The Direct Purchaser Plaintiffs ("DPPs") and End Purchaser Plaintiffs ("EPPs") in this multidistrict litigation move for class certification of their antitrust claims challenging defendants' "reverse payment" patent litigation settlement that they contend led to inflated costs for the brand name and generic versions of lidocaine patches. Defendants oppose, arguing that the highly stratified distribution chain for pharmaceutical drugs means that the DPPs and EPPs cannot show injury or damages through classwide proof. Instead, defendants assert that individual questions centered around each DPP's or EPP's place in the purchasing chain create a multitude of individualized issues that swamp any common ones. Defendants also seek to exclude plaintiffs' experts' opinions as based on inherently unreliable assumptions.

Plaintiffs' alleged antitrust injury is, fundamentally, that defendants were allowed to overcharge for their brand and generic lidocaine patches. Under plaintiffs' well-supported theory, with which defendants disagree but cannot effectively undercut at this stage because it depends on disputed facts that will be resolved at the merits stage, the lidocaine patches were sold by defendants at higher prices than would have existed in the but-for world.

Defendants' oppositions to the motions for class certification boil down to the following:

• The distribution chain in the prescription drug market is very stratified and complex.

• The direct purchasers and various types of end purchasers have numerous, different agreements between themselves so that determining whether any particular plaintiff was injured and how to apportion damages between the plaintiffs necessarily involves individualized questions that are undeniably complex.

Both statements are true. But common questions predominate. Did defendants engage in anticompetitive conduct? Did that conduct lead to overcharges for brand and generic lidocaine patches? What aggregate damages resulted from the overcharges? This case is more appropriate for class certification than not.

Defendants' arguments against certification would result in certification of very few antitrust cases as a class actions, despite repeated direction from the Supreme Court that the class device is particularly useful in the antitrust context. Plaintiffs' experts (and indeed, defendants' experts) have presented reliable, statistically-sound methods to determine not only classwide injury but also proof of aggregate damages. The disputes about what the but-for price should have been absent the anticompetitive conduct and when the generic lidocaine patches would have entered the market but-for that conduct will be resolved largely through common proof at summary judgment or trial. Once those issues are resolved, the experts have shown how to determine aggregate damages - carving out purchases made by entities or individuals who are not included within the class definitions - with classwide proof. That there may then be a need to conduct individualized analysis to determine which plaintiffs were injured and how much in damages they should receive does not negate the significant common and predominant legal and factual questions that will have been resolved previously. The motions for class certification are GRANTED.

BACKGROUND

I. LIDODERM PATENT LITIGATION AND SETTLEMENT

This case concerns the alleged antitrust and anticompetitive impact of a July 2012 settlement agreement ("Agreement") between defendants Endo Pharmaceuticals Inc. ("Endo"), Teikoku Seiyaku Co., Teikoku Pharma USA (collectively "Teikoku") and Watson Pharmaceuticals, Inc. (all collectively, "defendants"). The Agreement terminated ongoing patent litigation alleging invalidity of Teikoku's patents covering Lidoderm patches in exchange for giving brand-name Lidoderm patches to Watson, as well as a period of exclusivity to market its generic version of lidocaine patches without competition from Endo's generic patch.

Watson became part of Actavis, Inc., which became associated with Allergan, plc, and Watson then became part of Teva Pharmaceuticals Industries, Ltd. The defendant will be referred to here as Watson.

Lidoderm is the brand name for lidocaine 5% patches that are used to relieve the pain of post-herpetic neuralgia (also known as "after-shingles" pain). Declaration of Jeffrey J. Leitzinger (Dkt. No. 522-1) ¶ 18.

At the time of the patent litigation, Endo had an exclusive license from Teikoku to manufacture and sell the Lidoderm patches and subsequently manufactured and sold the authorized generic version of Lidoderm as well.

The primary terms of the Agreement were these. First, Watson agreed to delay launching its generic Lidoderm until September 15, 2013, about a year after the Food and Drug Administration's ("FDA") 30-month stay on Watson's Abbreviated New Drug Application ("ANDA") expired, a year before one of Teikoku's patents covering Lidoderm was to expire and two years before another of Teikoku's patents was due to expire. Second, Endo/Teikoku agreed to drop the pending patent infringement lawsuits and not further amend their Citizen Petition ("CP") pending with the FDA that asked the FDA to not approve ANDAs for Lidoderm unless they met more stringent scientific standards. Third, Endo/Teikoku agreed to give Watson $96 million worth of brand name Lidoderm patches to distribute or sell, on the condition that Watson honor Endo/Teikoku's existing price-related contracts. Fourth, Endo/Teikoku agreed not to release their authorized generic ("AG") lidocaine patch until seven and one half months after Watson began selling its generic version; during this "exclusivity period," Watson agreed to pay Endo/Teikoku a twenty-five percent royalty on the Gross Profit for sales of its generic.

A more detailed explanation of the factual background of this case, including the regulatory obligations of the FDA and patent litigation under the Hatch-Waxman Act (21 U.S.C. § 355(a)), is explained in my prior Orders, including Dkt. No. 117.

Plaintiffs filed suit, arguing that defendants' Agreement violated federal antitrust and related state laws. They assert that it harmed consumers because, absent the Agreement, Watson would have entered the market substantially before September 2013 with a generic lidocaine patch that was much less expensive than brand Lidoderm, that the Agreement allowed Endo/Teikoku to charge supracompetitive prices for their brand drug for longer and that the period of exclusivity allowed Watson to charger higher prices for its generic.

II. CLASSES SOUGHT TO BE CERTIFIED

The DPPs are pharmaceutical wholesalers, pharmacies, hospitals, and retail stores that purchased brand and generic Lidoderm patches directly from defendants and supplied the product to others. Expert Report of Gregory K. Leonard, Ph.D. (Dkt. No. 563-2) ¶¶ 8, 25 (characterizing the DPPs as falling into five different classes of trade - national wholesalers, regional wholesalers, mail order wholesalers, hospitals, and retail pharmacies). The DPPs seek to certify the following class:

The three named DPP plaintiffs are Drogueria Betances ("Betances"), Rochester Drug Co-operative, Inc. ("Rochester"), and American Sales Company, LLC ("ASC").

All persons or entities in the United States, including its territories, possessions, and the Commonwealth of Puerto Rico, who purchased brand or generic Lidoderm directly from any of the Defendants at any time during the period August 23, 2012 through May 1, 2014 (the "Class").

DPP Mot. 2-3. The end date of May 1, 2014, was chosen because that is the day before Endo launched its own authorized generic Lidoderm and it provides a "clear" cut-off date for class membership. Excluded from the class are defendants (and their officers, directors, management, employees, subsidiaries, and affiliates) and all federal government entities. DPP Mot. 3.

Under that definition, there are 55 DPP Class Members who are widely geographically dispersed across the United States. Leitzinger Decl., Exs. 4, 5. The DPP entities purchase drugs directly from the brand or generic (when available) manufacturers and provide them to hospitals, pharmacies, and retailers or re-sell the drugs to other wholesalers. Leonard Rep. ¶ 25.

The EPPs acknowledge that some wholesaler Class Members may have assigned some or all of their entitlement to overcharges to Retailer Plaintiffs. Leitzinger Decl. ¶ 19 n.57. The EPPs' expert contends he can "readily adjust Class volumes to reflect the assigned volumes." Id. I do not in this Order determine whether Retailer Plaintiffs can opt-out or, if allowed to opt-out, whether those separate actions will proceed in conjunction with the certified classes. See Dkt. No. 231 at 10-12.

The EPPs are employee health and welfare benefit plans, municipal corporations, employee unions, and individuals who purchased brand or generic Lidoderm from third parties, not from defendants directly. They seek to certify the following class:

The named EPPs are: (i) Allied Services Division Welfare Fund, (ii) City of Providence, (iii) International Union of Operating Engineers Local 49 Health and Welfare Fund, (iv) International Union of Operating Engineers Local 132 Health and Welfare Fund, (v) Iron Workers District Council of New England Welfare Fund, (vi) NECA-IBEW Welfare Trust Fund, (vii) United Food and Commercial Workers Local 1776 & Participating Employers Health and Welfare Fund, (viii) Welfare Plan of the International Union of Operating Engineers Locals 137, 137A, 137B, 137C, 137R, (ix) Letizia Gallotto, and (x) Steven Roller. EPP Mot. 1.

(a) All persons and entities in the United States and its territories who, in Arizona, California, Florida, Kansas, Maine, Massachusetts, Minnesota, Nevada, New Hampshire, New Mexico, New York, North Carolina, North Dakota, South Dakota, Tennessee, West Virginia, or Wisconsin ("Class States") for consumption by themselves or their family member, or by their insureds, plan participants or beneficiaries, paid and/or provided reimbursements for some or all of the purchase price of:

i. Branded Lidoderm for the time period August 23, 2012 through September 14, 2013; and/or
ii. AB-rated generic Lidoderm for the time period September 15, 2013 through August 1, 2014;

- and -

(b) Third-party payors CVS Caremark, Cigna, Envision Pharmaceutical Services, MedImpact Healthcare Systems, Inc., Comprehensive Health Management, Inc. Part D, and Express Scripts Senior Care to the extent they provided, under their Medicare Part D plans, reimbursements for some or all of the price of branded Lidoderm purchased in Class States for the time period September 15, 2013 through August 1, 2014.

Excluded from the Class are:

(a) Defendants and their officers, directors, management, employees, subsidiaries, and affiliates;
(b) Those who, after September 15, 2013, paid and/or provided reimbursements for branded Lidoderm and did not purchase or reimburse for generic Lidoderm, except third-party payors CVS Caremark, Cigna, Envision Pharmaceutical Services, MedImpact Healthcare Systems, Inc., Comprehensive Health Management, Inc. Part D, or Express Scripts Senior Care for their Part D insurance.
(c) Government entities, other than government-funded employee benefit plans;
(d) Fully insured health plans (i.e., plans that purchased insurance that covered 100 percent of the plan's reimbursement obligations to all of its members);

(e) "Single flat co-pay" consumers who purchased Lidoderm or generic Lidoderm only via a fixed dollar co-payment that does not vary on the basis of the purchased drug's status as branded or generic (e.g., $20 for both branded and generic drugs);
(f) "Flat generic co-pay" consumers who, after September 15, 2013, purchased generic Lidoderm via a fixed dollar co-payment (e.g. $10 for generic drugs) regardless of the co-payment applicable to branded drugs;
(g) Consumers who purchased or received Lidoderm or its AB-rated generic equivalent through a Medicaid program only;
(h) Pharmacy benefit managers; and
(i) The judges in this case and members of their immediate families.

EPP Mot. 1-2.

III. PHARMACEUTICAL DISTRIBUTION CHAIN

The class certification motions, and defendants' oppositions to them, turn in part on the roles the DPPs and other entities play in the pharmaceutical distribution chain. Neither side disputes that pharmaceutical manufacturers sell their drugs directly to wholesalers and other DPPs (hospitals, pharmacies, and retailers), typically at the wholesale acquisition cost ("WAC") minus discounts. EPP Mot. 11-12. The wholesalers, as their name implies, resell the drugs to pharmacies which sell the drugs to consumers. Id. 12. The WAC - as the first price in the chain - acts as a benchmark for all subsequent sales and purchases. Id. The EPPs assert - and defendants' experts do not challenge - that the higher the WAC the more the DPPs and the EPPs pay for the drugs. Id.

The other EPPs, including retailers, hospitals, and pharmacies, sell or provide the products to end consumers and (generally) not to entities who intend to pass them along to end users. "Consumers" or "end-users" as used in this Order refer to the individual end-users who the drugs were prescribed to.

End consumers do not typically pay the entire purchase price of the drug. Many consumers are covered by health insurance plans provided by third party payors ("TPPs"). The TPPs and the consumers are, unless they fall within the exclusions discussed above, both EPPs. That is because consumers often pay a portion of the purchase price at the point of purchase and the remainder is paid by the TPP. Expert Report of Hal J. Singer Ph.D. [Dkt. No. 524-1] ¶¶ 86, 127. The consumers and TPPs together pay the full price and neither passes the amount they paid onto the other. EPP Mot. 12; Singer Rep. ¶ 127.

These TPPs are insurers, health and welfare plans, plan sponsors, and self-insured employers. DeBree Decl. ¶ 3. As described somewhat simplistically by the EPPs, Lidoderm is offered in three end-payor markets: to cash payors, to commercial insurers, and to Part D Medicare insurers. EPP Mot. 9.

There are typically two types of payment arrangements for consumers with insurance: percentage of purchase price, where the consumer pays a set percentage of the price ("coinsurance"); and fixed copay, where the consumer pays a set amount for each drug purchased ("copay"). Expert Report of Robert Navarro [Docket No. 550-35] ¶¶ 16d, 28. How much a particular insured consumer will pay for a drug is governed by the specific design of their insurance coverage, including various annual deductible amounts as well as benefit or out of pocket maximums. Expert Report of James W. Hughes Ph.D. [Dkt. No. 550-34] ¶¶ 36, 45; Navarro Rep. ¶ 24. The amount of a copay or coinsurance payment may also depend on the particular drug's "formulary and tier structure" which often, but not always, results in higher co-payments for brand drugs than generic ones. Navarro Rep. ¶¶ 16g, 24c. Some plans allow consumers to choose brand name drugs over less expensive generics. Hughes Rep. ¶¶ 85-89 (describing behavior of "brand loyalists" who continue to purchase brand by choice, even after generics enter the market). Where a consumer does not have insurance coverage or a particular drug is not covered by the consumer's insurance, the consumer will pay 100% of the cost.

Defendants' expert Navarro states that in studies conducted in 2013-2014, health plan copayments for brand name drugs ranged from $7 to $300 and deductibles ranged from $500 to $2000. Navarro Rep. ¶ 26.

The EPPs assert that in no-insurance situations, pharmacies have records of consumers' prescription drug purchases. Declaration of W. Paul DeBree [Dkt. No. 524-2] ¶ 24.

Generally two entities are involved in providing health care coverage to consumers: plan sponsors (employers or self-funded employee health and welfare plans) and health plans (typically commercial insurers who the plan sponsors contract with). The extent to which the plan sponsor bears some portion of the cost of providing drugs to its participants varies. Plan sponsors contract with a health plan and/or a pharmacy benefits manager ("PBM") to administer prescription drug plans. Some plan sponsors contract with health plans or PBMs for "administrative services only" ("ASO") and there, the plan sponsor bears the full cost of claims for prescription drugs. Navarro Rep. ¶ 24h; DeBree Decl. ¶ 36. Under a "fully insured" contract, a plan sponsor pays premiums to a health plan and the health plan pays the prescription drug costs. Narvarro Rep. ¶ 24h. Plan sponsors may also contract to share with the health plans the costs of prescription drugs. Id. ¶ 16a. Plan sponsors who pay for prescriptions are TPPs and included in the EPP class. Health plans (typically commercial insurers) collect premiums from consumers and/or plan sponsors designed to cover the provision of prescription drugs (and health care more generally). Id. Health plans are TPPs and are included in the EPP class definition.

Many TPPs use a PBM to administer prescription drug benefits. Hughes Rep. ¶ 23; DeBree Decl. ¶ 3. PBMs who were involved in the lidocaine transactions covered by this case include Prime Therapeutics, OptumRX, Caremark, and Express Scripts. PBMs pay the pharmacies who provide the drugs to the end-consumers and then collect an agreed-to payment from the plan sponsor or health plan. Navarro Rep. ¶ 21i. Both sides agree that PBMs use "spread pricing" and "rebates" as part their business operations.

PBMs typically negotiate prices for drugs directly with retail pharmacies and earn profits on the "spread" between the prices they pay the pharmacies and the price they charge their TPP customers. Navarro Rep. ¶¶ 15, 24i. Defendants allege that PBMs may, if their estimates and negotiations are not on target and there is no "spread" on a particular transaction, end up bearing the cost of the drug transaction (e.g., they end up charging their customers less than what they paid the pharmacies for a particular prescription). Navarro Rep. ¶¶ 15, 24i, 75-78. Defendants do not, however, identify any instances of this happening with respect to the 5% lidocaine patches (brand or generic) at issue, much less how much of a real risk it represents to PBMs. Plaintiffs cite evidence from PBMs that they "do not suffer losses" on their contracts with TPPs to argue that the PBMs bear no real risk of harm from their spread pricing practices. DeBree Decl. ¶¶ 39-42.

PBMs (and some health plans) also negotiate rebates from drug manufacturers for formulary placement (e.g., the PBM's ability to offer preferential formulary placement to drug manufacturers) and other concessions favorable to manufacturers. Navarro Report ¶ 16g; Hughes Rep. ¶ 44; DeBree Decl. ¶ 38. Those rebates are apportioned between the PBMs and health plans or plan sponsors depending on how their contracts are structured, and can be a percentage of the rebate received by the PBM from the drug manufacturer or a fixed dollar rebate per prescription. In any event, the rates of rebates passed through to the health plan and plan sponsors differ widely and depend upon the specific deal agreed to by the PBM and individual plan or sponsor. Navarro Rep. ¶¶ 16f, 16g, 32-36. Some PBMs may "guarantee" rebates to plan sponsors at a fixed amount.

Drug formularies are the list of approved drugs accepted by the various TPPs, and created and administrated through the PBMs. DeBree Decl. ¶ 36; Singer Decl. ¶ 105; Hughes Decl. ¶¶ 11, 16, 20. They can be open or closed, with open formularies allowing access to more drugs. The formularies also have various tiers, typically 3 or 4 but up to 7, with Tier One typically including generic drugs and having the lowest amount of copay or coinsurance. For example, named EPP plaintiff Iron Workers District Council of New England Welfare Fund's plan had three formulary tiers where the copay was $15 for generic drugs (Tier 1), $30 for preferred brand drugs (Tier 2), and $45 for non-preferred brand drugs (Tier 3). Navarro Rep. ¶ 27. If the member used a mail order pharmacy, the copays rose to $30/$60/$90 and if the member went to an out-of-network pharmacy, they were required to pay in advance and then apply for reimbursement through the Fund's PBM. Id.

Defendants assert that, as with spread pricing, if PBM's estimates and negotiations are not on target (e.g., they could not secure as big a rebate from a manufacturer as they thought they could), the PBMs may end up taking a loss on particular drug sales if the rebates from the manufacturers are not as large as they estimated. Navarro Rep. ¶¶ 24f, 32-34, 75-76. The EPPs' expert DeBree agrees that it is possible, although "exceptionally unlikely," that a PBM could enter into an unprofitable contract because it promised too high a contractually guaranteed rebate to a TPP, a situation he has never seen happen. DeBree Decl. ¶ 39; see also Declaration of Brian Hansen (Dkt. No. 524-5) ¶ 7 (PBM Prime Therapeutics has not had to "perform" on a guarantees rebate since 2012); see also Sharp Supp. Decl., Ex. K (Response No. 13) ("OptumRx does not . . . incur losses on guaranteed rebates"). DeBree explains that as additional "insulation" for PBMs, their contracts with TPPs provide that the PBM can "equitably adjust" rates, administrative fees and rebates if unforeseen contingencies (like the availability of a generic) occur. DeBree Decl. ¶¶ 41, 53 (asserting that a PBM's failure to meet contractual guaranteed rebates promised to a TPP are paid for out of the PBM's contract expenses and not related to consumers' purchase of a specific drug like a 5% lidocaine patch).

As with spread pricing, defendants identify a general, theoretical risk without substantiating the true impact (if any) of that risk. Defendants' experts (Navarro and Hughes) each identify one example of a purported disparity between the rebate the PBM received from the manufacturer (none, according to defendants) and the contractual rebate amount promised by the PBM to the plan sponsor. Navarro Rep. ¶ 79; see also Hughes Rep. ¶ 26. Singer responds that this particular TPP may have been receiving rebates through a different PBM during the relevant timeframe. Singer Reply Decl. ¶¶ 55-56. PBMs are excluded from the EPP class, yet defendants argue that the EPPs damages model is overinclusive because it has not "removed" these theoretical PBM damages that resulted from the PBMs' theoretical failures to negotiate and accurately estimate their spread and rebates.

Another problem raised by defendants is that there is no reliable method to exclude the government's Part D damages from plaintiffs' aggregate damages calculations. Medicare Part D is the federal government's prescription drug benefit under Medicare ("Part D"). Part D contracts with private insurers to provide drug plans to participants. Navarro Rep. ¶ 72. According to the EPPs' expert Singer, 42% of lidocaine 5% patches at issue were procured through Part D. Singer Rep., Table 4.2. The government subsidizes premiums for Part D plans and provides the cost of Part D prescriptions for certain low income participants and participants who exceed their out of pocket threshold. Hughes Rep. ¶¶ 25, 121; Navarro Rep. ¶ 73. The EPP class definition includes Medicare Part D plans (the private insurance companies), but excludes government entities from membership.

A final issue, particularly relevant to the DPP's motion for certification, involves Group Purchasing Organizations ("GPOs"), which are membership organizations who negotiate on behalf of EPPs with manufacturers to secure contract prices on behalf of the EPP members. Leonard Rep. ¶ 15. Four GPOs - OptiSource Premier, Topco, Pharmacy Value Alliance, and Econdisc Contracting Solutions - negotiated prices on behalf of 29 of the EPPs. Id. ¶¶ 17-22. The GPOs, however, do not purchase any drugs and are not involved in payments between EPPs and manufacturers.

LEGAL STANDARD

Class actions are governed by Rule 23 of the Federal Rules of Civil Procedure. Plaintiffs bear the burden of showing that they have met each of the four requirements of Rule 23(a) and at least one subsection of Rule 23(b). Berger v. Home Depot USA, Inc., 741 F.3d 1061, 1067 (9th Cir. 2014) (citing Zinser v. Accufix Research Inst., Inc., 253 F.3d 1180, 1186 (9th Cir. 2001)). The plaintiff "must actually prove - not simply plead - that their proposed class satisfies each requirement of Rule 23, including (if applicable) the predominance requirement of Rule 23(b)(3)." Halliburton Co. v. Erica P. John Fund, Inc., 134 S.Ct. 2398, 2412 (2014) (citing Comcast Corp v. Behrend, 133 S.Ct. 1426, 1431-32 (2013); Wal-Mart Stores, Inc. v. Dukes, 131 S.Ct. 2541, 2551-52 (2011)).

The court's "class certification analysis must be rigorous and may entail some overlap with the merits of the plaintiff's underlying claim." Amgen Inc. v. Connecticut Retirement Plans and Trust Funds, 133 S.Ct. 1184, 1194 (2013) (quoting Dukes, 131 S.Ct. at 2551 (internal quotation marks omitted)). These analytical principles govern both Rule 23(a) and 23(b). Behrend, 133 S.Ct. at 1342. However, "Rule 23 grants courts no license to engage in free-ranging merits inquiries at the certification stage." Amgen, 133 S.Ct. at 1194-95. "Merits questions may be considered to the extent - but only to the extent - that they are relevant to determining whether Rule 23 prerequisites for class certification are satisfied." Id.

As the Ninth Circuit clarified in Ellis v. Costco Wholesale Corp., 657 F.3d 970, 982 (9th Cir. 2011), simply because an expert opinion clears the "scientifically reliable and relevant" hurdle of Daubert does not mean it passes the "rigorous analysis" required by Rule 23 to support class certification. At class certification, a court must determine whether the expert's evidence supporting certification is persuasive following a rigorous analysis of the same. Id. at 983-84. As part of that rigorous analysis, a court may be required to resolve factual disputes between the plaintiffs' and defendants' experts if those disputes go to whether or not the injury at issue can be shown on a classwide basis. Id.

Under Rule 23(a), the class may be certified only if: (1) the class is so numerous that joinder of all members is impracticable, (2) questions of law or fact exist that are common to the class, (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class, and (4) the representative parties will fairly and adequately protect the interests of the class. See Fed. R. Civ. P. 23(a). A plaintiff must also establish that one or more of the grounds for maintaining the suit are met under Rule 23(b): (1) that there is a risk of substantial prejudice from separate actions; (2) that declaratory or injunctive relief benefitting the class as a whole would be appropriate; or (3) that common questions of law or fact predominate and the class action is superior to other available methods of adjudication. See Fed. R. Civ. P. 23(b).

DISCUSSION

I. DIRECT PURCHASER PLAINTIFFS' MOTION FOR CLASS CERTIFICATION

The DPPs rely primarily on the declaration of their expert, Dr. Leitzinger, to support class certification. Leitzinger declares that common proof shows that the alleged suppression of generic competition by defendants resulted in classwide antitrust injury in the form of overcharges. In doing so, Leitzinger relies on: (i) economic and governmental studies on the market-wide effects of generic competition and delayed generic entry; (ii) defendants' own documents analyzing the projected and actual market-wide effects of generic entry and delayed generic entry; (iii) the actual pricing and sales experience of lidocaine 5% patches once generic patches finally entered the market; and (iv) the direct purchasers' role in the distribution chain. Leitzinger Decl. ¶ 22; see also ¶¶ 23-36. DPPs also rely on a "Trial Plan" (Dkt. No. 523-7), where they identify the common questions they intend to address as:

a) Whether the alleged reverse payment settlement and license agreement (the Agreement) violates the antitrust rule of reason;

b) Whether, by and through the Agreement, defendants conspired to suppress generic competition to Lidoderm;

c) Whether, pursuant to the Agreement, Watson agreed to, and did, delay its entry into the market with generic Lidoderm in exchange for large reverse payments from Endo and/or Teikoku;

d) Whether, pursuant to the Agreement, Endo and Teikoku made large reverse payments to Watson, and the magnitude of each such payment;

e) Whether the reverse payments suppressed generic competition to Lidoderm by delaying Watson's generic launch and Endo and Teikoku's authorized generic launch, and thereby

preventing prices for lidocaine patch 5% from falling;

f) When, absent the reverse payments, Watson would have launched its generic version of Lidoderm and Endo would have launched authorized generic Lidoderm;

g) Whether the reverse payments Endo and Teikoku made to Watson are explained by purposes other than delaying Watson's entry into the lidocaine patch 5% market, and, if so, what those explanations are;

h) Whether Endo and Teikoku's reverse payments to Watson were for a procompetitive purpose, and, if so, whether a reverse payment was reasonably necessary to achieve (and/or the least restrictive means of achieving) that procompetitive purpose;

i) Whether, on balance, the reverse payments harmed competition in the lidocaine patch 5% market;

j) Whether, by the reverse payments, defendants conspired or attempted to maintain Endo's market and/or monopoly power in the lidocaine patch 5% market;

k) Whether Endo had market or monopoly power in the lidocaine patch 5% market;

l) To the extent a relevant market or markets must be defined, what that definition is or those definitions are;

m) Whether the activities of Defendants substantially affected interstate commerce;

n) Whether, and to what extent, the challenged conduct caused antitrust injury to the business or property of Plaintiffs and the Class in the nature of overcharges; and

o) The quantum of overcharges paid by the Class in the aggregate.

DPPs' Trial Plan at 2-3. DPPs contend that common evidence to answer these questions includes testimony by defense witnesses and testifying experts, as well as internal documents from defendants, all of which will be evidence common to the Class as a whole. Id. at 3.

With respect to damages, as explained and supported by Leitzinger's declaration, the DPPs intend to establish damages in the aggregate using classwide evidence and to quantify the aggregate overcharge damages using a methodology that utilizes a "before and after" benchmark for generic Lidoderm prices based on actual generic rates and "backcasted" to calculate what the Class's expenditures would have been if generic Lidoderm entry had occurred earlier. Using that benchmark, EPP damages will be calculated by modeling the extent of generic substitution and the prices of generic and branded Lidoderm that would have occurred earlier but-for the paid-for delay in generic competition. These estimates will then be subtracted from the known quantities of Lidoderm actually purchased at known prices during the relevant time period to arrive at aggregate damages.

Plaintiffs intend to use the following types of common-to-the-class evidence:

a) Transactional data from Endo, Watson, and Endo's authorized generic seller Qualitest, showing unit and dollar sales, pricing, discounts, rebates, chargebacks, administrative fees, and other unit and/or dollar adjustments, for branded and generic Lidoderm;

b) Defendants' internal generic penetration models and forecasts, and the forecasts of generic manufacturers;

c) The extensive body of economic literature and empirical evidence regarding the effects of generic competition; and

d) Expert analysis and opinion.

DPP Trial Plan at 3-4. Under Leitzinger's preliminary analysis - using the backcast method described above and calculating a "Delay Period" as the time between March 31, 2013 and September 15, 2013, as well a longer "Overcharge Period" - the DPPs estimate that the Class suffered $295 million in overcharges due to the delayed generic entry. Leitzinger Decl. ¶ 45.

The Delay Period used by Leitzinger starts on the assumed March 31, 2013 entry date (but-for defendants' Agreement) and ends on the actual Watson entry date of September 15, 2013. The Overcharge Period is the period of time beginning on March 31, 2013 but extending until the point in time where according to Leitzinger the generic entry produced the full savings associated with generic competition. Leitzinger Decl. ¶¶ 38, 39.

Defendants challenge the DPPs' showing on the following grounds: (i) common questions of fact do not predominate and a class proceeding is not superior because given the differences between market position, purchasing power, and actual purchasing history of the DPPs, injury and damage questions cannot be resolved on a classwide basis; (ii) the number of DPPs in the class is small, relatively few DPPs control the vast majority of the market, and joinder is not impracticable under Rule 23(a)(1), so the DPP class fails the numerosity requirement; and (iii) because of the nature of the market and the different roles and contracts negotiated by the DPPs, conflicts between the DPPs preclude a finding of representativeness under Rule 23(a)(4).

Defendants do not challenge DPPs' motion as to the adequacy of interim class counsel. On May 20, 2014, I appointed Faruqi & Faruqi LLP, Garwin, Gerstein & Fisher LLP, and Hagens Berman Sobol Shapiro LLP, as Interim Co-Lead Counsel for the proposed Direct Purchaser Class, and Hagens Berman Sobol Shapiro LLP as Interim Liaison Counsel for the proposed DPP class, based on a showing of experience and adequacy. Dkt. No. 60. Those firms have ably and vigorously litigated this case, and nothing has occurred to undermine my initial determination of their experience and adequacy.

A. Predominance of Common Questions and Superiority

Defendants do not dispute that significant and numerous questions of law and fact identified by DPPs as to defendants' liability for anticompetitive conduct can be shown by common evidence. See DPP Trial Plan at 2-3. Instead, they argue that the DPPs cannot rely on their expert's model to establish classwide injury because it is flawed and cannot be used reliably to prove classwide damages from the alleged delay of generic introduction and inflated generic prices upon the Watson generic entry. Defendants obscure that the DPPs rely on Leitzinger's model to show a methodology of determining classwide aggregate damages, not to show classwide injury.

As to injury, neither the defendants nor their experts adequately address the academic and industry studies relied on by Leitzinger to support a showing of classwide impact. Those sources explain that, generally, the introduction of generic drugs creates significant cost savings for consumers at most levels of the distribution chain. Rebuttal Declaration of Jeffrey J. Leitzinger [Dkt. No. 591-1] ¶ 8; Leitzinger Decl. ¶¶ 22-27. Defendants simply dismiss those studies because they do not discuss what happened in this case. Oppo. to DPP Mot. 9. Defendants likewise do not persuasively rebut Leitzinger's reliance on defendants' own internal forecasts about the impact of generic entry in the Lidoderm market, predicting lower costs for both the brand after the generic entered the market, and for the generics, once more than one generic entered the market. Leitzinger Decl. ¶¶ 28-31. At most, defendants emphasize those forecasts do not exactly match "what actually happened." Oppo. to DPP Mot. 9.

Given the well-researched market at issue and the well-recognized type of antitrust injury alleged, this evidence is persuasive and supports the DPPs' argument that injury in this case can be and will be shown on a classwide basis. See, e.g., In re High-Tech Employee Antitrust Litig., 985 F. Supp. 2d 1167, 1215 (N.D. Cal. 2013) (relying on defendants' internal documents and economic literature); In re Dynamic Random Access Memory (DRAM) Antitrust Litig., No. M 02-1486 PJH, 2006 WL 1530166, at *9 (N.D. Cal. June 5, 2006) (relying on actual publication, market, and sales data).

In addition, defendants' expert testified that he could not recall and could not identify any instance where DPPs paid less for generic Lidoderm than for brand after generic entry and paid less for generic Lidoderm after Endo's AG entry. Declaration of Peter Kohn ISO DPP Reply [Dkt. No. 592], Ex. 52, Transcript of October 13, 2016 Deposition of Gregory Leonard at 91-93, 96, 108, 290.

Impermissible Aggregation. Defendants' main focus is to attack Leitzinger's aggregate damages model. Defendants argue that the model is unreliable because it fails to consider the "actual experience" of particular DPPs since it is based on aggregated purchases - combining brand only, generic and brand, and then generic only purchases to create aggregated purchasing figures - and then estimates damages flowing from the aggregated purchases based on Leitzinger's estimated "but-for" price (as opposed to actual prices). Oppo. to DPP Mot. 10-11. However, given the well-established academic and industry-accepted evidence of the swift and significant (in volume) switch to generic drugs mandated by state laws and the economic realities upon generic entry, that Leitzinger takes an aggregate approach to damages is not problematic here.

The cases defendants rely on that disapprove the use of averaged or aggregate approaches addressed materially different antitrust theories in materially different markets. See, e.g., In re Optical Disk Drive Antitrust Litig., 303 F.R.D. 311, 321 (N.D. Cal. 2014) (price-fixing conspiracy to prevent trending decline in prices); Food Lion, LLC v. Dean Foods Co., 312 F.R.D. 472, 489 (E.D. Tenn. 2016) (rejecting averaging approach to determining injury from alleged conspiracy to inflate prices of milk where model assumed price impact across areas where no impact was found); see also In re Celexa & Lexapro Mktg. & Sales Practices Litig., 315 F.R.D. 116, 128 (D. Mass. 2016) (rejecting plaintiffs' attempt to rely on aggregate statistical evidence to prove "but for causation" in an off-label promotion case because of flaws in the statistical analysis).

Defendants point out that the DPP class encompasses different types of entities (pharmacies vs. wholesalers) and that the economic circumstances and incentives vary between those different types of entities to argue that individual analyses of damages is necessary. Leonard Rep. ¶ 25. It is true that the different DPPs ultimately paid different prices for their brand and generic lidocaine patches because of their different sizes, purchase histories, and negotiating strength. But simply because they were injured in different amounts does not undermine the fact they were injured.

Contrary to defendants' assertion, even though the DPPs may have incurred significantly different amounts of damages, damages issues will not overwhelm the common liability questions. As a general matter, differences in damages will rarely suffice to defeat class certification. See, e.g., Vaquero v. Ashley Furniture Indus., Inc., 824 F.3d 1150, 1155 (9th Cir. 2016) ("We have repeatedly confirmed . . . that the need for individualized findings as to the amount of damages does not defeat class certification."). Under the DPPs' trial plan, aggregate damages can be determined using Leitzinger's backcasted model and then damages can be apportioned between the DPPs using on a pro rata formula based on each DPP's purchase of brand or generic Lidoderm. Cf. Briseno v. ConAgra Foods, Inc., 844 F.3d 1121, 1131 (9th Cir. 2017) (discussing the common techniques used for individualized claim determinations after a classwide finding of liability).

Uninjured DPPs. Defendants argue the unreliability of Leitzinger's model is demonstrated by reviewing the actual purchasing data that shows three of the putative DPP class members were not harmed. Two (Cesar Castillo, Inc. and DMS Pharmaceutical) were uninjured, according to defendants, because they purchased only branded Lidoderm post-generic release at higher prices than pre-release. The third, Drogueria Central, was uninjured because it did not purchase any Lidoderm after the generic entry date. Def. Oppo. to DPP Mot. 11.

As to these allegedly "false positives," plaintiffs argue that it is premature to exclude them from the class because they may have purchased generic product from other wholesalers or distributors, even if they did not purchase generics from defendants once they were available. Leitzinger Reb. Decl. ¶ 20. And simply because these three did not purchase any generic Lidoderm after generic entry (or for Drogueria Central, did not purchase Lidoderm after February 2013), plaintiffs point out that each of them may well have purchased generics if they had been available before the actual entry date, something which can be determined at the damages stage. Id.

Moreover, even if defendants could definitively show at this juncture that there are DPPs who were not harmed yet are included within the DPP class definition - and I do not find defendants have made that showing here - such overinclusiveness would not defeat class certification as long as the uninjured parties represent a de minimis portion of the class. See Torres v. Mercer Canyons, Inc., 835 F.3d 1125, 1136-37 (9th Cir. 2016) (presence of uninjured class members in class did not preclude predominance finding); In re: Lenovo Adware Litigation, No. 15-md-02624-RMW, 2016 WL 6277245, *15 (N.D. Cal. Oct. 27, 2016) (certifying a class of computer purchasers over defendants' objections that some class members were uninjured); Bernstein v. Virgin America, Inc., No. 15-cv-02277-JST, 2016 WL 6576621, *13 (N.D. Cal. Nov. 7, 2016) (certifying off-the-clock claims even though some class members may not have worked off-the-clock); see also In re Nexium Antitrust Litig 777 F.3d 9, 30-31 (1st Cir. 2015) (affirming certification in reverse-payment antitrust class action where de minimis number of uninjured end-payor plaintiffs included in the class, and defining de minimis as a "number of uninjured members . . . so large as to render the class impractical or improper, or to cause non-common issues to predominate."); In re Cathode Ray Tube (CRT) Antitrust Litig., 308 F.R.D. 606, 615 (N.D. Cal. 2015) ("Even if some individuals are thus able to join the class and then are later determined to not have valid claims against a proper defendant, this does not preclude class certification.").

Defendants do not dispute that there was impact to 52 DPPs and only challenge the injury as to three others. Even if these three are not properly included in the class, their inclusion at most has a de minimis impact and does not preclude certification.

Wrong Inputs. Defendants' expert Leonard attempts to show the weaknesses in Leitzinger's model by altering it in two respects. First, Leonard uses Leitzinger's estimated Aggregate Purchaser generic price but applies it to the actual DPP purchasing history after generic entry (as opposed to the but-for estimated purchases used by Leitzinger) and backcasts that to the Delay Period. Under that analysis, the aggregate damages would have been $49 million less; $245 million as opposed to $294 million. Leonard Rep. ¶ 51. Second, Leonard takes his analysis a step further and recalculates the overcharge per unit for branded and generic Lidoderm using the real world prices applied after actual generic entry and then aggregates the damages across the proposed class members' actual purchase history and ends up with an aggregate damage total of $218 million, or $76 million less than Leitzinger's. Id. ¶ 53.

These alterations, of course, do not show that some or any significant portion of the DPP class members suffered no injury or no damages. Instead, Leonard argues that the wide differences in aggregate damages resulting when Leonard's inputs are used in Leitzinger's model bolster his general argument that "internal inconsistencies" pervade Leitzinger's model, undermining its reliability. What the generic conversion rate would have been "but for" defendants' conduct is a matter of dispute between the experts. Same too for what the prices of brand and generic lidocaine patches would have been but for defendants' conduct. Those disputes are not appropriately resolved at this juncture; that the experts dispute what the appropriate inputs should be does not undermine the approach or the reliability of Leitzinger's model.

Defendants make additional arguments about Leitzinger's "incorrect factual assumptions," such as challenging Leitzinger's assumption on the date generic entry would have occurred but-for the antitrust conduct, and also argue that Leitzinger failed to address the effect of Watson's subsidiary Anda's sales of the free brand product as well as the effect of assignments in his damages model. See, e.g., Oppo. to DPP Mot. 13-19; Leonard Rep. ¶¶ 62, 65, 67, 73-75. Defendants assert that these issues require impact and damages to be assessed individually, undermining the commonality and predominance assertions of plaintiffs. I disagree. If further analysis or refinement of who is in the class and what purchases are relevant to the aggregate damages determination is necessary - because it becomes clear that some proposed DPP class members were not injured or they decide opt out - that can be readily managed. See, e.g., Leitzinger Decl. ¶ 19 n.57 (model can be "readily adjusted" to account for decreased volumes due to opt-outs); ¶ 47 (model can be adjusted to account for "bypass"); ¶ 49 (model can be adjusted for different generic entry dates).

At base, the criticisms of Leitzinger's model challenge the amount of damages suffered by the class (depending upon how the facts are determined), but those criticisms do not undermine Leitzinger's methodology to show aggregate damages or his conclusion that the vast bulk of class members were injured by overcharges. With respect to the amount of aggregate damages, Leonard's criticisms can be accommodated by Leitzinger's model depending on how facts are further developed, what questions are resolved on summary judgment, and the findings of the trier of fact (e.g., absent the agreement (i) when Watson would have entered the market, (ii) whether Watson would have been able to meet supply or when that ability would have been achieved, (iii) when Endo would have entered the market with its AG, and (iv) whether Endo would have charged higher prices for its branded drug if it had entered the market with an AG at same time as Watson).

For example, with respect to the generic entry date issue, defendants argue individualized inquiry is necessary because the date of entry - and how much generic Watson had on hand at the various possible earlier entry dates - impacts whether DPPs would be able to buy sufficient generic stock to fulfill their needs; if the supply from Watson was limited, price might well have increased or led to "rationing" by Watson. Def. DPP Oppo. at 14-15. The DPPs respond that Leonard's testimony and Watson's own documents support a factual finding that supply and rationing would not have been an issue. Leitzinger Reb. Decl. ¶¶ 28 - 30; DPP Reply at 6-9. With respect to Anda sales - Watson's subsidiary selling brand product it received from Endo during the Delay Period at a discount below other distributors - and defendants' argument that Leitzinger's model fails to account for those sales, the DPPs respond that: (i) the Anda sales occurred because of the settlement (and so are not relevant to a but-for world); (ii) the discounts offered by Anda were smaller than the discounts offered by Endo during the same period, so that the class did not receive greater price concessions; and (iii) those Anda sales were excluded in any event from Leitzinger's damages model. Def. DPP. Oppo. at 17; DPP Reply at 9; Leitzinger Reb. Decl. ¶¶ 24-25. With respect to assignments, the Big Three DPPs (AmerisourceBergen, Cardinal Health and McKesson) have allegedly assigned one-third of their claims to opt-out plaintiffs. Defendants argue that those assignments and exclusions mean that individualized inquiry into the terms of the assignments and underlying contracts is required to assure the correct numbers are put into Leitzinger's model. Def. DPP Oppo. at 18-19. However, Leitzinger's model can be adjusted to account for opt-outs and related assignments. Leitzinger Reb. Decl. ¶ 31, DPP Reply at 10. That type of inquiry does not undermine the superiority of proceeding as a class action or call into question the methodology behind or reliability of Leitzinger's aggregate damages model.

The DPPs have shown that common questions as to injury and damages are sufficiently predominant and that resolution of these questions through a class action is superior. That Leitzinger has made a number of assumptions at this stage of the proceedings in order to show how he can (at summary judgment and trial) calculate classwide aggregate damages does not undermine that he has made a solid preliminary showing of impact and demonstrated a reliable method that can accommodate future judicial rulings, findings of fact, and changes to class membership or damages depending on opt-outs and assignments to prove aggregate damages on a common and classwide basis.

B. Numerosity and Practicality of Joinder

Defendants also argue that the DPPs fail to show that the class is sufficiently numerous to make joinder impractical under Rule 23(a)(1). Generally, a "class of 41 or more is usually sufficiently numerous." 5-23 Moore's Federal Practice - Civil § 23.22 (2016). "Although the absolute number of class members is not the sole determining factor, where a class is large in numbers, joinder will usually be impracticable." Jordan v. Cty. of L.A., 669 F.2d 1311, 1319 (9th Cir. 1982), vacated on other grounds, 459 U.S. 810 (1982); see also id. (court "inclined to find the numerosity requirement in the present case satisfied solely on the basis of the number of ascertained class members, i.e., 39, 64, and 71"). "Where the class is not so numerous, however, the number of class members does not weigh as heavily in determining whether joinder would be infeasible. In the latter situation, other factors such as the geographical diversity of class members, the ability of individual claimants to institute separate suits, and whether injunctive or declaratory relief is sought, should be considered in determining impracticability of joinder." Id.; see also Pa. Pub. Sch. Emples. Ret. Sys. v. Morgan Stanley & Co., 772 F.3d 111, 120 (2nd Cir. 2014) ("the numerosity inquiry is not strictly mathematical but must take into account the context of the particular case, in particular whether a class is superior to joinder based on other relevant factors including: (i) judicial economy, (ii) geographic dispersion, (iii) the financial resources of class members, (iv) their ability to sue separately, and (v) requests for injunctive relief that would involve future class members.").

The DPPs contend that there are 55 Class Members who are widely geographically dispersed across the United States. Leitzinger Decl., Exs. 4, 5. According to defendants' expert, there are at most 54 DPPs in the class as defined (but more likely 53), and because GPOs negotiated prices for some of the DPPs as a group, those GPO/DPPs should be considered to be one entity, reducing the number of class members to less than 30 entities. Leonard Rep. ¶ 8.

With respect to the number of class members, whether 55 or 54 or 53, the DPP class is sufficiently numerous to make joinder impracticable. As to defendants' GPO-members equal one entity argument, I am not persuaded. Leonard agrees that GPOs are merely membership organizations that negotiate prices and secure contract guarantees with manufacturers for smaller DPPs, and do not actually buy or pay for the drugs. Leonard Rep. ¶ 15. Simply because 29 of the Class Members belong to five different GPOs in order to secure better prices from defendants, that does not mean that the individual DPPs were not the ones to suffer the impact and harm of the alleged overcharges. The DPPs who were members of GPOs still made their own purchasing decisions (i.e., how much to purchase) and were the ones who paid the overcharges. That the smaller DPPs used GPOs in an effort to match the purchasing and negotiating power of the larger DPPs does not mean that they are not separate and independent members of the DPP class.

There is a factual dispute over whether Cedardale Distributors is a part of Cardinal Health or a separate legal entity. Compare Leonard Rep. Decl. ¶ 12 with Leitzinger Reb. Decl. at 12 n.39. The status of Cedardale Distributors is not determinative to this motion. Leonard also argues that if you take March 31, 2013 as the entry date - which is what Leitzinger did for his primary analysis - then Drogueria Central is likewise not a class member. Leonard Rep. ¶ 13. Leitzinger responds that simply because Drogueria Central stopped buying Lidoderm as of February 2013, should not be excluded from the class since the start date may end up being earlier than March 2013. Leitzinger Reb. Decl. ¶ 20. At this stage of the proceedings, Drogueria Central should not be excluded from the class.

The purchases by the 29 DPPs who used GPO-secured prices account for under 10% of defendants' sales to direct purchasers during the relevant time frame.

Even though the number of class members - 52 at a minimum - makes joinder impracticable, other relevant factors support this conclusion. One is the judicial economy from proceeding as a class action, which is especially true since 44 DPPs have claims worth less than it would realistically cost to litigate an expert- and discovery-intensive case like this one. Leitzinger Reb. Rep. ¶ 32. These smaller DPPs also may not have the market-power security to challenge defendants when they need to negotiate to purchase drugs from these same entities in the future. The wide geographic dispersion of the DPPs also weighs against joinder. Finally, that the "Big Three" DPP class members (McKesson, Cardinal Health, and AmerisourceBergen) account for 86% of the purchases only heightens the conclusion as to impracticality of joinder given the smaller-size of the other DPPs' claims. Leonard Rep. ¶¶ 8, 14.

In a notice of recent authority, Watson points to a decision from the Third Circuit in In re Modafinil Antitrust Litigation, 837 F.3d 238 (2016). There, the Third Circuit reversed the trial court's order certifying a DPP class because there were only 22 putative class members and the district court did not adequately explain why joinder of that few entities - three of which accounted for 97% of the total value of class claims - would be impracticable under Rule 23(a)(1). Modafinil does not control and is not persuasive. There are far more DPP class members here than in that case (53 versus 22) and the market concentration of the larger players is less significant (97% versus 86%). And, as explained above, both judicial economy and the geographic distribution of the DPPs support a finding that joinder is impractical (a showing that was missing from the district court's analysis in Modafinil according to the Third Circuit). The DPP class is adequately numerous. Joinder is impracticable.

C. Representativeness, Conflicts, and Superiority

Rule 23(a)(4) covers "adequacy of representations" and requires that the class representatives will fairly and adequately protect the interests of other members of the class. Ellis v. Costco Wholesale Corp., 657 F.3d 970, 980 (9th Cir. 2011). "Adequate representation depends on, among other factors, an absence of antagonism between representatives and absentees, and a sharing of interest between representatives and absentees." Id. at 985. Defendants argue that because the class includes brand only, generic only, and brand/generic purchasers, there is an inherent conflict between class members. They assert that this means that the named DPP plaintiffs are not representative and that a class cannot be certified covering all the types of DPPs given their various market positions.

As one source of alleged conflict, defendants contend that actual entry of generic Lidoderm caused some DPPs to lose sales volume as a consequence of "generic by-pass," where customers shift to purchasing generic product from the generic manufacturer instead of from other wholesalers who formerly supplied them brand drugs. According to defendants, generic by-pass means that some of the DPPs would have lost sales earlier in a but-for world where defendants did not delay competition and, therefore, some of the DPPs benefitted from the generic delay creating intra-class conflicts precluding certification. Def. DPP Oppo. 23; Leonard Rep. ¶¶ 77-78. However, the majority of courts to consider the issue have found that the "generic by-pass" theory does not create conflicts precluding certification. See, e.g., In re Niaspan Antitrust Litig., 2015 WL 4197590, at *1-2 (E.D. Pa. July 9, 2015) (generic bypass is "irrelevant as a matter of law"); Meijer, Inc. v. Abbott Labs., 2008 WL 4065839 (N.D. Cal. August 27, 2008) ("answering this question [whether it would be in the interest of some class members to operate under allegedly illegal pricing structure] would require a great deal of speculation. This fact alone negates the possibility that there is a present and apparent fundamental conflict between class members."); Meijer, Inc. v. Warner Chilcott Holdings Co. III, 246 F.R.D. 293, 304 (D.D.C. 2007) (generic bypass phenomenon does not create a conflict).

As Leonard explains it "[b]randed drug manufacturers are more likely to use wholesalers, while generic drug manufacturers are more likely to eliminate intermediaries and sell directly, e.g., to retailers. Thus, after generic entry, direct purchasers that purchased and resold the branded drug from the branded manufacturer prior to generic entry may find that their volumes had declined, with the losses flowing to other direct purchasers (such as retailers or other generic-only direct purchasers)." Leonard Decl. ¶ 78; see also Leitzinger Decl. ¶ 46 ("the circumstance in which, following generic entry, some Class members' customers buy generics directly from generic manufacturers and 'bypass' the wholesaler").

In support of their by-pass theory, defendants admit that they rely on Valley Drug Co. v. Geneva Pharm., Inc., 350 F.3d 1181, 1189 (11th Cir. 2003), a case which has been widely rejected by courts, including courts in this District. See, e.g., Braintree Laboratories, Inc. v. McKesson Corp., No. 11-80233 MISC JSW (JSC), 2011 WL 5025096 (N.D. Cal. Oct. 20, 2011) ("whether McKesson and other class members somehow benefitted from the delay of the introduction of generics to the market is irrelevant to the merits of the underlying action" and denying discovery into whether DPP profited from antitrust conduct). Courts have also rejected attempts to decrease damages under that theory. See, e.g., In re Prograf Antitrust Litig., 2014 WL 7641156, at *4 (D. Mass. Dec. 23, 2014) ("reducing damages to plaintiff wholesalers under a bypass defense is inconsistent with Hanover Shoe") (quotation and citations omitted); In re Skelaxin (Metaxalone) Antitrust Litig., 2014 WL 2002887, at *4-6 (E.D. Tenn. May 15, 2014) (same); cf. Wellbutrin XL, 2011 WL 3563385, at *16 (noting that Leitzinger could account for bypass if necessary in aggregate damages model).

Defendants and Leonard also posit that some class members "have characteristics that suggest that they were harmed by generic entry and thus would have benefited economically from the alleged delay in generic entry," and therefore "might" prefer a lost profit measure of damages as opposed to overcharges. Leonard Decl. ¶¶ 84-86. However, these hypothetical class members could protect any such interest by opting out of the class. That the DPPs are presently relying on the widely-accepted "overcharge" method of damages calculation to prove aggregate damages on behalf of the class does not create an inherent conflict precluding certification. See, e.g., Meijer, Inc. v. Abbott Labs., No. C 07-5985 CW, 2008 WL 4065839, at *7 (N.D. Cal. Aug. 27, 2008) (recognizing that while "it is theoretically possible that some class members may wish to pursue damages for lost profits rather than for overcharges, given the difficulties of proof involved and the consequent potential that a class member would be denied recovery, it is not likely" and rejecting "conflict" based on fact plaintiffs class sought overcharge damages).

Relatedly, defendants argue that because the country's three largest pharmacy chains (taking their assignments from the Big Three and covering 29.5% of the DPP class purchases) have opted-out, the DPP class suffers from "fragmentation" showing both that proceeding as a class action is not superior and that individual class members will be interested in controlling their own claims. Def. DPP Oppo. 21-22. Assuming that the DPPs with smaller claims and fewer resources to litigate their claims on their own remain, the possibility that a number of additional "large claim" DPPs might opt out to control their own cases or seek lost profits damages only increases the utility of the class device.

Finally, defendants challenge the ability of two DPPs, ASC and Betances, to act as named representatives because they lack standing to pursue their own claims. The dispute over ASC depends on to whom McKesson assigned the relevant claims, ASC or its parent corporation, Ahold USA. Compare Def. DPP Oppo. 24-25 (arguing McKesson assigned its rights to Ahold, not ASC) with DPP Reply 15 (arguing Watson's contracts and communications were with ASC). Defendants do not dispute that either ASC or Ahold is an appropriate DPP, and plaintiffs ask for leave to substitute Ahold in as necessary. DPP Reply 5. That request will be granted, if necessary, if ASC agrees that the claims were assigned to Ahold USA and that Ahold USA should be substituted in as a named DPP.

With respect to Betances, defendants argue that because Betances is organized under the laws of Puerto Rico (a territory, not a state) and antitrust conduct in a territory is not actionable under the Sections 1 and 2 Sherman Act claims, Betances does not have any claims and cannot act as a named DPP. Def. DPP Oppo. 25. The First Circuit, however, has repeatedly recognized that the Sherman Act applies to Puerto Rico. See, e.g., United States v. Peake, 804 F.3d 81, 86 (1st Cir. 2015), cert. denied, 137 S. Ct. 36 (2016) ("First, it is well-settled that, for purposes of the Sherman Act, Puerto Rico is "to be treated like a state and not like a territory."). Absent persuasive circuit authority to the contrary, Betances may pursue its claims under the Sherman Act. Defendants' standing challenges do not undermine the named plaintiffs' representativeness under Rule 23(a).

The fact that the Supreme Court recently reemphasized the Puerto Rico is not a "separate sovereign" for purposes of the double jeopardy clause does not undermine the First Circuit's case law interpreting the Sherman Act. Def. DPP Oppo. at 25 n.110 (citing Puerto Rico v. Sanchez Valle, 136 S. Ct. 1863 (2016)).

The DPPs have shown that they have a reliable methodology for proving classwide injury and damages through Leitzinger (in addition to their other sources of evidence of classwide injury). The DPPs have also shown that the class is numerous, a class action is a superior method to litigate the Sherman Act claims, and the named DPPs are adequate representatives of the DPP class. Therefore, the DPPs' motion for class certification is GRANTED. The Named Plaintiffs Betances, RDC, and ASC are hereby appointed as representatives of the Class. Faruqi & Faruqi LLP, Garwin Gerstein & Fisher, LLP, and Hagens Berman Sobol Shapiro LLP are appointed as Co-Lead Counsel and Hagens Berman Sobol Shapiro LLP is appointed as Liaison Counsel for the Certified Class.

II. END PAYOR PLAINTIFFS' MOTION FOR CLASS CERTIFICATION

Defendants make many of the same attacks on the EPPs' motion for class certification as leveled against the DPPs' motion. I will not go in depth to dispel identical arguments that have similar impact - in reality no or limited impact - as to the EPPs. But there are a number of differences in chain of distribution position of the different EPPs, stark differences in damages (both as to amount and calculation methods)and a much more complex class definition (with multiple layers of exclusions) that require further analysis and discussion.

Similar to the DPPs, the EPPs submit a Trial Plan where in Phase I liability would be determined as to an antitrust violation by using common proof to show: (i) defendants intended for their Agreement to prevent the risk of competition from less expensive generic versions of Lidoderm; (ii) Watson was ready and able to launch generic Lidoderm as early as August 23, 2012; (iii) Watson agreed to, and did in fact, delay the launch of its generic Lidoderm product; (iv) defendants' Agreement has no countervailing procompetitive justifications; (v) any proffered procompetitive justifications were not reasonable necessary to accomplish their goals; and (vi) the relevant market is Lidoderm and AB-rated generic versions of Lidoderm. EPP Trial Plan (Dkt. No. 526-15) at 2.

The EPPs propose to show antitrust impact and injury by common proof that: (i) generic drugs are significantly less expensive than the branded version of the same drug product; (ii) purchasers pay significantly less for generic drugs than they do for branded drugs; (iii) the presence of a second generic drug product on the market - such as an authorized generic - further drives down branded and generic drug prices; (iv) state laws and health benefit plans promote or require the substitution of less expensive generic drugs for branded versions once the generic drug products are on the market; (v) defendants' Agreement delayed the availability of, and competition from, generic Lidoderm; (vi) defendants understood that Watson could have launched its generic Lidoderm product at a significantly lower price than that of Endo's and Teikoku's branded product; (vii) defendants' conduct impacted all or nearly all Class members; and (viii) the pricing of Lidoderm and generic Lidoderm once Watson launched its generic Lidoderm product confirm the impact of defendants' Agreement. Id. at 3. The EPPs argue that the special verdict form submitted to the jury will track Section 1 of the Direct Purchasers' Sherman Act claims and will encompass all of the elements of plaintiffs' state law claims, allowing the jury to make findings that will be equally applicable to all plaintiffs' claims. Id.

Examples of the special verdict questions proposed by the EPPS include: (i) did defendants reach an agreement delaying competition in the Lidoderm market? (ii) Did defendants' Agreement have the effect of artificially maintaining and inflating the price of Lidoderm and generic Lidoderm? (iii) Did defendants' Agreement impact plaintiffs and members of the class by forcing them to pay more for Lidoderm and generic Lidoderm than they would have in the absence of defendants' Agreement? (iv) Absent the Agreement, would a generic version of Lidoderm have come to the market before September 15, 2013? (v) If so, what is a reasonable estimate as to when? (vi) Would an authorized generic have entered at or about the same time? Id.

In Phase II, the EPPs propose to prove class-wide aggregate damages based on the answers provided in Phase I, and evidence including: (i) the rate at which Watson's generic Lidoderm product and Endo's and Teikoku's authorized generic would have taken market share from branded Lidoderm; (ii) the prices of generic and branded Lidoderm that would have prevailed in the absence of defendants' anticompetitive Agreement; (iii) the number of units of Lidoderm purchased during the Class period; (iv) the percentage of purchases made by uninjured class members, if any; (v) that plaintiffs and class members paid more for their Lidoderm purchases or reimbursements than they would have in the absence of defendants' anticompetitive Agreement; and (vi) what plaintiffs and class members would have been charged in the absence of defendants' anticompetitive Agreement. Id. at 5.

In Phase III, damages would be allocated through an administrative process and claims form where class members would verify their generic and/or branded Lidoderm purchases during the class period and what their intent would have been if generic Lidoderm had been on the market earlier (to exclude Brand Loyalists). TPPs would also submit claims data showing the reimbursements for purchases made by their members to pharmacies or through PBMs. Id. at 6.

Defendants oppose the EPPs motion and contest the EPPs' ability to prove injury on a classwide basis, arguing that the model proposed by Singer is overinclusive and includes consumers who were not injured, specifically: (i) Brand Loyalists who would have stuck with brand Lidoderm even if generic was available earlier, and those Brand Loyalists cannot be identified with common proof; (ii) EPPs who purchased generic Lidoderm at costs above the brand costs, and were not injured; (iii) consumers who reached out of pocket maximums (and therefore were not injured); and (iv) consumers whose plans would not allow them to purchase generic if available. Those consumers, according to defendants, cannot be identified and excluded from the class with common proof.

Defendants similarly argue the model proposed by Singer is overinclusive as to TPPs because: (i) TPPs may have received rebates from Endo that exceeded their payments for brand Lidoderm; (ii) TPPs with "high consumer contributions" were not injured (although their members were); and (iii) the TPPs passed on the costs of any overcharges to their members through premiums. These TPPs likewise cannot be identified and excluded with common proof.

Defendants further argue that classwide proof cannot be used to show that the six Medicare Part D ("Part D") EPPs were injured because of government contributions. As to common questions and predominance, defendants challenge Singer's model as inherently unreliable under Daubert. And under Rule 23(a)'s other inquiries, defendants contend that the EPP class is not readily ascertainable without significant individualized inquiries, the consumer plaintiffs are inadequate class representatives, there are conflicts between the TPPs, consumers, and Part D plans that preclude certification, and differences in the applicable state laws make the case unmanageable.

While defendants challenge the adequacy of the named EPP plaintiffs, they do not challenge the adequacy of the counsel I appointed on an interim basis to prosecute the EPPs claims; Girard Gibbs LLP, Cohen Milstein Sellers & Toll PLLC, and Heins Mills & Olson PLC as Interim Co-Lead Counsel; Joseph Saveri Law Firm, Inc. as Interim Liaison Counsel; and an Executive Committee comprised of Hilliard & Shadowen LLP, Miller Law LLC, Motley Rice LLC, Robbins Geller Rudman & Dowd LLP, and The Dugan Law Firm, APLC. Dkt. No. 63. These firms have ably and vigorously litigated this case, and nothing has occurred to undermine my initial determination of their experience and adequacy.

A. Dueling Expert Approaches and Assumptions

But-For Price. In general, EPP expert Singer's approach is to rely on internal documents produced by defendants (as well as academic literature and research) to estimate the but-for prices generic and branded Lidoderm would have had in the "but-for" world of earlier generic entry. Singer Decl. ¶¶ 96-101. Defendants challenge Singer's but-for price because it is significantly lower than the actual prices EPPs paid after generic entry occurred, according to PBM data produced in this case. Defendants also attack Singer's but-for price because it allegedly does not accurately account for Watson's expected or actual production costs, or what happened in the "real world" after Watson introduced its generic.

Defendants' expert, Hughes, uses a different method to determine the but-for price. He takes the actual prices charged to EPPs after Watson's generic entry and "backcasts" them to the purchases made in the Delay Period. Hughes Decl. ¶ 13. Singer criticizes Hughes' approach, arguing that the actual prices are "tainted" by the antitrust conduct and were higher than they would have been in the immediate post-generic entry and post-AG entry periods because of that conduct. Singer Reply Decl. ¶ 2.

As discussed above, the DPPs' expert Dr. Leitzinger takes a third approach and uses the purchase ratios between generic and brand Lidoderm after Watson's entry and applies that ratio to determine his but-for cost.

As above, however, what the but-for price should have been is not appropriately resolved on this motion. It is sufficient at this juncture to note that both Singer's and Hughes' methods for determining but-for price are plausible approaches based on classwide proof, and do not rely or implicate individualized questions that would predominate over common ones. What, in the end, the but-for price is determined to be is subject to further merits-based determinations and findings by the trier of fact.

Watson's Entry. In addition to the disputed assumptions underlying the competing but-for prices, the parties dispute other assumptions in each expert's model, including when Watson would have been able to enter the but-for market and whether Watson would have had sufficient product to meet demand or would have needed to "ration" product between purchasers. The resolution of these disputes is appropriately reserved for the trier of fact (or possibly resolution on summary judgment depending on what the facts show). The existence of these disputes at this juncture does not mean that any of the experts' models are inherently unreliable. The disputes simply highlight that common proof can be used in the competing economic models to show both impact (or lack of impact) and aggregate damages (or that aggregate damages are inflated).

B. Classwide Proof of Injury and Damages

Defendants argue that the EPPs cannot prove injury from the alleged antitrust agreement on a classwide basis and, therefore, that common questions do not predominate and the class mechanism is not a superior method to resolve the antitrust claims. As with the DPPs' motion, defendants contend that some of the individual consumers and TPPs were not injured and those uninjured EPPs cannot be identified with common proof.

1. Uninjured Consumers

a. Brand Loyalists

"Brand Loyalists" are consumers who continue to purchase brand by choice, even after generics enter the market. See, e.g., Hughes Rep. ¶¶ 85-89. A Brand Loyalist would not be injured because she would continue to purchase the brand drug despite the entry of a lower-priced generic. Both sides admit that Brand Loyalists exist and are not possible to identify individually through common evidence so that they can be individually excluded from the EPP class. According to defendants, that makes the class fatally overbroad and uncertifiable because identifying them will create predominant individualized issues. See, e.g., Wellbutrin SR, 2010 WL 3855552, at *25 (E.D. Pa. Sept. 30, 2010) (rejecting certification of EPP class in part because plaintiffs did not provide "a method for identifying which individual purchasers would remain brand loyal through analysis of common information" and therefore failed to demonstrate that common proof is available to show that supra-competitive prices passed through to purchasers of both branded and generic purchasers); Provigil, 2015 WL 4737288 (E.D. Pa. Aug. 4, 2015) (where plaintiffs did not identify "a class-wide methodology for identifying those persons who purchased" the brand or generic drug but who fall within the brand loyalist exclusion from the class definition, they failed to show that common issues will predominate).

Here, however, the EPPs and their expert have developed a method for approximating the number of Brand Loyalists in the class using common evidence. Singer defines Brand Loyalists as consumers who voluntarily choose to buy the brand after generic entry (excluding those whose health plans forced them to continue to purchase the brand post-generic entry). He estimates that Brand Loyalists account for 6.1% of the EPP class purchases, and excludes those purchases from his aggregate damages model. Singer Reply Decl. ¶¶ 30, 82; Singer Sur-Reply Decl. ¶ 10. Hughes estimates that Brand Loyalists account for 24% of the EPP class. Hughes Reply Rep. ¶ 39.

Purchasers who purchased brand Lidoderm after generic entry are excluded from the class as Brand Loyalists.

The experts disagree over which consumers are uninjured Brand Loyalists who should not be in the class and whether Singer appropriately included Medicare Part D purchases and TPP payments on behalf of insureds with a "flat generic co-pays" among those injured, despite the fact that some of these consumers - according to defendants and Hughes - were Brand Loyalists and continued to purchase branded Lidoderm after generic entry. Singer explains that he continues to count the Medicare Part D consumers as injured members of the class (and did not exclude them as Brand Loyalists) because under his theory, they were injured in their pre-generic entry purchases because of the delay in brand prices moving to a more preferred/cheaper copay tier (which is what happened after generic entry when Endo renegotiated their contracts with the Part D entities). Singer points out that these Part D consumers are in fact excluded from the post-generic entry class (because they continued to purchase brand under their Medicare D plans) and those post-generic entry purchases are not included in the aggregate damages. Singer Sur-Reply Decl. (Dkt. No. 619) ¶ 3. Correcting for this "error" by Hughes, Singer argues that Hughes' own estimate of uninjured class members drops from 24% to 9.3%. Id. ¶¶ 4-7. Singer also points out that Hughes attempted to inflate the number of Brand Loyalists by excluding transactions made between September 2013 and July 2014 by consumers with flat generic copayments for brand purchases. Singer contends that while many of these transactions may have been made by flat copayors consumers (who were uninjured), the TPPs are still injured on these transactions and it is appropriate to include them in the calculation. Id. ¶ 8. In the end, Singer sticks with his 6.1% Brand Loyalist estimate after "double-checking" that figure by reviewing actual purchase data from OptumRX PBM. Id. ¶¶ 9-10.

I find that, at this juncture, Singer has appropriately accounted for Brand Loyalists in his model by excluding 6.1% of purchases from his aggregate damages estimate. While Hughes believes that the number of Brand Loyalists is higher (or the amount of Brand Loyalist purchases is higher), that dispute does not undermine the fact that both experts rely on common proof (as opposed to individualized proof) to estimate the impact Brand Loyalists have on the aggregate damages number under both of their models. Estimating the number of Brand Loyalist purchases (using a common proof methodology) is a sufficiently reliable method to remove purchases from the aggregate damages award. At the claims administration stage (if the jury finds liability and awards aggregate damages), there are a number of ways that Brand Loyalists can be identified and excluded from the damages distribution process.

b. Consumers Who Had No Cost Savings from Purchasing Lidoderm

Defendants also argue that plaintiffs' class improperly includes consumers whose health plans provided access to generic Lidoderm at the same copay tier or a less expensive copay tier as branded Lidoderm because these consumers were not injured and cannot be identified with common proof, but can only be identified by looking to the terms of individual plans.

The EPPs respond by relying on Singer's findings that only 2.92% of consumers in the class period had this type of copay structure, and within that 2.92% are many EPPs who have already been excluded from the class as "flat copayors." Singer Reb. Decl. ¶ 15. In addition, plaintiffs cite to OptumRX's data that shows when faced with a choice of whether to buy generic or brand at the same copay tier, only 25% of consumers still purchased branded. Therefore, at most 25 percent of the 2.92% (or 1.19% as adjusted) of consumers faced with identical copays are effectively Brand Loyalists, who according to Singer are either already accounted for in his uninjured Brand Loyalists estimate or, if their purchase was post-generic entry, excluded from the Class. Singer Reb. Decl. ¶ 16. That low figure is fairly consistent, according to Singer, with his initial estimate based on OptumRx data that only 1% of actual transactions for generic Lidocaine were of the same or higher price as branded Lidoderm, a figure which Singer already included in the 6.1% class-carve-out discussed above. Id. At most, therefore, and assuming that Singer's prior calculations do not already adequately account for these purchasers, Singer's 6.1% carve-out could be revised upward by another 1.19% to 7.2%. Id. No matter; Singer has articulated a sound evidence-based methodology by which the "no cost savings" and Brand Loyalist purchases can be excluded from aggregate damages using common proof, ranging from 6% to 7% of the class purchases.

Defendants also argue that consumers who reached their out of pocket maximums could not have been harmed whether or not generic Lidoderm had been available earlier. Def. EPP Oppo. at 11-12. Plaintiffs point out that if that type of consumer purchased Lidoderm even once before hitting the maximum, the consumer would be injured and appropriately included in the class. In any event, if a consumer paid nothing for the Lidoderm in this situation, the consumer would not be injured by that transaction, but the TPP who actually paid the costs would and the aggregate damages award would be unaffected.

c. Impact of Noninjured Consumers in the Class

Defendants argue that Singer's admission that the class includes uninjured purchasers (who made the 6% to 7% of the purchases) prevents certification. This does not show overinclusiveness or predominance of individualized uninjured or Brand Loyalist issues. Instead, that figure represents at most a de minimis portion of the EPP class. It is a figure that has a basis in the data regarding actual sales of Lidoderm and is arrived at by common proof. As such, and under the case law discussed above with respect to the DPPs' motion, the class is certifiable despite their inclusion in the EPP class definition. See supra at 18.

The EPPs have a classwide method to "account" for their existence, so that the 6-7% purchases are excluded from the aggregate damage award. As already noted, various methodologies can be employed at the damages allocation phase to ensure that uninjured brand loyalists are not allocated any damages. Use of those methodologies at allocation will not overwhelm the common proof issues already discussed.

It is also possible, depending on the facts found by the jury, that in absence of the Agreement and if Watson entered at risk earlier than it eventually did, Endo could have implemented a "discount brand" strategy contemplated in some of Endo's documents (Endo would have discounted its brand drug to compete with Watson's generic) and these Brand Loyalists would have been injured and properly considered part of the class. In reality, Endo did not implement that strategy and instead followed a profit maximizing strategy where it increased its prices on Watson's entry.

2. Uninjured TPPs

Defendants also assert that the EPP class is overinclusive and that injury cannot accurately be determined through common proof because various TPPs have not been injured by defendants' alleged conduct.

a. TPP Rebates

Both sides agree that TPPs received rebates provided by drug manufacturers and secured and paid through PBMs. Defendants argue that the terms of the rebates vary across the board and require individualized review depending on the size of the TPP, the type of TPP (retail or mail order pharmacy), and the TPP plan (e.g., rebates depend on at what tier a brand or generic drug is offered). Defendants posit that these rebates "may have caused" TPPs to pay less for generic Lidoderm than branded and point out two examples where named TPPs were not injured because they paid less per patch "on average" for generic than branded Lidoderm. Def. EPP Oppo. at 13; Hughes Rep. ¶¶ 100 - 108.

Plaintiffs respond, first, by noting that whether a TPP suffered damage "on average" is irrelevant because the TPP need only suffer damage on one purchase to be injured. See, e.g., In re Nexium Antitrust Litig., 777 F.3d 9, 27 (1st Cir. 2015) (where class member paid one overcharge, injury established even if suffered no damage because injury later offset); In re Delta/AirTran Baggage Fee Antitrust Litig., No. CV 1:09-MD-2089-TCB, 2016 WL 3770957, at *7 (N.D. Ga. July 12, 2016) (the "Court concludes that a person suffers a cognizable injury and is impacted by a price-fixing conspiracy at the moment he pays an antitrust overcharge, even if the anticompetitive conduct at issue also results in offsetting benefits."). That is not to say that the rebates are irrelevant; they are relevant to damages and, if large enough, to class membership. But the rebates have been addressed through a common method of proof by the EPPs' expert, who determined the existence of overcharges on all purchases after determining a but-for price and after taking into account rebates.

Defendants rely on In re Class 8 Transmission Indirect Purchaser Antitrust Litig., 140 F. Supp. 3d 339, 349 (D. Del. 2015), but the section of the decision relied on by defendants addresses alleged conflicts and inadequacy of class representatives, and is not otherwise persuasive.

Defendants criticize Singer's analysis on a number of grounds. First, he assumes that PBMs uniformly pass on 100% of the rebates to TPPs. Def. EPP Oppo. 14. But this approach is merely a conservative one that cannot result in an overestimation of impact or damages, but only an under-estimation. Second, defendants object to his determination of the but-for price of branded and generic Lidoderm and, instead, rely on Hughes's higher but-for price which results (not surprisingly) in a finding that at least four TPPs paid more for Lidoderm after generic entry than before net of rebates. See, e.g., Hughes Rep, Ex. 10c. But Hughes was only able to make that showing by using a significantly higher but-for price. As described elsewhere, Singer's but-for estimated price is based on academic research, defendants' own forecasts about the Lidoderm market, and analysis of what actually happened when a generic was introduced. Based on Singer's but-for price, all named TPPs were injured. Singer Reply Decl. ¶¶ 42-43. Hughes' but-for price is based on a different set of assumptions. The parties dispute the appropriate but-for price in this case; the determination will likely have to be made by the jury.

Singer's analysis suffices for purposes of the class certification motion and can (if necessary) be altered based on further legal rulings or jury determinations as to disputed facts.

b. TPPs with High Consumer Copayments

Similar to the argument above, Hughes applied his higher but-for price to the OptumRX data and determined that 26% of TPPs and one opt-out EPP GEHA "escaped injury" because even before rebates were factored in, their members' high copayments offset any cost difference between a brand and generic prescription. After rebates are factored in, 90% of TPPs would have paid more for generic after Endo's AG entry. Hughes Rep. ¶¶ 110, 111. Not surprisingly, Singer finds Hughes' analysis faulty because it is based on Hughes' inflated but-for price. Based on Singer's but-for price and analyzing the OptumRX data, every one of the TPPs paid more for branded Lidoderm on at least one transaction and overpaid on 90% of total purchases per month (after considering rebates and copays). Singer Reply Decl. ¶¶ 47, 67, 68. Obviously, these analyses depend upon disputed facts that underlie the determination of the but-for price. But both of these analyses also rely on common methodologies, even if the inputs of each differ based on disputed assumptions.

Singer also finds that even using Hughes' but-for price, GEHA paid more for brand Lidoderm in some of its transactions. Singer Reply Decl. ¶ 46.

c. TPPs Could "Pass On" Costs of Branded Lidoderm Through Premiums

Defendants, supported by the expert declaration of John F. Fritz (which plaintiffs seek to exclude under Daubert, discussed below), argue that the health insurance and welfare plan TPPs were not harmed because they could "pass on" and otherwise avoid injury by setting and resetting their premiums to cover prescription drug overcharges. While defendants seem to recognize that the pass on defense is not viable under federal law, they argue it is viable under the state laws at issue because TPP plaintiffs "absorbed" the overcharges. Def. EPP Oppo. 15-17. Fritz opines that the TPP insurance and welfare plans are able to recoup overcharge costs through premium adjustments and argues that the parties will be forced to analyze a myriad of individualized inquiries to determine the extent of the overcharge absorption, making class certification inappropriate. Id. 16.

See, e.g., In re Nexium Antitrust Litig., 777 F.3d 9, 27 (1st Cir. 2015) ("defendants incorrectly assume that if a class member offsets an overcharge through later savings attributable to the same or related transaction, there is no injury. But antitrust injury occurs the moment the purchaser incurs an overcharge, whether or not that injury is later offset. . . . Here, if a class member is overcharged, there is an injury, even if that class member suffers no damages." (internal citations and quotations omitted)).

However, the class as defined here is an end-payor class - by definition it only includes members who were at the end of the distribution chain and who did not resell the product to another. The cases relied on by defendants recognizing that the premiums might shield health plans from incremental costs caused by unlawful behavior are inapposite because in the antitrust or end payor context, the alleged harm is unexpected overcharges. See, e.g., Serv. Employees Int'l Union Health & Welfare Fund v. Philip Morris Inc., 249 F.3d 1068, 1075 (D.C. Cir. 2001) (plaintiffs could not recover costs of providing smoking-related health care costs from tobacco companies on RICO and fraud claims because costs of providing medical coverage generally were offset by premiums); Int'l Bhd. of Teamsters, Local 734 Health & Welfare Trust Fund v. Philip Morris Inc., 196 F.3d 818, 824 (7th Cir. 1999) (same and noting health care premiums for smokers were set higher); see also In re Methionine Antitrust Litig., 204 F.R.D. 161, 165 (N.D. Cal. 2001) (plaintiffs failed to show how they intended to show that the indirect purchaser resellers did not pass on the overcharge; there are no resellers in the EPP class here).

Defendants rely on Ironworkers Local Union 68 v. AstraZeneca Pharm., LP, 634 F.3d 1352, 1364 (11th Cir. 2011), a RICO and fraud case based on the manufacture's promotion of off-label drug uses. The court concluded that because "the insurers assumed the risk of paying for all prescriptions of drugs covered by their policies, including medically unnecessary or inappropriate prescriptions—even those caused by fraudulent marketing" the premiums were adequate to compensate for that "known risk." Id. at 1364. Not only is this case outside the antitrust/end-payor context, but it has also been disagreed with by more recent cases. See In re Avandia Mktg., Sales Practices & Prod. Liab. Litig., 804 F.3d 633, 641 (3d Cir. 2015), cert. denied sub nom. GlaxoSmithKline LLC v. Allied Servs. Div. Welfare Fund, 136 S. Ct. 2409 (2016); In re Testosterone Replacement Therapy Prod. Liab. Litig. Coordinated Pretrial Proceedings, 159 F. Supp. 3d 898, 920 (N.D. Ill. 2016); In re Neurontin Mktg. & Sales Practices Litig., 799 F. Supp. 2d 110, 120 (D. Mass. 2011), aff'd, 712 F.3d 21 (1st Cir. 2013). Defendants' reliance on pay-for-delay cases were the injured parties were not ascertainable by common proof (based a legal standard that has been rejected by the Ninth Circuit in Briseno) are not persuasive. In re Skelaxin (Metaxalone) Antitrust Litig., 299 F.R.D. 555, 571 (E.D. Tenn. 2014), reconsideration denied, No. 1:12-MD-2343, 2014 WL 1623705 (E.D. Tenn. Apr. 23, 2014); Vista Healthplan, Inc. v. Cephalon, Inc., No. 2:06-CV-1833, 2015 WL 3623005, at *12 (E.D. Pa. June 10, 2015), reconsideration denied, No. 2:06-CV-1833, 2015 WL 4737288 (E.D. Pa. Aug. 4, 2015).

Even if the pass-on defense could apply to these end-payor health insurance and welfare plan TPPs, there is no evidence that premiums are calculated either to account for antitrust overcharges or prices of specific drugs. Instead, the evidence is that the premiums are set to cover future (not past) costs based on what actuaries determine those future costs will be and known market dynamics. Fritz Rep. ¶ 1 (premiums set to cover "future" and "projected" or "expected" costs). Here the EPP health plans and welfare funds were injured as of the date they paid the overcharges; that these plans and funds may have - as part of their annual premium setting - increased premiums to cover for future health care and prescription drugs for their members in general does not show that these plans and funds did not bear the risk of or actual damage from the overcharges at issue here. For purposes of the class certification analysis, the premium-setting and recoup issues posited by defendants do not create individualized issues that undermine the predominance of the legal questions identified above.

3. Medical Part D

The proposed class includes six Medical Part D providers with whom Endo renegotiated contracts to preclude them from providing generic coverage when Watson entered. Defendants argue that this portion of the class creates more significant individualized issues, requiring analysis of each of these providers' contracts to figure out what the providers and Endo might have agreed to if Watson had entered earlier in the but-for world. Defendants' argument, again, depends on disputed factual assumptions - e.g., the date of early entry, whether Endo would have agreed to enhanced rebates in the but-for world, etc. The damages for these providers, according to Singer, is established similarly to the other EPPs (calculating each of the provider's actual purchase quantity multiplied by the eventual rebate differential and as applied to the Delay Period). Singer Rep. Decl. ¶ 138; Singer Reply Reb. ¶ 72. The jury may or may not accept the factual assumptions underlying Singer's analysis; at this stage, the theory is appropriate and supports certification as to the Part D providers.

4. Predominance

As discussed above, Singer's model - and some of the factual assumptions it relied on - are sufficient at this stage to support class certification as to commonality and predominance. If Singer's model needs to be adjusted based on summary judgment or findings at trial, it can be. Plaintiffs point to In re Nexium Antitrust Litig., 777 F.3d 9 (1st Cir. 2015), a case that affirmed certification of a similar EPP class. Defendants argue that there was no "real world" data regarding the generic in Nexium, as there is here and which, according to defendants, shows that the EPP class is fatally overinclusive. But as discussed above, Singer's estimation that the EPP class at most is 6-7% overinclusive is evidence-based and does not defeat certification.

I will not determine the impact of that "real world" data on a motion for class certification. A rigorous review of Singer's (and Hughes') opinions and their reasoning, as required under recent Supreme Court precedent, establishes that the concepts and designs of their models are solid. What facts and assumptions are appropriate to include in those models (and which model is preferred) are not issues I can or should resolve on this motion.

To be clear, I am not relying on a presumption of antitrust injury. I am concluding that plaintiffs have shown that they can attempt to prove classwide impact through common evidence, including defendants' own forecasts, academic research applicable to the generic/brand drug pricing market, and Singer's model. Whether they succeed depends in large part on assumptions and facts to be tested on summary judgment or by the trier of fact. While defendants and Navarro assert that individual determinations of actual injury can be based on documentary evidence, that does not mean individual injury determinations are required. If it did, few if any antitrust class actions would be permissible.

Navarro asserts that to determine whether EPPs were injured, he needed to consult PBM records, TTP-PBM agreements, and PBM-manufacturer agreements. If that analysis indeed showed no injury to a significant portion of the EPPs, it could undermine certifiability. However, he does not attempt to conduct that sort of analysis and does not show that there is a wide swath of EPPs who were uninjured. Instead, his actual analysis of a discrete number of documents relevant to a few EPPs essentially shows that the apportionment of damages will require analysis of individual PBM and health plan contracts and purchase records. See, e.g., Navarro Rep. ¶¶ 45, 47, 52. That issue is not in dispute and does not preclude certification.

C. Damages Not Attributed to the Class

Similar to their overinclusiveness argument as to injury, defendants argue that certification should be denied because the damages model includes damages not "attributable to the class." They rely heavily on the Supreme Court's decision in Comcast Corp. v. Behrend, 133 S.Ct. 1426, 1432 (2013). There, plaintiffs initially relied on four theories of antitrust liability and calculated aggregate damages based on each of the four theories. 133 S.Ct. at 1434. However, the district court certified the class based on only one of the four theories, and plaintiffs did not provide a damages calculation for that one theory standing alone. Id. Because the plaintiffs relied on "a methodology that identifies damages that are not the result of the wrong" alleged, they did not establish that "damages are capable of measurement on a classwide basis," failing to meet the Rule 23(b)(3) requirement. Id. at 1433-34.

Comcast presents no problem to plaintiffs. They have one theory of injury and one consistent theory of damages as explained by Singer. See also In re Nexium Antitrust Litig., 777 F.3d 9, 19 (1st Cir. 2015) (rejecting challenge under Comcast where "the plaintiffs' theory and model for damages would only require that the defendants pay aggregate damages equivalent to the injury that they caused."); In re Urethane Antitrust Litig., 768 F.3d 1245, 1258-59 (10th Cir. 2014) (explaining the expert's benchmarks in Comcast became "useless" upon a ruling that three of the liability theories could not be used); In re Deepwater Horizon, 739 F.3d 790, 815 (5th Cir. 2014) (explaining that Comcast stands for the proposition that formulas for classwide measurement of damages should not be "incompatible" with liability theories); Butler v. Sears, 727 F.3d 796, 799 (7th Cir. 2013) (A damages model must "measure only those damages attributable to [the liability] theory. If the model does not even attempt to do that, it cannot" meet the requirements of Rule 23(b)(3) (citing Comcast, 133 S.Ct. at 1433)), cert. denied, 134 S.Ct. 1277 (2014)); Leyva v. Medline Indus. Inc., 716 F.3d 510, 514 (9th Cir. 2013) ("[P]laintiffs must be able to show that their damages stemmed from the defendant's actions that created the legal liability." (citing Comcast, 133 S.Ct. at 1435)).

Defendants argue that In re Wellbutrin XL Antitrust Litig., 308 F.R.D. 134 (E.D. Pa. 2015), the court relied on Comcast to find the class was not ascertainable. Def. EPP Oppo. at 23. However, the Wellbutrin court recognized that case did not present "a pure Comcast problem" and actually cited favorably Comcast's finding that at the class certification stage, damages "[c]alculations need not be exact." Id. at 149. In Skelaxin, the court noted the ongoing dispute over how far courts should stretch Comcast and simply noted (but did not rest on) that "[g]iven Comcast's requirement that the damages model and the theory of liability match, [an overinclusive damages mode] could be problematic." 299 F.R.D. at 575 (emphasis added).

Also, as noted above, in estimating aggregate damages plaintiffs have shown how purchases attributable to class members who were not damaged can be excluded on a classwide basis (e.g., aggregate damages reduced by 6-7%), and therefore avoid any Rule 23 or Rules Enabling Act problem. As to apportioning the aggregate damages, it bears repeating that the need for individualized determinations concerning damages generally does create a lack of predominance. See, e.g., In re Nexium Antitrust Litig., 777 F.3d 9, 21 (1st Cir. 2015) ("the Supreme Court in Amgen and the circuits in other cases have made clear that the need for some individualized determinations at the liability and damages stage does not defeat class certification.").

Aside from the overinclusiveness arguments discussed and rejected above, defendants also claim that aggregate damages cannot be adequately shown using Singer's model because it does not reliably distinguish between EPP purchases in the states included in the class from purchases in non-included states. Def. EPP Oppo. 24. These issues do not preclude class certification. Singer opines that while any one state may be a net exporter of Lidoderm, other states will be net importers and the differences are likely to even out. Singer Reply Decl. ¶ 60. Moreover, the issues Hughes attempts to identify as possibly occurring (i.e., resellers may have resold Lidoderm outside of their respective states and the large and arguably anomalous amount of sales in Arizona) are factors that can be accommodated by altering the inputs to the experts' models.

Defendants also challenge Singer's alleged failure to account for damages attributable to the federal government for payments to Part D providers under the low-income subsidy ("LIS"). Hughes Rep. ¶¶ 122-23. If, on summary judgment or at trial, facts are shown that TPPs were reimbursed for these overcharges by the federal government (facts currently in dispute), Singer calculates that the maximum government payment under the LIS amounts to only 1.1% of class damages; like the damages attributed to the Brand Loyalists, these can be excluded from the aggregate damages. Singer Reply Decl. ¶¶ 53-54.

Finally, defendants criticize Singer's model for its failure to exclude damages born by the PBMs that resulted from the speculated failure of the PBMs to effectively negotiate rebates and set spread prices. However, as discussed above, there is no evidence either of these scenarios actually occurred to PBMs with respect to lidocaine patches. Defendants' speculation cannot defeat certification.

D. Ascertainability

Defendants argue that given the very complex class definitions at issue, including the numerous exceptions, as well as the lack of reliable data to identify EPPs, plaintiffs have not shown that the EPP class is "administratively ascertainable." However, the class definition - while somewhat complex - is based on objective criteria that allow potential class members to determine whether they are included in the class. See, e.g., Philips v. Ford Motor Co., No. 14-CV-02989-LHK, 2016 WL 7428810, at *12 (N.D. Cal. Dec. 22, 2016).

As the Ninth Circuit recently explained, acertainability (much less "administrative acertainability") is not a requirement under Rule 23. Briseno v. ConAgra Foods, Inc., 844 F.3d 1121, 1125 (9th Cir. 2017). Concerns about illegitimate claims and manageability, such as those expressed by defendants here, are accounted for by other provisions of Rule 23; that consumers may not have documentation to support their claims of injury or damages does not mean a class of consumers cannot be certified. Briseno, 844 F.3d at 1129-30; see Kumar v. Salov N. Am. Corp., No. 14-CV-2411-YGR, 2016 WL 3844334, at *6 (N.D. Cal. July 15, 2016) (finding class members ascertainable despite defendant's arguments that class members would have to self-identify and show "what they paid, where they purchased it, and how many times, plus whether they saw and were deceived" by a product's label). Post-judgment claims forms and other tools can be used to allow defendants to test a class member's purported entitlement to damages and to apportion damages appropriately between class members. Id. at *7; see also Briseno, 844 F.3d at 1131 (at "the claims administration stage, parties have long relied on 'claim administrators, various auditing processes, sampling for fraud detection, follow-up notices to explain the claims process, and other techniques tailored by the parties and the court' to validate claims").

Therefore, the cases defendants rely on rejecting certification because the class sought was not administratively ascertainable are inapposite. See, e.g., In re Skelaxin (Metaxalone) Antitrust Litig., 299 F.R.D. 555, 572 (E.D. Tenn. 2014), reconsideration denied, No. 1:12-MD-2343, 2014 WL 1623705 (E.D. Tenn. Apr. 23, 2014) (denying class certification because of lack of ascertainability of EPPs, given role of PBMs and others in the distribution and payment chains).

E. Adequacy

According to defendants, the individual consumer plaintiffs are inadequate class representatives. Ms. Gallotto (the only class representative with standing under Massachusetts law) is inadequate because: (a) she purchased branded Lidoderm during the class period only in May 2013; (b) she could not confirm or prove that her purchase was through Medicare Part D and not Part B (which is excluded from class); (c) she could not recall or produce records to show what her copay or co-insurance was for the purchase; and (d) her ill health prevented her from sitting for a deposition. Plaintiffs respond that Gallotto provided purchase price details that established the cost to her TPP and her 25% co-pay (standard for Part D coverage), along with documents demonstrating it was a "MPD" copay. While she did not sit for a deposition, her interrogatory responses confirm that she is adequate (she did not have a flat co-pay structure, she is not a Brand Loyalist), and her interrogatory responses in lieu of a deposition do not undermine her adequacy. In these circumstances, I find that Gallotto is an adequate class representative.

Defendants allege that Mr. Roller, the other individual EPP plaintiff, is inadequate because he is undamaged: he purchased Lidoderm with cash, lacked insurance in December 2013, and used a coupon which resulted in a payment lower than the generic price that month. Plaintiffs do not dispute that Roller is undamaged, but argue that the allegedly "unique" defense that could be applied to Roller (comparing actual purchase price to but-for price) is not an individualized defense because that comparison will be performed for all EPP class members. Plaintiffs also argue that Roller's interests remain aligned with the class because he was injured if not damaged. It appears to me that Roller is not an adequate class member because there is no evidence that he was injured, given his use of the coupon that lowered the price he paid to below the but-for price estimated by Singer. There is no evidence that Roller would have bought more lidocaine patches had the prices been lower or other theory of injury. In this circumstance, he is subject to a unique defense that makes him inadequate as a class representative.

As noted above, defendants do not challenge the adequacy of the counsel appointed to represent the EPP class on an interim basis, and nothing since that time has undermined that finding.

At oral argument, plaintiffs' counsel asked for leave to substitute in a new class representative if I were to find one or both of them inadequate. That request is granted and plaintiffs may substitute in a new class representative for Roller with 45 days of the date of this Order.

Defendants will be allowed to take limited fact discovery to test the adequacy of any proposed additional named class representative.

F. Representativeness and Conflicts

Because the TPPs and consumers are in different positions in the distribution chain - and make different choices along the way - defendants assert that TPPs may be "in conflict" over fact of injury and amount of damages in any given transaction. The specific conflicts asserted are that: (i) the amount of overcharge damages will need to be assigned between TPPs and end consumers, putting the parties into conflict over who gets what recovery; and (ii) some of the Part D plans and other TPPs would prefer different legal theories about what would have happened in the but-for world, creating conflicts.

As to the first theory of conflicts, defendants have not shown that the alleged conflict would permeate the aggregate damages calculation. Instead, it arises at the time damages are allocated. And at that juncture, claims mechanisms (which rely on EPP documentation or sworn affidavits) may be employed to resolve any theoretical disputes between, for example, an end payor consumer and her health insurance plan over how their overcharge damages should be split. This does not create a type of conflict that precludes certification.

Defendants rely on In re Skelaxin (Metaxalone) Antitrust Litig., 299 F.R.D. 555, 577 (E.D. Tenn. 2014), where the court found that there were conflicts precluding a finding of adequacy. There, PBMs were included in the class definition (unlike here) and the district court found that because PBMs bore some-price risk in the transaction (which I have rejected here) and that price-risk was not accurately accounted for in the plaintiffs' expert's modeling. That is not so here. Moreover, this is an apportionment case, not a case where "each class member will have to offer proof that necessarily will involve arguing that a threshold number of other [class members] would not have gotten" damages. See In re NCAA I-A Walk-On Football Players Litig., No. C04-1254C, 2006 WL 1207915, at *8 (W.D. Wash. May 3, 2006).

As to the second theory of conflicts, defendants argue that because some Part D and other TPPs may have actually benefitted if Endo had employed a "discounted brand" strategy instead of immediately launching an AG, those EPPs are incentivized to pursue different but-for theories to calculate aggregate damages, creating a conflict. This is not a case where there were two major segments of the class, one segment who were harmed by defendants' conduct and the others who benefitted. Allied Orthopedic Appliances, Inc. v. Tyco Healthcare Grp. L.P., 247 F.R.D. 156, 177 (C.D. Cal. 2007). Instead, it is a case where EPPs have chosen one damages theory over another. If this is a real concern (and I am not finding it is), EPPs who wanted to pursue different damage theories could opt-out to do so. At this juncture, however, the theoretical conflict identified by defendants does not preclude certification.

G. State Law Claims

Finally, the defendants argue that variations among the state laws invoked by the EPPs bar class certification because of material differences in those laws. See Declaration of Daniel B. Asimow, Exs. 24 & 26 [Dkt. Nos. 550-25, 550-27]. The material differences identified by defendants in their Opposition are: (i) impact on intrastate commerce (statutes use different phrasing or it is not a requirement); (ii) when enhanced damages apply (flagrant conduct, willful or knowing conduct); and (iii) differences in statutes of limitations. Plaintiffs respond that most of the state laws at issue are interpreted consistently with federal antitrust law (and therefore will rise and fall with the DPPs' Sherman Act claims) and any differences are not really material because the core elements of the state laws in play are identical.

Numerous courts in this District have certified cases involving indirect purchaser claims under different state laws. See, e.g., In re TFT-LCD (Flat Panel) Antitrust Litig., 267 F.R.D. 583, 608 (N.D. Cal. 2010), amended in part, No. M 07-1827 SI, 2011 WL 3268649 (N.D. Cal. July 28, 2011); In re Static Random Access memory (SRAM) Antitrust Litig., 264 F.R.D. 603 (N.D. Cal. 2009); see also In re Nexium (Esomeprazole) Antitrust Litig., 297 F.R.D. 168, 176 (D. Mass. 2013), aff'd sub nom. In re Nexium Antitrust Litig., 777 F.3d 9 (1st Cir. 2015) (variance in state laws and statutes of limitations do not bar class certification under Rule 23(b)(3)); see also In re Terazosin Hydrochloride, 220 F.R.D. 672, 701 (S.D. Fla. 2004) ("the Court acknowledges that management of the several state classes will raise numerous challenges. However, these challenges are ones that routinely arise in complex litigation, and they are insufficient to overcome the innumerable advantages that class treatment will afford.").

The differences in the applicable state laws identified by defendants do not appear to be material or even significant. But if they are, those differences can be readily accommodated on a special verdict form or through other mechanisms routinely employed in complex litigations like this one.

Presumably, defendants will also move for summary judgment on some of these state law claims, which could reduce the number of state laws at issue.

For the foregoing reasons, the EPPs motion for class certification is GRANTED. Girard Gibbs LLP, Cohen Milstein Sellers & Toll PLLC, and Heins Mills & Olson PLC are appointed as Co-Lead Counsel; the Joseph Saveri Law Firm, Inc. is appointed as Interim Liaison Counsel; and the following firms are approved as the Executive Committee Hilliard & Shadowen LLP, Miller Law LLC, Motley Rice LLC, Robbins Geller Rudman & Dowd LLP, and The Dugan Law Firm, APLC.

III. MOTIONS TO EXCLUDE

A. Legal Standard

Federal Rule of Evidence 702 allows a qualified expert to testify "in the form of an opinion or otherwise" where:

(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.

Expert testimony is admissible under Rule 702 if it is both relevant and reliable. See Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 589 (1993). "[R]elevance means that the evidence will assist the trier of fact to understand or determine a fact in issue." Cooper v. Brown, 510 F.3d 870, 942 (9th Cir. 2007); see also Primiano v. Cook, 598 F.3d 558, 564 (9th Cir. 2010) ("The requirement that the opinion testimony assist the trier of fact goes primarily to relevance.") (internal quotation marks omitted).

Under the reliability requirement, the expert testimony must "ha[ve] a reliable basis in the knowledge and experience of the relevant discipline." Primiano, 598 F.3d at 565. To ensure reliability, the court must "assess the [expert's] reasoning or methodology, using as appropriate such criteria as testability, publication in peer reviewed literature, and general acceptance." Id. These factors are "helpful, not definitive," and a court has discretion to decide how to test reliability "based on the particular circumstances of the particular case." Id. (internal quotation marks and footnotes omitted). "When evaluating specialized or technical expert opinion testimony, the relevant reliability concerns may focus upon personal knowledge or experience." United States v. Sandoval-Mendoza, 472 F.3d 645, 655 (9th Cir. 2006).

The inquiry into the admissibility of expert testimony is "a flexible one" where "[s]haky but admissible evidence is to be attacked by cross examination, contrary evidence, and attention to the burden of proof, not exclusion." Primiano, 598 F.3d at 564. "When the methodology is sound, and the evidence relied upon sufficiently related to the case at hand, disputes about the degree of relevance or accuracy (above this minimum threshold) may go to the testimony's weight, but not its admissibility." i4i Ltd. P'ship v. Microsoft Corp., 598 F.3d 831, 852 (Fed. Cir. 2010). The burden is on the proponent of the expert testimony to show, by a preponderance of the evidence, that the admissibility requirements are satisfied. Lust By & Through Lust v. Merrell Dow Pharm., Inc., 89 F.3d 594, 598 (9th Cir. 1996); see also Fed. R. Evid. 702 Advisory Cttee. Notes.

B. Expert Opinion of Dr. Hal Singer

Endo moves to exclude in full the expert report of Dr. Hal Singer (and presumably his rebuttal and sur-reply declarations), and his opinion that antitrust impact and aggregate damages may be established with common proof under Daubert and Federal Rule of Evidence 702. Endo argues that Singer's opinions are without actual support and his model is inherently unreliable because: (1) Singer's impact model (showing how much a generic Lidoderm would have cost but-for the delay in its release) is based on a wholly hypothetical generic Lidoderm price created by "projecting forward" based on estimates of cost generated before Lidoderm was actually on the market, whereas a more "reliable" method is like the one employed by the DPPs' expert Leitzinger, which is to work backwards from the actual prices Watson and Endo charged once their generic versions were on the market; (2) Singer's aggregate damages model is inherently unreliable because: (a) it includes data from entities that are not part of the class and purportedly but not actually excluded (e.g., PBMs and certain government payors); (b) he uses an unreliable method to attempt to exclude damages for purchased made in 33 states that are not part of the class; and (3) Singer's model does not differentiate between allegedly elevated prices paid by TPPs and consumers, so his damages model conveys "nothing about whether all or nearly all class members were impacted." Dkt. No. 522.

The motion is DENIED. Singer's calculation of but-for date and but-for prices for his model are based on reasonable assumptions and evidence, and supported by reasoned principles as well as academic scholarship. That some of those assumptions are disputed does not make Singer's reliance on them improper. The trier of fact will ultimately weigh some of these fact disputes and determine but-for dates and but-for prices, and those determinations can be input into Singer's model. As discussed above, a de minimis overstatement of class members within his aggregate damages calculations does not fatally undermine the model's utility or Singer's opinions at this juncture. Overinclusiveness (e.g., for Brand Loyalists, excluded states) can be dealt with by further refinement of the class or by reasoned deductions from the aggregate damages sought. Apportionment of the aggregate damages can be managed after the liability phase and with readily available tools that ensure the damages are provided only to those who have been injured by defendants' conduct. The attacks against Singer's model are relevant, and may be persuasive to a finder of fact, but they do not make his opinions so inherently unreliable that they should be excluded under Daubert.

C. Expert Opinion of W. Paul DeBree

Defendants also move to exclude the expert report of W. Paul DeBree under Daubert and Federal Rule of Evidence 702. Defendants argue that DeBree fails to provide reliable, relevant, and admissible evidence to support his opinions. Dkt. No. 554. DeBree is relied on by plaintiffs as an expert regarding PBMs. Defendants contend that he has "limited relevant experience" with PBMs, he only performed minimal case-specific analysis in support of his opinions, and his opinions lack foundation. Dkt. No. 554.

More specifically, defendants challenge DeBree's opinion that PBMs never pay any portion of the cost of drugs. Defendants rely on their expert, Navarrao, who opines that PBM's bear "risk" if they fail to negotiate well and might be harmed if negative spreads or unfunded guaranteed rebates to TPPs occur. As discussed above, defendants fail to identify any instance of these scenarios actually happening to any PBM in the class period, much less that it happened with respect to Lidoderm sales. Instead, admissible evidence shows that PBM did not suffer from these theoretical risks during the relevant class period. See, e.g., Sharp Supp. Decl., Ex. K (Response No. 13) ("OptumRx does not . . . incur losses on guaranteed rebates"); Declaration of Brian Hansen (Dkt. No. 524-5) ¶ 7 (PBM Prime Therapeutics has not had to "perform" on a guarantees rebate since 2012).

At most, defendants identify a general risk (which DeBree explains can be contained by contractual provisions in the PBM agreements) and point to one instance where a PBM gave a TPP sponsor a rebate of $11 per Lidoderm transaction, while Endo's own records show no rebates given to that PBM during that time frame. Plaintiffs challenge that assertion, arguing that the PBM at issue likely received rebates through another PBM.

Defendants also challenge DeBree's opinion that "pharmacy records" together with "claim processing records from PBMs" provide plaintiffs a feasible method for ascertaining class members. Defendants argue that the court in In re Wellbutrin XL Antitrust Litig., 308 F.R.D. 134 (E.D. Pa. 2015) has criticized prior similar testimony from DeBree as being too conclusory and inadequate to support class certification. Dkt. No. 554. Defendants contend that DeBree's opinion as to what the PBM records might show or their utility should be disregarded because he did not attempt to develop a list of class members identifiable from the action PBM records at issue, was unable to explain discrete examples of information from the PBM records secured from OptumRX, and could not show how PBM records could be used to identify individuals and TPPs that have been excluded from the proposed class definition.

In In re Wellbutrin XL Antitrust Litig., 308 F.R.D. 134 (E.D. Pa. 2015), the court granted a motion to decertify a class of indirect purchasers (here called end payors) because under Third Circuit precedent which is not regularly followed in the Ninth Circuit, the indirect purchaser class was not "administratively feasible" to ascertain. In reaching that decision, with respect to DeBree and another expert's testimony, the court noted: "Neither expert, however, examined or analyzed these pharmaceutical records, or the Aetna data analyzed by [defense expert], to show that they could be used to ascertain PBMs and individual consumers. The Court is not persuaded by these experts' conclusory statements. Even if it were established that such records exist, the IPC has not introduced any evidence showing that such records are obtainable or can be used in an administratively feasible fashion to ascertain class members. The IPC's own expert testified that it could be difficult to obtain purchase data from PBMs. DeBree Dep. 286:22-288:16. Indeed, the IPC served subpoenas on several PBMs during the recent discovery period, but did not obtain any records from those PBMs. This heightens the Court's concern that such pharmaceutical records may not be obtainable for use in the ascertainability inquiry." Id. at 150. The Wellbutrin court did not exclude DeBree. Here significant PBM records (covering 16% of the class) have been secured from OptumRX and those records support plaintiffs' ascertainability argument.

The record in this case is starkly different than it was in Wellbutrin. A significant amount of PBM records have been secured and reviewed by DeBree and the other experts. He also relies on statements by PBMs themselves, as well as identified PBM documents, to support his opinions. He explained his extensive experience working for and advising clients regarding PBMs. His experience is sufficient to meet the Daubert threshold and allow him to give his expert opinions. Although DeBree may have overstated his position (e.g., "PBMs never pay for a portion of the drugs" as opposed to "evidence shows that PBMs bear some theoretical but rarely practical risk with respect to a particular drug"), that does not mean his opinions are without any weight. Finally, DeBree's failure to decipher all categories of PMB data at his deposition without more explanation or context does not fatally undermine his opinion that PBM records can be used to ascertain class members.

Of course, post-Briseno, acertainability is no longer the hurdle it might have been at the class certification stage.

In sum, defendants do not put forth evidence showing that PBMs - despite their spread pricing and rebates - have actually borne injury from poor negotiating or overpromising rebates to TPPs on Lidoderm. At summary judgment and trial, defendants will be free to argue that DeBree's over-statements undermine his opinions. But for purposes of a solid evidentiary basis and persuasive showing on class certification, DeBree's opinions are admissible and defendants' motion is DENIED.

Navarro points to the conclusion of the FTC that "PBMs do bear some risk of their plan client's drug spending," because spread pricing may not cover the total cost of any particular prescription. (Navarro Rep. ¶¶ 15, 24(i), 75-78) That is not evidence that PBM records will not be able to provide a reliable source of proof about class ascertainability and a source of common proof on damages.

D. Expert Opinion of John F. Fritz

The EPPs move to exclude the September 2, 2016 Report of John F. Fritz ("Fritz Report," Dkt. No. 550-33), arguing that Fritz does not meet the requirements in Federal Rule of Evidence 702 and Daubert. The EPPs argue that Fritz is not qualified to provide his opinion that no "health insurer" members of the EPP class "suffered any economic harm because of the alleged delay in the availability of generic alternative(s) for brand Lidoderm" because of the insurers' collection of premiums to cover their costs and because they can recoup any prescription overcharges when they reset premiums annually. Fritz Rep. ¶ 1; Dkt. No. 588. They also contend that his opinion is unreliable and irrelevant. Dkt. No. 587.

More specifically, plaintiffs object to Fritz's "no harm" opinion because: (1) Fritz bases it solely on his personal employment as an actuary and his purported awareness of general premium setting processes, and not on any evidence in this case and not based on any recognized methodology or professional publications; (2) it is unreliable as it is not based on evidence regarding a class member, but instead is based on an analysis of information provided or alleged by former (but now opted-out) class member GEHA, and his opinion is disproved by the only Lidoderm-specific document he considered; and (3) it is based in part on the impact of premiums and contributions collected by third-party payors ("TPPs"), but that evidence shows that plaintiffs and class members do not pass-on the overcharges they paid through premiums or contributions, and premium setting dynamics are irrelevant to injury in this type of case. See Dkt. No. 435 (denying premium-setting discovery based on lack of relevance).

Defense counsel had sought premium-related discovery from specific EPP class members. I denied them access to that discovery because: "the requests are more burdensome than probative and not proportional. The application of the pass-on defense does not appear to be appropriate in this context. For example, defendants have not cited any testimony from the 30(b)(6) of Local 49 (Johnson), despite having taken that deposition, that Local 49 was able to "recoup" amounts spent in the past on prescription drugs when setting employer contribution rates for the future. As to premium and prescription benefits plan structures, defendants have deposed the 30(b)(6) representative from Local 49 regarding the structure and design of the prescription drug benefit plans actually adopted. Additional discovery concerning alternative plans which may have been considered is overbroad, not directly relevant, and not proportional." Dkt. No. 435. --------

The motion is DENIED. The weight to be given Fritz's opinion (based on his experience, or lack thereof, and based on the information he did or did not review) is more appropriately challenged at summary judgment or trial. Moreover, while the relevance of Fritz's opinion has not been fully briefed or finally determined (although I have expressed skepticism that the "pass-on" defense will be allowed in this case at least with respect to the federal claims), as explained above, the opinion is not persuasive in my determination of the EPPs' certification motion.

CONCLUSION

The DPP and EPP motions for class certification are GRANTED. The Daubert motions are DENIED.

IT IS SO ORDERED. Dated: February 21, 2017

/s/_________

WILLIAM H. ORRICK

United States District Judge

Summaries of

In re Lidoderm Antitrust Litig.

UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF CALIFORNIA

Feb 21, 2017

Case No. 14-md-02521-WHO (N.D. Cal. Feb. 21, 2017)

finding the fact "that the experts dispute what the appropriate inputs should be does not undermine the approach or the reliability of model"

Summary of this case from In re Lamictal Indirect Purchaser & Antitrust Consumer Litig.

denying motion to exclude expert opinion because it was based on "reasonable assumptions and evidence, and supported by reasoned principles as well as academic scholarship" and while "some of those assumptions [were] disputed," those disputes did not "make [the expert's] reliance on them improper"

Summary of this case from In re EpiPen (Epinephrine Injection, USP) Mktg., Sales Practices & Antitrust Litig.

approving aggregate damages model using pro rata formula

Summary of this case from In re Suboxone (Buprenorphine Hydrochlorine & Naloxone) Antitrust Litig.

rejecting defendants' argument that "the terms of the rebates [for third-party payors ("TPPs")] vary across the board and require individualized review depending on the size of the TPP, the type of TPP . . . , and the TPP plan" because plaintiffs had provided a "common method of proof by [their] expert, who determined the existence of overcharges on all purchases after determining a but-for price and after taking into account rebates" and thus differences in TPP rebates didn't defeat predominance requirement

Summary of this case from In re EpiPen (Epinephrine Injection, USP) Mktg., Sales Practices & Antitrust Litig.

rejecting argument that "a class cannot be certified covering all the types of DPPs given their various market positions" and certifying a class of "national wholesalers, regional wholesalers, mail order wholesalers, hospitals, and retail pharmacies"

Summary of this case from In re Intuniv Antitrust Litig.

recognizing that, "in the end," deciding which expert opinion accurately estimates the but-for prices of generic and branded drugs "is subject to further merits-based determinations and findings by the trier of fact"

Summary of this case from In re EpiPen (Epinephrine Injection, USP) Mktg., Sales Practices & Antitrust Litig.

In re Lidoderm Antitrust Litig.

Summary

Opinion

INTRODUCTION

BACKGROUND

I. LIDODERM PATENT LITIGATION AND SETTLEMENT

II. CLASSES SOUGHT TO BE CERTIFIED

III. PHARMACEUTICAL DISTRIBUTION CHAIN

LEGAL STANDARD

DISCUSSION

I. DIRECT PURCHASER PLAINTIFFS' MOTION FOR CLASS CERTIFICATION

II. END PAYOR PLAINTIFFS' MOTION FOR CLASS CERTIFICATION

III. MOTIONS TO EXCLUDE

CONCLUSION

In re Lidoderm Antitrust Litig.

In re Lidoderm Antitrust Litig.

Case Details

Citations

Citing Cases

In re Lidoderm Antitrust Litig.

Summary

Opinion

INTRODUCTION

BACKGROUND

I. LIDODERM PATENT LITIGATION AND SETTLEMENT

II. CLASSES SOUGHT TO BE CERTIFIED

III. PHARMACEUTICAL DISTRIBUTION CHAIN

LEGAL STANDARD

DISCUSSION

I. DIRECT PURCHASER PLAINTIFFS' MOTION FOR CLASS CERTIFICATION

II. END PAYOR PLAINTIFFS' MOTION FOR CLASS CERTIFICATION

III. MOTIONS TO EXCLUDE

CONCLUSION

In re Lidoderm Antitrust Litig.

In re Lidoderm Antitrust Litig.

Case Details

CitationsCopy Citation

Citing Cases

Citations