinion 71 Doc. #307 at 44. 19 testimony.” Amorgianos v. Nat’l R.R. Passenger Corp., 303 F.3d 256, 266 (2d Cir. 2002). But as “long as an expert’s analysis is reliable at every step, it is admissible.” In re Mirena IUS Levonorgestrel-Related Prods. Liability Litig. (No. II), 982 F.3d 113, 123 (2d Cir. 2020). Neither the Supreme Court nor the Second Circuit have definitively decided whether the Daubert standard governs the admissibility of expert evidence submitted at the class certification stage. See Kurtz v. Costco Wholesale Corp., 818 F. App’x 57, 61 n.3 (2d Cir. 2020) (declining to address the issue). Some district courts have applied the Daubert standard at the class certification stage. See, e.g., In re Teva Sec. Litig., 2021 WL 872156, at *10 (D. Conn. 2021); In re LIBOR, 299 F. Supp. 3d at 470-71; Chen-Oster v. Goldman, Sachs & Co., 114 F.Supp.3d 110, 114 (S.D.N.Y. 2015). These courts do, however, take into account that the purpose of the Daubert inquiry at the class certification stage is different from that at the trial stage: “The question is not whether a jury at trial should be permitted to rely on the expert’s report to find facts as to liability, but rather whether the Court may utilize it in deciding whether the requisites of Rule 23 have been met.” In re LIBOR, 299 F. Supp. 3d at 471. Accordingly, “to the extent that flaws in expert testimony proffered at class certification do not warrant that testimony’s exclusion by the Court as gatekeeper under Daubert at the threshold, those flaws may nonetheless be considered in the Rule 23 analysis undertaken by the Court as trier of fact.” Ibid. Plaintiffs engaged Mustoe as an expert to provide his opinion as to “whether it is possible to calculate ‘clawbacks’ . . . for each proposed class and subclass on a classwide basis.” 72 Mustoe’s declaration defines a clawback to be when a member “pays a copayment or deductible that exceeds the Pharmacy [R]ate and Cigna claws back the overcharges.”73 According to 72 Doc. #278-1 at 8; see also Doc. #207-1 at 4 (filed under seal). 73 Doc. #309 at 26; see also Doc. #207-1 at 15 (filed under seal). 20 Mustoe, clawbacks can be calculated using the prescription drug transaction data set and the DST reports produced by Cigna. 74 Mustoe relies on a pharmacy transaction data set produced by Cigna that contains transactions for both Class and non-Class members. 75 The transaction data set includes more than 500 million transactions.76 Mustoe relies on twenty relevant data fields within the set that he found were “material to calculating ‘Clawbacks’” on a transaction-by-transaction basis.77 The DST reports Mustoe used provide the account number and name; the document type; the “CN” number—a document ID for a particular document of a particular client; the funding arrangement, that is, whether the plan is subject to ERISA, fully insured, a cash management program, or an Advanced ERISA benefit; and the effective date of the document. 78 Cigna produced four DST reports that each contain one of the following phrases: (1) “exceed the amount paid by the plan to the pharmacy” (DST Report 1); (2) “incurs expenses for charges made by a pharmacy” (DST Report 2); (3) “required to pay a portion of the Covered Expenses for Prescription Drugs and Related Supplies” (DST Report 3); and (4) “prescription drug charge” (DST Report 4). 79 At the time of Mustoe’s declaration, plaintiffs also sought a fifth DST report that includes the phrase “may be required to pay a Deductible, Copayment or Coinsurance requirement for Covered Expenses for Prescription Drug Products.”80 Mustoe claims that using DST Reports 1-3, he can identify the plans related to the Classes and Subclasses, as well as when 74 Doc. #278-1 at 9-11; see also Doc. #207-1 at 4 (filed under seal). 75 Doc. #278-1 at 9; see also Doc. #207-1 at 9 (filed under seal). 76 Doc. #207-1 at 10 (filed under seal). 77 Doc. #309 at 21; see also Doc. #207-1 at 9-10 (filed under seal). 78 Doc. #309 at 19; see also Doc. #207-1 at 12 (filed under seal). 79 Doc. #278-1 at 12; see also Doc. #207-1 at 13 (filed under seal). 80 Doc. #278-1 at 12; see also Doc. #207-1 at 13 (filed under seal). 21 a given member began making claims under particular plans. 81 Mustoe proposes to use DST Reports 4 and 5 to determine when the employer no longer used a plan with the actionable Class language. 82 Using the transaction data set and the DST reports, Mustoe believes he can calculate the aggregate clawbacks for each Class and Subclass.83 Mustoe asserts that the clawback amount can be calculated by identifying the negative amounts in the PHARM_PD_AMT field, which Mustoe understands to be the field “normally used to account for the amounts that Cigna pays to the pharmacy, but when the amount in the field is negative, it signifies a Clawback.” 84 Mustoe then believes that he can match claims in the transaction data set to particular plan documents.85 Plaintiffs believe that Mustoe’s methodology will enable them to “(1) identify[] each specific plan associated with each specific prescription drug claim and (2) electronically review[] each such plan to determine if it has the operative language.” 86 Mustoe believes that he can calculate the total amount of overcharges, that is, the total amount of clawbacks, by “summing the negative amounts in the PHARM_PD_AMT field.” 87 Mustoe then plans to allocate damages to the ERISA or RICO Classes or Subclasses based on whether a given plan is recorded as a plan subject to ERISA in the DST Reports. 88 81 Doc. #278-1 at 13-14; see also Doc. #207-1 at 13-14 (filed under seal). 82 Doc. #278-1 at 13-14; see also Doc. #207-1 at 13-14 (filed under seal). 83 Doc. #278-1 at 15; Doc. #309 at 26; see also Doc. #207-1 at 14 (filed under seal). 84 Doc. #309 at 26; see also Doc. #207-1 at 15 (filed under seal). 85 Doc. #309 at 21; see also Doc. #207-1 at 15 (filed under seal). 86 Doc. #309 at 21-22. 87 Doc. #309 at 26; see also Doc. #207-1 at 17 (filed under seal). 88 Doc. #278-1 at 14-15; see also Doc. #207-1 at 18 (filed under seal). 22 Cigna moves to exclude Mustoe’s declaration and testimony.89 Cigna’s argument focuses on two main grounds: (1) Mustoe’s methodology to calculate damages through the clawbacks is “unreliable and inapposite to plaintiff’s theory of the case,” 90 and (2) Mustoe’s methodology to identify Class and Subclass members is “unreliable and produces incongruous results.” 91 Cigna first argues that “clawbacks” are not the correct measure of relief.92 Cigna instead asserts that the true measure of damages, if plaintiffs succeed on their claims, is the difference between what members paid for prescriptions and what they should have paid under the terms of the plan. 93 Instead, Mustoe’s method appears to assume that any clawback necessarily has a oneto-one relationship to any overcharge, while Cigna asserts that clawbacks are really “credits permitted by a series of contractual revenue sharing arrangements” and are therefore an “arbitrary measure of class-wide damages.” 94 Cigna further argues that an individual’s deductibles and out-of-pocket maximums would affect what amount that individual should have paid on a claim-to-claim basis over the course of a plan year, and that looking at each claim in isolation, as Mustoe’s aggregate method appears to do, does not correctly measure the amount of alleged damages.95 According to Cigna, in order to calculate the alleged damages to any given Class member, one must look cumulatively at each Class member’s individual claims history and plan-specific terms to determine how much the member should have paid in a given transaction, and that, further, resolving each claim on its 89 Doc. #278. 90 Doc. #278-1 at 18. 91 Id. at 28. 92 Id. at 18. 93 Id. at 19. 94 Id. at 18-19. 95 Id. at 19. 23 own would have a “ripple” effect on subsequent claims. 96 That is, if a Class member paid more than they should have paid on their first claim, that means they also would have paid more towards meeting their deductible for the plan year. If the Class member paid what plaintiffs assert they should have, then they would also have paid less towards their deductible. This in turn, according to Cigna’s theory, could affect subsequent claims by operation of a kind of “deductible shifting,” which would mean that “some class members would have been further from satisfying any applicable deductible requirement had their transactions [been] processed under Plaintiffs’ theory and, consequently, the amount those class members would have to pay toward their deductible for later transactions would have increased.”97 Cigna also argues that Mustoe’s methodology ignores aspects of the plan designs or plan-specific terms, the effect of out-of-pocket maximums, and plaintiffs’ own theory of liability. 98 Mustoe does admit in his deposition that his method does not re-adjudicate or reprocess each claim or prescription drug transaction.99 Nor does it take into account the effect of deductibles. 100 Mustoe also admitted that there could be a “ripple” effect of sorts from changing the deductible transactions.101 Cigna also asserts that Mustoe admitted that “had he been asked to calculate the true measure of harm for Plaintiffs’ claims—what members of the class and subclass should have paid under the terms of their plans, as interpreted by Plaintiffs—he would have used a completely different method,” 102 akin to that he uses for conducting pharmacy 96 Id. at 21-24. 97 Id. at 23. 98 Id. at 21-27. 99 Id. at 15; see also Doc. #277-4 at 22, 185:1-13 (filed under seal). 100 Doc. #309 at 29. 101 Doc. #278-1 at 23 n.12. 102 Doc. #344 at 19. 24 benefit audits in his normal course of work, “with participants and plan sponsors paying the correct amounts and accounting for the effect of adjustments on subsequent transactions.” 103 But like Cigna’s other objections to Mustoe’s declaration, what Cigna is really attacking is plaintiffs’ theory of liability in this case. As plaintiffs have asserted, under their theory, while the clawback is itself “not the harm,” the amount of the clawback is “indisputable evidence of the amount of each overcharge.”104 Plaintiffs’ theory may very well be wrong. But at this stage, where I am only deciding whether to consider Mustoe’s declaration and testimony for the purposes of deciding plaintiffs’ motion for class certification, I am not all that concerned that Mustoe’s methodology is at odds with Cigna’s theory of alleged liability. Cigna’s objections go to the nature of plaintiffs’ allegations and the theory they are seeking to advance through a class action. Mustoe’s declaration and testimony, on this point, is simply how one would calculate the damages, given the relevant data sets, if one accepts plaintiffs’ theory of liability: that the clawbacks are necessarily the amount of the overcharge and the correct measure of harm. I am well aware of the difference in opinion between plaintiffs and Cigna on the liability measurement issue and the allegations in this case more broadly, and I do not find Mustoe’s methodology so unreliable as to preclude its consideration under Daubert. Cigna’s second objection is to Mustoe’s method of identifying Class and Subclass members. Cigna argues that the phrases used to generate the DST reports do not match the Class definition language, and that Mustoe’s use of the DST reports is at times underinclusive and at other times overinclusive.105 Cigna provides examples of plans that it argues would meet the 103 Doc. #278-1 at 20 n.10. 104 Doc. #307 at 15. 105 Doc. #278-1 at 28-29. 25 Class definition, but that would not be picked up by the DST reports. 106 Cigna also challenges Mustoe’s assertion that his methodology can determine when a Class plan with the relevant definitions becomes operative (or when it stopped being operative).107 Finally, Cigna takes issue with plaintiffs’ position that certain variations of plan language are still in the Class or Subclass, including variations such as use of the word “incur” versus “incurs,” 108 and that Mustoe’s method includes some participants as Subclass members who are not Class members.109 In reply, Cigna also argues that Mustoe failed to test his method’s validity.110 Similar to Cigna’s objection to Mustoe’s method for calculating damages, at least some of these objections go to Cigna’s overarching argument that sharing mere “snippets” of plan language across thousands of plans does not a Class make. Cigna’s objections also go to the question of ascertainability, which is part of the class certification analysis under Rule 23(b)(3). I think that, like Cigna’s objection to Mustoe’s damages calculation method, Mustoe’s method is not so unreliable as to require preclusion at the class certification stage. Further, I credit plaintiffs’ assertion that “mismatches” between Mustoe’s method and plans that meet the Class and Subclass definitions can be identified and then resolved through revised or updated DST reports. 111 At the very least, I can take Cigna’s objections as to the flaws in Mustoe’s method to identify Class and Subclass members into account in considering plaintiffs’ motion for class certification rather than as grounds for preclusion. 106 Id. at 29-31. 107 Id. at 31-33. 108 Id. at 34-35. 109 Id. at 35-36. 110 Doc. #344 at 20. 111 Doc. #309 at 35. 26 Because my inquiry under Daubert is guided by the nature of my gatekeeper position at this point in the case—the class certification stage versus a jury trial—I find that Mustoe’s declaration and testimony, with Cigna’s objections in mind, can properly be considered when deciding plaintiffs’ motion for class certification. Accordingly, I will deny Cigna’s motion to preclude the testimony of Launce B. Mustoe. C. Plaintiffs’ motion for class certification Rule 23 of the Federal Rules of Civil Procedure permits a federal court to certify a class action by which named plaintiffs may litigate claims on behalf of a class of similarly situated aggrieved class members. Class actions “are an exception to the general rule that one person cannot litigate injuries on behalf of another.” Langan v. Johnson & Johnson Consumer Cos., Inc., 897 F.3d 88, 93 (2d Cir. 2018) (citing Wal-Mart Stores, Inc. v. Duke, 564 U.S. 338, 348 (2011)). In order for the Court to grant plaintiffs’ motion to certify a class under Rule 23(b)(3), plaintiffs must satisfy seven requirements. First, plaintiffs must satisfy the four threshold requirements of Rule 23(a)—numerosity, commonality, typicality, and adequate representation of the class, and then they must satisfy two more requirements under Rule 23(b)(3)— predominance and superiority. See Scott v. Chipotle Mexican Grill, Inc., 954 F.3d 502, 512 (2d Cir. 2020); In re Petrobras Sec., 862 F.3d 250, 260 (2d Cir. 2017). In addition, plaintiffs must also satisfy “an implied requirement of ascertainability” to ensure that the class is sufficiently definite so that the Court can determine whether any particular individual is a class member. See ibid. I will initially focus on just one of the requirements: commonality. Rule 23(a)(2) requires the existence of “questions of law or fact common to the class.” Fed. R. Civ. P. 23(a)(2). Each 27 class must therefore involve a common question of law or fact capable of resolving an issue central to the validity of each class member’s claim at once. See Wal-Mart, 564 U.S. at 350-51. “Where the same conduct or practice by the same defendant gives rise to the same kind of claims from all class members, there is a common question.” Johnson v. Nextel Commc’ns Inc., 780 F.3d 128, 137 (2d Cir. 2015).While the class members’ claims “need not be identical for them to be common,” ibid., “[w]hat matters to class certification . . . is not the raising of common ‘questions’—even in droves—but rather, the capacity of a class-wide proceeding to generate common answers apt to drive the resolution of the litigation. Dissimilarities within the proposed class are what have the potential to impede the generation of common answers.” Wal-Mart, 564 U.S. at 350 (emphasis in original). ERISA plans are essentially contracts, and courts use “familiar rules of contract interpretation” when addressing an ERISA plan. Lifson v. INA Life Ins. Co. of N.Y., 333 F.3d 349, 353 (2d Cir. 2003) (per curiam). One such well-established rule is that I must read a plan “as a whole, [and] giv[e] terms their plain meanings.” Fay v. Oxford Health Plan, 287 F.3d 96, 104 (2d Cir. 2002). To be sure, “contract claims generally may be appropriate for class certification where form agreements are at issue.” Wing v. Metro. Life Ins. Co., 2007 WL 9814564, at *6 (S.D.N.Y. 2007). On the other hand, “courts properly refuse to certify breach of contract class actions where the claims require examination of individual contract language” where the language variations are material to the issue of breach. In re U.S. Foodservice Inc. Pricing Litig., 729 F.3d 108, 124 (2d Cir. 2013). Further, “class-wide resolution of contract claims becomes problematic in the absence of form agreements, or where a number of different form agreements are at issue.” Wing, 2007 WL 9814564, at *7. 28 The heart of the dispute between plaintiffs and Cigna on the motion for class certification is whether the construction and interpretation of the various plans at issue defeat commonality. According to plaintiffs, the common questions in this case are “whether class members paid too much in violation of the uniform language of their Plans,” because Cigna “incorrectly calculated the cost-share payments of every member of the ERISA Classes by not using the Pharmacy Rate as required by the Plan language,” 112 and “whether Defendants’ contract language prohibited them from ‘clawing back’ the copayment and deductible ‘overpayments’ that Defendants charged to Class Members.” 113 Plaintiffs assert that there are only three key provisions, uniform across the Class and Subclass plans, that matter for the Class and Subclass claims: (1) “Schedule of Benefits;” (2) “Covered Expenses”; and (3) “Your Payments.” 114 Cigna in turn argues that there are key material variations among the Class and Subclass plans, namely: (1) the meaning of the term “Covered Expenses”;115 (2) how the plans describe and define “Your Payments”; 116 and (3) how much participants pay for certain prescription drugs. 117 Cigna also argues that resolving ambiguities in the various plans’ language would require reference to extrinsic evidence of the individual plan-sponsor’s intent.118 If the issue is, as plaintiffs frame it, whether Cigna miscalculated the cost-share payments of every Class member by not using the Pharmacy Rate, then any variations in how each Class or 112 Doc. #209 at 15. 113 Doc. #205 at 3. Although plaintiffs’ briefing refers to “defendants” in the plural, two previously named defendants—Cigna Corporation and OptumRx, Inc.—have been dismissed from this action. Docs. #119, #318. 114 Doc. #307 at 16-17. 115 Doc. #274 at 24-25. 116 Id. at 26-27. 117 Id. at 27-30. 118 Id. at 30-32. 29 Subclass member’s cost-share payment is calculated or in how the Pharmacy Rate is defined would be material. To put it more simply, the common question that must exist across the Class and Subclass plans hangs on the determination of how much each putative Class member should have paid for prescription drugs. If the plans provide for different methods or means of calculating that amount—rather than a single method or means—then the calculation of the amount each putative Class member should have paid requires reference to the specific terms of individual plans apart from only those particular terms chosen by plaintiffs. If the amount each putative Class member should have paid can only be determined by consulting varying terms of individual plans, then there is no common question. Indeed, some of the plans whose language meets the Class or Subclass definition also have other provisions that may impact the calculation of what members should have paid. These variations can be illustrated by reference to a few examples that demonstrate a number of the variations Cigna has identified. Plaintiffs, for their part, maintain that these are not variations, or, if they are, they are not material ones. Take, for example, the O. Berk Company, LLC Plan (the “O. Berk Plan”). The O. Berk Plan contains the relevant Class language: it states that the member “may be required to pay a portion of the Covered Expenses,” 119 it provides that Covered Expenses are when an individual “incurs expenses for charges made by a Pharmacy,”120 and it does not include language stating that the deductible payment “will be based on the plan’s Prescription Drug Charge.” The O. Berk Plan defines “Charges” as “the discounted amount that the pharmacy benefits manager makes available to the Insurance Company with respect to Participating 119 Doc. #275-47 at 37. 120 Id. at 39. 30 Pharmacies.” 121 For retail prescription drugs, the O. Berk Plan provides for “[n]o charge after $15 copay” for generic drugs and “[n]o charge after $35 copay” for certain brand-name drugs. 122 The O. Berk Plan’s “Your Payments” section also provides that coverage for prescription drug purchases at a pharmacy is “subject to the Copayment or Coinsurance shown in the Schedule,” and that, “[i]n no event will the Copayment exceed the retail cost of the Prescription Drug or Related Supply.” 123 As a second example, consider the GL&V USA Inc. Plan (the “GL&V Plan”). This Plan also meets the Class definition. 124 The GL&V Plan’s Schedule defines “Charges” differently from the O. Berk Plan, defining the term to mean the “amount charged by the Insurance Company to the plan when the Pharmacy is a Participating Pharmacy,” 125 or what Cigna terms the “Client Rate.” 126 In the table describing the benefits at participating versus non-participating pharmacies, the GL&V plan provides that at participating pharmacies, generic drugs have “[n]o charge after $10 copay,” while brand-name drugs are charged to the participant at “30% subject to a minimum copay of $25 and a maximum copay of $50.” 127 The “Your Payments” section then states that “Coverage for Prescription Drugs and Related Supplies purchased at a Pharmacy is subject to the Copayment or Coinsurance shown in the Schedule, after you have satisfied your Prescription Drug Deductible, if applicable,” 128 but it does not contain any “in no event” 121 Id. at 37. 122 Ibid. 123 Id. at 39. 124 Doc. #275-39 at 44, 46. 125 Id. at 44. 126 Doc. #274 at 15. 127 Doc. #275-39 at 44. 128 Id. at 47. 31 language like either the O. Berk Plan or the Subclass definition. For a third example, consider the Law Offices of Peter G. Angelos Plan (the “Angelos Plan”), which contains the Class, but not Subclass, language.129 Like the GL&V Plan, the Angelos Plan also references the Client Rate, defining “Charges” to mean the “amount charged by the Insurance Company to the plan when the Pharmacy is a Participating Pharmacy.” 130 But unlike the GL&V Plan, the Angelos Plan provides for different payments for retail prescription drugs. For generic prescription drugs, the Angelos Plan provides for “[n]o charge after $10 copay,” while providing for “[n]o charge after $20 copay” for brand-name drugs. 131 The Angelos Plan’s “Your Payments” section also contains “in no event” language, this time providing that “[i]n no event will any Copayment or Coinsurance, as applicable, exceed the cost of the Prescription Drug or Related Supply.” 132 There are a number of differences among these three plans despite all three meeting the Class definition. For one, each of the three Plans provide the “amount you pay” for generic and brand-name drugs, but these amounts are each calculated differently. Further, while the GL&V and Angelos Plan define “Charges” the same way, the O. Berk Plan defines the term differently. And the O. Berk and Angelos Plans both provide “in no event” language that differ both from each other and from the Subclass “in no event” language, while the GL&V Plan contains no such language at all. I will begin with the issue of the variation in the plans for the “amount you pay” section for generic and brand-name retail prescription drugs under each Plan (putting aside for now the 129 Doc. #275-44 at 43, 45. 130 Id. at 43. 131 Ibid. 132 Id. at 46. 32 “in no event” language present in both the O. Berk and Angelos plan that differs from that in the Subclass language). The Angelos Plan states that there is “[n]o charge after $10 copay” for generic drugs and “[n]o charge after $20 copay” for brand-name drugs, while the O. Berk Plan states that there is “[n]o charge after $15 copay,” and “[n]o charge after $35 copay” for each, respectively. The GL&V Plan has the same language for generic drugs as the Angelos Plan, but has a different calculation for brand-name drugs: “30% subject to a minimum copay of $25 and a maximum copay of $50.” Consider, for example, a brand-name drug that costs $36. Under the Angelos and O. Berk Plans, members would presumably pay the copay of $20 and $35, while under the GL&V Plan, members would pay $25, as 30% of $36 is $10.80, but the drug is subject to a minimum co-pay of $25. Yet if the drug only costs $10, presumably that would mean that members under the O. Berk Plan would still pay $35, those under the Angelos Plan would pay $20, and those under the GL&V Plan would pay $25 (again, setting aside the potential effect of the “in no event” language in two of those three plans). Of course, plaintiffs contend that the challenged “clawbacks” “rarely occurred on branded drugs” and that “Tier 1 generics” are the “heart of the case.”133 But even a generic drug presents this kind of variation under each of the plans. Consider a $9 generic drug. Under the Angelos and GL&V Plans, members would only pay $10, that is, the co-pay for generic drugs. Certainly for the GL&V Plan, which lacks any “in no event” language, this means that members would pay $1 more than the cost of the drug. In contrast, under the O. Berk Plan, members would presumably still pay the $15 copay, which is $5 more than the cost of the drug or the Pharmacy Rate, again setting aside the effects of that plan’s “in no event” language. 133 Doc. #308 at 20. 33 Plaintiffs’ response at oral argument to these variations in the “amount you pay” sections of these Plans was that language like “no charge after $10 copay” is “subordinate to the master definition of ‘a portion of Covered Expenses’” 134 and that there is a class-wide question: whether the “no charge after $10 copay take[s] precedence over the class language which says you are limited to pay a portion of Covered Expenses.”135 In plaintiffs’ view, the class language is controlling over any other variations in the plan and the amount members pay is therefore necessarily limited by the Pharmacy Rate. And in that view, even if the “no charge after $10 copay” language conflicts with the Class language, the Class language controls. But it is “well-established that [courts] disfavor readings of a contract that render provisions of an agreement superfluous.” CP III Rincon Towers, Inc. v. Cohen, 666 F. App’x 46, 51 (2d Cir. 2016). Even though “superfluity is not necessarily fatal to a contract, [courts] do take the fact that a given interpretation would render a provision superfluous into account in evaluating whether the relevant provisions, read together, are ambiguous.” Id. at 52; see also Utica Mut. Ins. Co. v. Fireman’s Fund Ins. Co., 957 F.3d 337, 346 (2d Cir. 2020) (rejecting argument for interpretation of limits prescribed in insurance policy that “would render significant portions of the Schedules meaningless”). And it is “a fundamental rule of contract construction that ‘specific terms and exact terms are given greater weight than general language.’” Aramony v. United Way of Am., 254 F.3d 403, 413 (2d Cir. 2001) (quoting Restatement (Second) of Contracts § 203(c) (1981)). Is the Class language “general” language? Cigna argues that it is. Cigna asserts that the Class language should be read as “a basic description of prescription drug benefits,” namely, that 134 Doc. #378 at 102:22-25. 135 Id. at 102:13-16. 34 Cigna or the plan sponsor “‘will provide coverage’ if a participant ‘incurs expenses’ for nonspecific ‘charges made by a Pharmacy,’ and the amount of such coverage is ‘shown in the Schedule.’” 136 Plaintiffs respond that the Schedule of Benefits “expressly incorporates the definition of Covered Expenses,”—that is, where a member “incurs expenses for charges made by a Pharmacy”—“into the deductible and copayment definition (by providing that a member ‘may be required to pay a portion of the Covered Expenses’ as the amount of the copayment or deductible).” 137 According to plaintiffs, “[b]ecause this is the only definition for the amount of the copayment and deductible in the Plan boilerplate, there can be no alternative interpretation.” 138 But the different “amount you pay” language in the O. Berk, GL&V, and Angelos Plans conflicts with plaintiffs’ assertion that this is the “only definition,” or at least the only explanation in the plans of the amount members pay for copayments. After all, the GL&V Plan expressly provides that there is “[n]o charge after $10 copay” for generic drugs, which seems to unilaterally impose a $10 copay for generic drugs without regard to the Pharmacy Rate. Giving this provision its plain meaning does not lead to absurd results: while it would mean that a member would overpay by $5 for a $5 generic drug, the same member would save $40 on a $50 generic drug.139 To me, this appears to be an intractable conflict between different provisions within a given plan. If the Class language is controlling, that seems to make any “amount you pay” 136 Doc. #274 at 30-31. 137 Doc. #307 at 32. 138 Ibid. (emphasis in original). 139 To be sure, plaintiffs stress that prescription drug transactions or claims where a member’s “cost-share payments were lower than the Pharmacy Rate” are not part of the Class because in these cases, “there was no overcharge and no clawback.” Doc. #307 at 34-35. And, of course, plaintiffs’ Class definitions only include transactions where the copayment or deductible payment exceeds the Pharmacy Rate. Doc. #253 at 2. While the member who saves $40 on a $50 generic drug may not be in the Class, the provision of their plan that gives them that savings does matter for the purposes of interpreting the plan and determining whether purported variations are material. 35 language superfluous—after all, why provide that there is “[n]o charge after $10 copay” if the amount a member pays is necessarily limited to no more than the Pharmacy Rate, whatever that happens to be? If the Pharmacy Rate is truly controlling, why wouldn’t the “amount you pay” just be “no greater than the charges made by a Pharmacy,” to use the Class language? Or, to reuse the earlier example, if a plan wanted to prevent members from paying the $50 Pharmacy Rate for generic drugs, why wouldn’t the “amount you pay” be “the charges made by the Pharmacy, up to a total of $10,” incorporating the $10 copay limit in the plan as written? And if specific and exact terms are to be given greater weight than general language, how should I determine which terms are specific and exact and which terms are general without reference to the entire plan? Of course, contract interpretation is a task that courts routinely undertake, and I could try to apply the rules of contract interpretation to each of these three plans to determine what amount members should pay for prescription drugs. But the problem is that there is not just one contract at issue; rather, plaintiffs seek to certify as a Class members of potentially thousands of plans and involving as many as 500 million transactions. The variations or ambiguities a court could resolve in one or three or even ten plans could become infeasible once the court is asked to consider thousands of plans, especially if those variations or ambiguities require reference to plan language outside of the Class language that will control whether and how much any Class member has overpaid. Perhaps I could resolve the question of whether the “amount you pay” language is subordinate to the Class language across all tens or hundreds or thousands of plans that contain the “amount you pay” language, thereby establishing that it is a common question. Given that these plans are largely comprised of boilerplate language, that could be possible. But the 36 variations in the “amount you pay” language are not the only potential variations among the Class and Subclass plans that Cigna has identified. For that matter, plaintiffs’ responses to the plans Cigna identifies seem to undermine their position that there is a common question across thousands of putative Class plans. Returning to the issue of the “in no event” variations referenced above and what effect those may have on the Class and Subclass language (in addition to the “amount you pay” provisions), the O. Berk and Angelos Plans both contain “in no event” language that is different from each other and from the Subclass definition, which includes plans that contain the language: “in no event will the Copayment . . . exceed the amount paid by the plan to the Pharmacy.” 140 According to plaintiffs, the Subclass “in no event” language “reinforc[es] the Pharmacy Rate cap.” 141 This would mean that the “in no event” language in the O. Berk and Angelos Plans similarly provides some kind of cap on the copayment paid by the member. But what is that cap? For the O. Berk Plan, it is the “retail cost” of the prescription drug. For the Angelos Plan, it is the “cost” of the prescription drug. But is the “retail cost” of the drug necessarily the same as the “cost” of the drug? Are those terms the same as the Pharmacy Rate, the cap on the Subclass Plans? The differing language in the O. Berk and Angelos Plans affects the calculation of what members should have paid to the pharmacy, the question purportedly common to all the Class and Subclass plans. But if these terms have different meanings—and the interpretation of their meanings would require reference to the respective plans—can it really be said, as plaintiffs argue, that the calculation of how much each member should have paid to the pharmacy is a common question? 140 Doc. #253 at 2. 141 Doc. #209 at 7. 37 In their response to Cigna’s surreply, plaintiffs assert that “Cigna ignores the fact that Plans define the terms ‘actual cost’ and ‘full cost’ through the overarching cost-share definition as a portion of expenses for charges made by a pharmacy.” 142 But this argument is simply a retreat to plaintiffs’ interpretation of the Class language and their position that the Class language controls over individual-specific provisions of the plans. Another Class plan, this one for Hampton Affiliates (the “Hampton Plan”), contains a fourth variation on the “in no event” language, adding on to the Subclass language. This Plan provides “[i]n no event will the Copayment or Coinsurance for the Prescription Drug or Related Supply exceed the amount paid by the plan to the Pharmacy, or the Pharmacy’s Usual and Customary (U&C) charge.” 143 The Hampton Plan goes on to state that the U&C charge means “the established Pharmacy retail cash price, less all applicable customer discounts that Pharmacy usually applies to its customers regardless of the customer’s payment source.”144 Is the “retail cash price” in the Hampton Plan the same as the “retail cost” in the O. Berk Plan or the “cost” in the Angelos Plan? And how does the U&C charge interact with the “amount paid by the plan to the Pharmacy” that plaintiffs argue is a cap on members’ payments? Does it simply serve to explain the meaning of the latter, or is it a separate term? Both of these provisions are in the Hampton Plan, and to determine how much a given member would have paid would require a plan-specific interpretation of these provisions. Plaintiffs note that “[t]he Subclass Member Plans all stated that ‘[i]n no event’ would copayments ‘exceed the amount paid by the plan to the Pharmacy, or the Pharmacy’s Usual and 142 Doc. #361 at 6. 143 Doc. #275-27 at 35 (emphasis added). 144 Ibid. 38 Customary (U&C) charge.’” 145 Of course, this version of the Subclass definition matches the Hampton Plan, but tellingly, the U&C portion does not appear in the amended Class and Subclass definitions plaintiffs submitted to the Court. 146 It may be the case that plaintiffs are now defining the Subclass language to include the U&C portion, but the plaintiffs’ ever-evolving class definitions undermine their argument that there are common issues across thousands of plans that are suitable for class-wide resolution. Further, in response to Cigna’s citation of variations in the “in no event” language in various plans’ “Your Payments” provisions, plaintiffs argue that “by definition, those plans are not part of the Subclasses because they do not have the specific language that defines the Subclasses.” 147 But plaintiffs miss the point. These plans, while not containing the Subclass language as plaintiffs now choose to define it in their briefing (but not in their already onceamended class definitions), do still contain the relevant Class language and are therefore still part of the Class. The extent to which these “in no event” variants complicate the interpretation of the Class language for these Class plans has bearing on the Rule 23(a) question of commonality across the Class. The Hampton Plan has another variation, one that this time applies to members who “insist on a more expensive ‘brand-name’ drug where a ‘generic’ drug would otherwise have been dispensed.”148 Those members “will be financially responsible for the amount by which the cost of the ‘brand-name’ drug exceeds the cost of the ‘generic’ drug, plus the required 145 Doc. #307 at 12 (emphasis in original); see also Doc. #209 at 9 (same). 146 Doc. #253 at 2. 147 Doc. #307 at 17 (emphasis in original). 148 Doc. #275-27 at 34. 39 Copayment identified in the Schedule.” 149 Thus, for a $40 brand-name drug with a $10 generic equivalent, members would presumably, under the plain meaning of the Plan, pay the $30 difference plus the $50 copay provided for in the Schedule, for a total of $80, even though that Schedule also provides that there will be “[n]o charge after $50 copay” for brand-name drugs with a generic equivalent, 150 and the $80 total is greater than the $40 amount presumably charged by the pharmacy. According to plaintiffs, this simply means that “a member must pay the difference between cheaper generic and more expensive branded drugs” and “has nothing to do with copayments or deductibles,”151 without any elaboration on how this does not conflict with the Class language plaintiffs argue limits the amount members should pay to not more than the Pharmacy Rate. And further complicating things is the Hampton Plan’s “in no event” language. If that language refers to the $40 cost of the brand-name drug, it could mean a limit on the $50 copay to $40. But even under that interpretation, the member would presumably pay the $30 difference plus the capped $40 copay, for a total of $70, which is still greater than the cost of the $40 drug. In another variation, as Cigna points out, some plans’ “Covered Expenses” sections provide specific language on the amount a member must pay. 152 One of these plans, belonging to the Boyd Gaming Corporation (the “Boyd Plan”), provides that “Covered Expenses” are where a member “incurs expenses for charges made by a Pharmacy,” as required by the Class language, but also provides that certain drug classes are “covered at 100% member liability (member pays 149 Ibid. (emphasis added). 150 Id. at 32. 151 Doc. #307 at 22. 152 Doc. #274 at 25. 40 full cost with CIGNA’s discount).” 153 This appears, by its plain meaning, to be a potentially different amount than the “charges made by a Pharmacy” or the Pharmacy Rate. Plaintiffs argue that the Boyd Plan “merely provides that the member will pay as a deductible the ‘full cost with CIGNA’s discount’ (i.e., the amount ‘charged by the pharmacy’),” and assert that this is therefore not a material variation. 154 But plaintiffs fail to elaborate on why the “full cost with CIGNA’s discount” is the same as the amount “charged by the pharmacy” and I cannot determine from the Class language alone why that would necessarily be the case. In yet another variation, Cigna notes that some plans “may also describe participants’ prescription drug payments by reference to a variety of unspecified ‘costs.’” 155 For example, the M/A-COM Technology Solutions, Inc. Plan (the “M/A-COM Plan”)—a Subclass plan— provides that for generic maintenance drugs at participating pharmacies, members pay “$10 for the first 3 fills, then the plan pays 100% after plan deductible. For refills after the 3rd fill, you pay 100% of the cost.” 156 To what “cost” does “100% of the cost” refer? Is it the same as the “retail cash price” in the Hampton Plan or the “retail cost” in the O. Berk Plan or the “cost” in the Angelos Plan? Plaintiffs’ only response in their reply is that because the deductible “is still defined as a ‘portion’ of ‘Covered Expenses,’” the “deductible payment under this Plan is like the deductible payment under any other Class deductible plan.”157 And in their response to Cigna’s surreply, plaintiffs only state in a footnote that because the M/A-COM Plan is “subject to the ‘in no event’ maximum regardless of any other copayment provision,” there is no material 153 Doc. #275-31 at 41. 154 Doc. #307 at 19 (emphasis in original). 155 Doc. #274 at 29. 156 Doc. #275-45 at 34. 157 Doc. #307 at 21. 41 variation. 158 But again, just as with the “amount you pay” variations, plaintiffs have not established how, relying on the Class language alone and without reference to the plans as a whole, I can determine that the “in no event” language necessarily controls over any other copayment provisions. Another variation Cigna identifies appears in the Greenbrier Companies, Inc. Plan (the “Greenbrier Plan”), which provides that maintenance medications “must be filled through home delivery; otherwise after 3 retail refills, the normal copay will double (up to a maximum of the actual cost of the drug).” 159 Plaintiffs’ response is simply that the term “actual cost of the drug” is the same as the “amount charged by the pharmacy,” and is therefore consistent with the Class definition, without elaborating on why these terms necessarily have the same meaning. 160 Again, is the “actual cost” in the Greenbrier Plan the same as the “retail cash price” in the Hampton Plan or the “retail cost” in the O. Berk Plan or the “cost” in the Angelos Plan? Under plaintiffs’ theory, these terms all have the same meaning, and that meaning matches the Class language, the “charges made by a Pharmacy.” But if that’s the case, then why do any of these plans use any language other than the “charges made by a Pharmacy?” And why should that general language control over what appears, from their plain meaning, to be more specific or exact terms, such as “actual cost” or “retail cost”? Plaintiffs misplace their reliance on Smith v. United HealthCare Servs., Inc., 2002 WL 192565 (D. Minn. 2002). That case concerned health plan terms that entitled subscribers to “pay the ‘lesser of’ the fixed dollar co-pay or the ‘Prescription Drug Cost.’” Id. at *1 (emphasis in original). Indeed, the construction in the named plaintiff’s plan was substantially more explicit 158 Doc. #355 at 7-8 n.5. 159 Doc. #275-40 at 34. 160 Doc. #307 at 21. 42 than plaintiffs’ interpretation of the Class language here. That plaintiff’s plan provided that “[i]f the Prescription Drug Cost is less than the Co-payment, the Co-payment does not apply and the Covered Person pays the Prescription Drug Cost.” Id. at *2. To be sure, the court in Smith determined that variations in the plans’ terms “Prescription Drug Cost,” “actual cost,” and “cost” “may have the same effective meaning” and that “[h]ow these terms will be defined will involve issues of legal construction common to all potential class members.” Id. at *3. But the plan language at issue in Smith that provided the same limit that plaintiffs see in their plans was quite clear. Those plans do not appear to have had the same conflict that is present here between terms that purport to limit Class members’ payments to one amount and terms that seem to require another limit. The variations don’t stop there. The Metropolitan Nashville and Davidson County Employee Benefit Board Plan (the “Nashville Plan”) contains the Class language that members “may be required to pay a portion of the Covered Expenses,” but goes on to state, “[t]hat portion is the Coinsurance,” without a reference to copayments or deductibles.161 While plaintiffs’ counsel maintained that the “portion” is “expressly defined to include copayments and deductibles,”162 the Nashville Plan—which is a Class plan under the Class definition—appears to define that portion differently. Plaintiffs’ response is that the Nashville Plan is “not an ERISA plan, and so is only relevant to Plaintiffs’ RICO claim” and plaintiffs then assert that while the Plan “does not provide for copayments,” it does “limit[] deductible payments to the amounts paid to the pharmacy.”163 Then, in a footnote, plaintiffs go through a series of definitions in the Plan, namely, how the Plan defines “charges” and “Deductible,” which then requires the 161 Doc. #275-46 at 50. 162 Doc. #210 at 2 (¶ 5). 163 Doc. #307 at 19. 43 definition of “Covered Service.” 164 According to plaintiffs, taking these definitions together, the “amount of the deductible is the amount of the actual billed charges, the amount charged by the pharmacy,” which is “completely consistent with the Class Plan language.” 165 Plaintiffs’ argument for why the language variations of the Nashville Plan are consistent with the Class language depends on reference to other parts of the Plan—something plaintiffs otherwise insist is not necessary because the Class and Subclass Plans purportedly all raise common questions. In plaintiffs’ response to Cigna’s surreply, plaintiffs then simply assert that the Nashville Plan is a “‘one off’ client-generated plan for Nashville and Davidson County. It does not have the typical boilerplate language,” and that “[a]lthough it defines the portion to be coinsurance, it is a deductible plan with the class definition language.”166 What troubles me about these back-and-forth exchanges between Cigna and plaintiffs is that plaintiffs repeatedly put forth rationalizations about why a purported variation is not really a variation or at least is not a material one. Plaintiffs very well may be correct about some of these interpretations and about how to read certain plan-specific variant terms to be equivalent to the “charges made by a Pharmacy” Class language. But the issue here is exactly that: plaintiffs’ responses to these variations in language require interpretations of countless individual plans to explain why this-or-that variation is or is not consistent with the Class language. In a case with thousands of Class and Subclass Plans and some 500 million transactions, this kind of whack-a-mole approach to what appear to be material variations is not tenable. Nor can I simply take plaintiffs’ word that other variations do not exist in all the other thousands of plans that fall under the Class and Subclass definitions. As far as I can tell, it is holes with moles 164 Id. at 19 n.8. 165 Ibid. 166 Doc. #355 at 7-8 n.5. 44 all the way down. Cf. Rapanos v. United States, 547 U.S. 715, 754 & n.14 (2006) (plurality opinion of Scalia, J.). Plaintiffs argue that “Cigna suggests that Plaintiffs have the duty to identify every Plan with the variations Cigna has identified, regardless of how unreasonable Cigna’s views may be,” and that the “fact that Cigna has not come forward with a list of Plans with these allegedly contradictory terms speaks volumes.”167 But it is plaintiffs who bear the burden of satisfying Rule 23(a)’s threshold requirements, including that of commonality, not Cigna’s burden to disprove it. See In re Patriot Nat’l, Inc., Sec. Litig., 828 F. App’x 760, 764 (2d Cir. 2020). In addition, for examples where—as with the Nashville Plan—plaintiffs admit that a plan “does not have the typical boilerplate language” but still has the Class definition language, this only raises additional questions as to how many other plans in the thousands of plans with the Class language also—by plaintiffs’ own admission—do not have the “typical boilerplate language” and therefore require some kind of interpretation of the Plan’s definitions. This is not a case where the determination of the “truth or falsity” of plaintiffs’ contention— whether Class members paid too much because Cigna incorrectly calculated members’ cost-share payments by not using the Pharmacy Rate—will “resolve an issue that is central to the validity of each one of the claims in one stroke.” Wal-Mart, 564 U.S. at 350. Cigna goes on to argue that plaintiffs have failed to show that there is a common question across the Class as to the clawbacks, 168 and as to the discretionary authority Cigna does or does not have to interpret and apply the terms of each given plan.169 I need not consider these arguments, because I find that the number of variations among the specific plans identified by 167 Doc. #361 at 7. 168 Doc. #274 at 33-34. 169 Id. at 34-38. 45 Cigna and the fact that they require individual-specific reference to terms in the plans outside the Class definition defeat any assertion of commonality across the ERISA Class and Subclass. I need not consider the other class certification requirements under Rule 23, because plaintiffs’ failure to establish commonality as required by Rule 23(a) is fatal to their motion for class certification. I note, however, that plaintiffs would likely face at least as many difficulties satisfying the predominance requirement as they do the commonality requirement. See Petrobras, 862 F.3d at 270-71 (discussing “predominance” requirements); Gorss Motels Inc. v. Sprint Commc’ns Co., L.P., 2021 WL 1207440, at *5 (D. Conn. 2021) (no predominance because “consent issues will vastly vary between class members who have had years of business dealings with Wyndham and/or Sprint on the basis of multiple and successive agreements and related documents that set the terms for their dealings and their communications”). Plaintiffs’ RICO Class and Subclass fail for similar reasons. In their motion for class certification, plaintiffs state that the alleged uniform misrepresentation at issue is that “copayments and deductibles would only be a ‘portion’ of Covered Expenses and, for Subclass members, that ‘in no event’ would copayments ‘exceed the amount paid by the plan to the Pharmacy.’” 170 As plaintiffs assert, the ERISA Class and Subclass members are subsumed within the RICO Class and Subclass. 171 But for the same reason I cannot resolve the ERISA Class and Subclass claims, I also cannot resolve the RICO Class and Subclass claims, as I cannot determine whether the allegedly uniform misrepresentations are actually uniform—or even misrepresentations—without reference to the individual plans. Cf. U.S. Foodservice, 729 F.3d at 118 (noting that “fraud claims based on uniform misrepresentations to all members of a class are 170 Doc. #209 at 26. 171 Id. at 9. 46 appropriate subjects for class certification because, unlike fraud claims in which there are material variations in the misrepresentations made to each class member, uniform misrepresentations create no need for a series of mini-trials.”). Plaintiffs argue that the Second Circuit’s decision in In re U.S. Foodservice Inc. Pricing Litigation, concerning a RICO and breach of contract class action, provides the “blueprint for this case and is dispositive on most Rule 23 issues.”172 But an examination of the decision in U.S. Foodservice makes clear the stark differences between the facts in that case and the facts here. The decision in U.S. Foodservice concerned USF, a food distributor who provided food products and services to customers under cost-plus contracts: the “cost” being the price at which USF purchased the goods from its supplier and the “plus” being a surcharge, often expressed as a percentage increase over the cost. 729 F.3d at 112. While USF’s contracts had various ways of calculating the cost (some on based on price lists, others based on the local market), the class action concerned those contracts that set the cost at the “invoice cost,” that is, the price on the invoice from the supplier. Ibid. The suppliers often also provided promotional allowances or discounts to distributors. Under the contracts, USF was typically permitted to keep the benefit of any such allowances and was not required to pass the savings onto the customer. Id. at 112-13. The plaintiffs alleged that USF engaged in fraudulent schemes using a set of shell companies that enabled USF to inflate the cost in the cost-plus contracts and then disguise the proceeds of the inflation through purported promotional allowances. Id. at 113. The plaintiffs sought to certify a class of customers with these cost-plus contracts with USF on their breach of contract claims. USF argued, as is relevant here, that common questions would not predominate because of material variations across the contracts and because some of 172 Doc. #209 at 13. 47 the contracts required customers to satisfy minimum purchase requirements before they were entitled to cost-plus pricing. Id. at 123. USF argued that to determine whether USF had breached the contracts pursuant to the contracts’ materially different terms (especially due to numerous different definitions of the terms “vendor” and “promotional allowance”) would require reference to individualized extrinsic evidence. Id. at 123-24. The Second Circuit, while noting that courts “properly refuse to certify breach of contract class actions where the claims require examination of individual contract language,” found that the language variations USF identified were not material to the issue of breach. Id. at 124. But the Second Circuit was considering different contracts and different evidence than the case here. There, USF’s “own expert testified that the contracts ‘essentially all [say] the same thing’ and that in the food service industry, ‘[i]t [is] well understood . . . what a cost plus contract is,” while USF’s “own auditor found that USF’s contracts are consistent in how they define invoice cost.” Ibid. Here, there has been no such testimony, and in fact, plaintiffs’ own expert did not even review any plan documents.173 It is also particularly significant that the class contracts in U.S. Foodservice all defined “cost” to mean the “invoice cost.” But as I have recounted above, the various Class plans use a variety of different provisions to define how much a member owes for a particular prescription. Other than plaintiffs’ repeated assertions that “charges made by a Pharmacy” is a limiting definition of sorts on every other term or provision in a given plan (indeed, even ones that appear to contradict that definition), plaintiffs have not shown that the relevant definitions are consistent across the Class plans, at least not without an examination of the individual plans themselves. 173 Doc. #278-1 at 13; Doc. #344 at 21. 48 Further, the fact that the Uniform Commercial Code (“UCC”) governed all of the costplus contracts also supported the conclusion in U.S. Foodservice that common issues would predominate. See 729 F.3d at 125; see also In re U.S. Foodservice Inc. Pricing Litig., 2011 WL 6013551, at *13 (D. Conn. 2011) (noting that the UCC “provides uniformity to any issue that may arise over the meaning of ‘vendor’ in USF’s cost-plus contracts”). There is no such code providing a similar unifying role for interpretation of the thousands of health plans at issue in this case. Plaintiffs’ reliance on U.S Foodservice is also flawed because of the nature of the claims asserted. The U.S. Foodservice plaintiffs sought to bring contract claims that “focus[ed] predominantly on common evidence to determine whether, in fact, USF used controlled middlemen to inflate invoice prices and whether such a practice departs from prevailing commercial standards of fair dealing so as to constitute a breach.” 729 F.3d at 125. The plaintiffs’ allegations were that USF breached the cost-plus contracts because using the shell companies “to inflate costs was dishonest, commercially unreasonable, and a breach of USF’s implied duty of good faith,” all claims that could be evaluated on a class-wide basis because the contracts were “materially uniform insofar as they imposed the same duty of good faith,” and the question of whether USF violated that duty was common to all class members. Ibid. The Second Circuit anticipated certain common issues across the class on the plaintiffs’ claims: USF’s creation and control of the shell companies, the services the shell companies provided, USF’s efforts to hide the nature of the shell companies from the customers, and “trade usage concerning controlled middlemen” like the shell companies. Id. at 125-26. Here, the question is not whether Cigna violated a duty of good faith to the class members. Nor is there an external code like the UCC that may shed light on how certain terms 49 are uniformly used across the industry. Rather, the question is, as framed by plaintiffs, whether Cigna miscalculated the cost-share payments of every Class member by not using the Pharmacy Rate. And to determine what the Class members should have paid or whether Cigna should have used the Pharmacy Rate requires reference to the individual plans, with material variations in terms and provisions that may affect how that calculation is done. This is a question that is not common across all plans, and it is fatal to plaintiff’s efforts to certify a class across thousands of plans. And importantly for the plaintiffs’ RICO claims, the very misrepresentations at issue in U.S. Foodservice were not in the cost-plus contracts themselves, but in the shell-companyrelated invoices USF mailed to its cost-plus customers. 729 F.3d at 118. The misrepresentation was uniform across the class: “that the cost component of USF’s billing was based on the invoice cost from a legitimate supplier and not from a shell [company] controlled by USF and established for the purpose of inflating the cost component.” Ibid. “While each invoice obviously concerned different bills of goods with different mark-ups, the material misrepresentation— concealment of the fact of a mark-up inserted by the [shell company]—was the same in each.” Ibid. While plaintiffs here allege that Cigna “went to great lengths to hide [its] ‘clawback’ scheme,” by enforcing “gag clauses” against network pharmacies and hiding the clawbacks from its employer clients, 174 and that there is some common evidence as to the relationship between Cigna and other entities to create the clawback mechanism, 175 the basis of plaintiffs’ RICO Class in their motion for class certification is the alleged misrepresentations within the language of the 174 Doc. #209 at 11-12. 175 Doc. #307 at 47. 50 plans themselves. With the sheer number of plans at issue, and the variations within the plan language already identified, I cannot find that plaintiffs have met their burden under Rule 23 to certify a class. Plaintiffs’ briefing highlights evidence of internal corporate communications that they believe shows Cigna knew that it was wrongly profiting from clawbacks and how some at Cigna feared it might face a class action lawsuit one day.176 But the issue before me is not whether Cigna may be liable to someone. It is whether any liability of Cigna can be adjudged on the massive class-wide basis that plaintiffs have chosen to propose as the vehicle for deciding this case. It cannot. Accordingly, I will deny the motion for class certification. CONCLUSION For the reasons set forth above, the Court GRANTS in part and DENIES in part plaintiffs’ motion to strike (Doc. #311). The Court DENIES Cigna’s motion to preclude the declaration and testimony of plaintiffs’ expert Launce B. Mustoe (Doc. #278). The Court DENIES plaintiffs’ motion for class certification. (Doc. #205). It is so ordered. Dated at New Haven this 20th day of May 2021. /s/ Jeffrey Alker Meyer Jeffrey Alker Meyer United States District Judge 176 See, e.g., Doc. #209 at 10. 51

Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You should read the full case before relying on it for legal research purposes.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.