248 F.R.D. 363 | D. Mass. | 2008
MEMORANDUM AND ORDER
INTRODUCTION
In this proposed national class action, Plaintiffs allege that First DataBank, Inc. and McKesson Corporation engaged in a racketeering enterprise (the “Scheme”) to fraudulently state the “average wholesale price” (“AWP”) for numerous prescription pharmaceuticals beginning in late 2001, in violation of 18 U.S.C. § 1964 and California state law. The Scheme allegedly jacked up the AWP by five percent for over 400 brand-name, self-administered drugs sold through retail pharmacies, including mail order (the “Marked Up Drugs”). This allegedly fraudulent price hike caused damages to consumers and 11,000 third party payors (“TPPs”) across the nation.
Plaintiffs move to certify nationwide classes of consumers and TPPs that purchased or made reimbursements for the Marked Up Drugs. McKesson
After hearing, and a review of the extensive submissions, the motion is ALLOWED, but not to the full extent sought by Plaintiffs.
PROCEDURAL BACKGROUND
In its previous Memorandum and Order, the Court certified a class of consumers who, pursuant to a TPP plan, paid a percentage co-pay for the Marked Up Drugs (“Class 1,” or the “Co-Pay Class”) for a period covering August 1, 2001 to March 15, 2005.
The Court also found that Plaintiffs satisfied the requirements of Rule 23(a) with respect to a proposed class of TPPs (“Class 2,” or the “TPP Class”) for the same class period. Id. at 81, 84-85. The Court indicated it would certify the TPP Class for liability and equitable relief, but deferred making a decision on certification with respect to damages because the Plaintiffs failed to satisfy the predominance and superiority requirements of Rule 23(b)(3). Id. at 89.
With respect to predominance, the Court described the issue as follows:
McKesson’s better argument is that each TPP in the proposed class had unique contracts with its [Pharmacy Benefit Managers] for pharmaceuticals, and that these variations in how TPPs purchased and reimbursed for drugs would render any common inquiry into causation or damages fruitless. In the related multi-district litigation, plaintiffs’ expert Dr. Hartman testified that some TPPs employed “heat seeking missiles” in their contracts to automatically reduce their costs in the event that AWPs increased. Dr. Willig refers to these contracts as “self-adjusting” or “pass throughs.” According to defendant, for example, some plans have rebates which increased as AWP increased, or had reimbursement caps. Dr. Willig states that “the TPPs had a wide variety of mechanisms to protect themselves from an artificial rise in some AWP levels, and that individual TPPs used different means to counteract artificial rise in some AWP levels resulting from the alleged Scheme.” (Willig Exp. Report 5.)
*366 Other TPPs directly pushed back against the price increases by renegotiating their contracts with the PBMs. ESI testified that in renegotiating, amending, and renewing contracts, the FDB spread increases were important factors considered, and that other PBMs were acting in similar manner. As such, McKesson argues that some TPPs were able to recoup losses through aggressive pricing renegotiations, by obtaining additional discounts, increasing rebate pass-through provisions, lower dispensing fees, and changing to a reimbursement formula of (WAC + %) rather than an (AWP - %) benchmark.
Defendant’s expert, Dr. Willig, supported this position by stating that as AWPs rose the discounts off AWPs paid by TPPs increased over the class period. (Willig Exp. Report ¶ 48, ¶ 8.) Some TPPs shifted costs to consumers, increasing their co-pays. Still others did nothing, perhaps unaware of the AWP markup or contractually unable to mitigate the price increase. Due to all of these variables, McKesson argues that class certification is inappropriate here because issues of damages and causation are governed by substantially different contracts over the entire class period of 3.5 years, and plaintiffs would require an individualized calculation of damages for each TPP.
Id. at 86. The Court gave Plaintiffs the opportunity to address the Court’s predominance concerns by redefining the proposed class to include only TPPs which had contracts based on AWP before the start of the Scheme, and to exclude reimbursements for renegotiated contracts. See id. at 87.
With respect to the related issues of superiority and manageability, the Court found that the methodology proposed by Plaintiffs to determine aggregate damages for the TPP Class was inadequate because it did not sufficiently take into account individual TPP mitigation of damages through renegotiation of contract price terms with Pharmacy Benefit Managers (“PBMs”). Plaintiffs expert, Dr. Raymond Hartman, had calculated damages based on a “zero-knowledge — zero-mitigation” damages model, which would likely have resulted in a significant overstatement of damages if there were subsequent push-backs by the PBMs and/or TPPs in a way not captured by the model. Moreover, the Court was concerned that individualized trials or hearings on damages for those TPPs with renewed or renegotiated contracts would be unmanageable. The Court gave Plaintiffs’ expert the opportunity to submit an aggregate damage methodology alternative to address these concerns. Id. at 87-89.
McKesson sought leave to file an interlocutory appeal pursuant to Fed.R.Civ.P. 23(f), which was denied on November 16, 2007 because McKesson had not developed its reliance argument and this Court’s certification order did not “turn on” any question of law preserved by the petition.
Plaintiffs now set forth (1) their full, revised aggregate TPP damage calculation for different class periods, addressing the Court’s concerns, and also propose, as alternatives, (2) limiting the class period or (3) simply certifying the TPP class as to liability and equitable relief only. In response, McKesson contends that no TPP Class should be certified, and that the Co-Pay Class should be decertified because Plaintiffs rely upon the same flawed methodology to determine the Co-Pay Class’s aggregate damages.
FACTUAL BACKGROUND
The Court assumes familiarity with its previous Memorandum and Order, as well as certain facts established during the related multidistrict litigation. See In re Pharm. Indus. Average Wholesale Price Litig., 230 F.R.D. 61 (D.Mass.2005) (“Pharm.1”); In re Pharm. Industry Average Wholesale Price Litig., 491 F.Supp.2d 20 (D.Mass.2007). The Court also relies on the newly filed Third Amended Complaint and the extensive submissions provided by both parties, including the multiple expert declarations by Dr. Raymond Hartman,
To recap the allegations, beginning in late 2001, First DataBank, a drug pricing publisher, and McKesson, a drug wholesaler, reached a secret agreement to raise the Wholesale Acquisition Cost (“WAC”) to AWP spread from 20% to 25% for the over four hundred Marked Up Drugs. McKesson communicated these new 25% WAC to AWP markups to First DataBank, which then published AWPs with the new markup. To conceal the Scheme, McKesson and First DataBank agreed to effectuate price changes only when some other WAC-based price announcement was made by a drug manufacturer. By 2002, McKesson estimated that 95% of all prescription drug manufacturers used the inflated 25% markup, and that, by 2004, 99% of all prescription drug manufacturers did so. The Scheme ended on March 15, 2005, when First DataBank disclosed that it had ceased to conduct surveys of the market to obtain AWP information, contradicting prior statements.
The Scheme resulted in higher profits for retail pharmacies that purchase drugs on the basis of WAC, but get reimbursed on the basis of AWP. McKesson implemented the Scheme in order to provide this greater AWP “spread” to important retail pharmacy clients like Rite Aid and Walmart as well as its own pharmacy related businesses.
A hotly disputed factual issue is the extent to which TPPs mitigated any damage caused by the inflated AWP promulgated by the Scheme through renegotiation of the TPPs’ contracts with PBMs. Nearly all TPPs contract with PBMs to assist in the reimbursement process. PBMs are the “800-pound gorillas of pharmaceutical reimbursement” and their relationships with TPPs are heavily negotiated and highly individualized. See Phwrm. I at 71. TPPs negotiate drug pricing discounts with PBMs based on AWP, and PBMs in turn, negotiate discounts with the pharmacy networks based on AWP, although TPPs sometimes negotiate discounts with pharmacy networks directly. Typically, TPPs enter into contracts with PBMs to reimburse pharmacies for drugs at AWP minus 15 to 17 percent plus a dispensing fee. Sometimes PBMs earn a spread or profit margin between what they pay a retailer and what a TPP pays the PBM. Sometimes, though, TPPs use PBMs only as claims administrators and there is no such “spread.” Very few TPPs track AWP and WAC prices over time. Contracts between a TPP and PBM typically have three year terms.
DISCUSSION
A. TPP Mitigation
McKesson vigorously argues that individual issues predominate because some TPPs gained knowledge of the AWP inflation caused by the Scheme (though not of the Scheme itself) during the proposed class period, and adjusted their contracts terms with PBMs or retailers accordingly. The Court found that factual premise persuasive in McKesson I. Now, with the benefit of a much expanded record, Plaintiffs argue that the premise is wrong for two reasons.
First, they point to evidence that very few TPPs had knowledge of the increased AWP spreads caused by the Scheme. Dr. Hartman opines that the depositions, declarations and other evidence submitted by TPPs
Moreover, Plaintiffs have submitted expert declarations and a tutorial from Kimberly P. McDonough, an expert on TPP/PBM contracting who is the President of Advanced Pharmacy Concepts, an independent pharma
McKesson responds that it does not claim that every TPP had knowledge of the increased AWP spreads, only that some had knowledge, and thus any proof of impact or damage would necessarily have to be individualized. (See Docket No. 326 (‘Willig Dec!.”) ¶ 11). To support its claim, McKesson submitted a chart of evidence showing “knowledge” of the increased AWP spreads by various TPPs and PBMs. Some TPPs found out about these increases from an April 2002 letter sent by a large PBM, Express Scripts, Inc. (“ESI”). (See Docket No. 327 app. A; see also Hartman Decl. ¶46 (quoting contents of the letter)). There is a factual dispute over how many TPPs received this letter.
Second, Plaintiffs dispute McKesson’s contention — and the Court’s prior finding — that PBM competition for TPP business was “fierce” or “vigorous.” McKesson also argues that PBMs “squeezed” out of retailers any financial benefit they obtained by the spread increase. As Plaintiffs’ industry expert Kimberly McDonough says, “As the PBM industry learned of the change in AWP to WAC ratio[s], they were able to renegotiate pharmacy contract rates, reducing the prices paid to pharmacies to compensate wholly or partially for increased profit margins.” (McDonough Report at 13). In McKesson’s view, TPPs were able to “claw-back” against the increase by renegotiating their contracts with PBMs competing for their business.
In response, Plaintiffs point to the fact that some PBMs either owned or were vertically integrated with mail order and other retail pharmacies. Because these PBMs benefitted from the increased AWP spreads perpetuated by the Scheme, Plaintiffs argue that they had no incentive to inform TPPs of the inflated AWP, let alone fiercely compete to mitigate any damage. As proof, Plaintiffs quote an April 26, 2002 internal ESI e-mail, sent around the same time as the ESI letter, that states that “the AWP increases being pushed through by First Data Bank [are] having a very favorable impact on our mail margins.” (Docket No. 345 Ex. 5, at ESI-414-00005439). The e-mail goes on to state, “Our clients will not be sympathetic to our financial situation since we [will have benefit-ted] from the AWP increase in the mail and they hired us to control drug trend.” (Id.). The e-mail includes a handwritten note, in response, “Let’s put a lid on it and not make it a big deal.” (Id.).
The predominance hurdle can only be addressed by redefining the class period. Plaintiffs have presented sufficient evidence (albeit disputed) that common issues predominate over individual ones through the end of 2003. The bulk of the allegedly fraudulent increase in drug prices happened by 2002, when the inflated spread caused by the Scheme applied to approximately 95% of all drug manufacturers. McKesson provided a chart of evidence showing PBM and TPP knowledge of spread increases, which indicates that PBMs and TPPs began to learn about the spread increases beginning in April 2002. (See Docket No. 327 app. A). However, the typical TPP did not renegotiate its contract with a PBM based on knowledge of the AWP increase in mid-stream in the contract. For example, Plaintiffs claim that Blue Cross of California, which learned of the increases in early 2002, maybe even late 2001, took two years after discovery of the price hike to mitigate.
Accordingly, the Court finds that the typical TPP did not successfully negotiate contract terms to “clawback” against the hike until 2004. Beginning in 2004, individual issues of TPP knowledge and renegotiation would predominate. Thus, the Court finds that common issues predominate from August 2001 to December 2003.
B. IMS DATA
The newest debate concerns the use of IMS National Prescription Audit (“NPA”) data to determine aggregate TPP damages. IMS data “tracks the total retail method of payment for each NDC, arranged by method of payment and party of payment.” McKes-son I at 88. The database summarizes unit sales and dollar sales (by month, by drug, and by drug dosage) for transactions reimbursed at retail by (1) TPPs; (2) uninsured payors; and (3) Medicaid. Plaintiffs’ proposed methodology for calculating TPP aggregate damages relies on IMS data, which they tout as the “gold standard” in data. They argue that any individual TPP mitigation would be captured by the IMS data, and thus aggregated damages will not be overstated.
The reliability of the methodology for calculating aggregate damages on a classwide basis matters not only to the TPP Class, but to the certified Co-Pay Class as well. As put by Dr. Hartman, “since damages to Class 1 are simply a percentage of the damages (paid as coinsurance) to the TPPs/insurers that insure those consumers,” aggregate damages for Class 1 would “require calculation of the damages to the TPPs insuring those consumers.” (Hartman Deck ¶ 12). As McKesson’s expert, Arthur F. Shinn, President of Managed Pharmacy Consultants LLC, puts it, the co-pay is not a percentage of AWP but of the net cost to the TPP in reimbursing drug costs, i.e., “the consumer pays the pharmacy a specified percentage of what the TPP is obligated to pay the PBM.” (Docket No. 372 Ex. 4 (“Shinn Deck”) ¶ 11). Evidence provided by both parties
While not challenging the accuracy of this database for other commercial purposes like market research, McKesson argues that IMS data does not measure TPP mitigation, as IMS data reports payments made to pharma
In response, Plaintiffs argue that IMS data is a reasonable proxy of TPP reimbursements for drugs. To show that IMS data is a reasonable proxy (and to show that no mitigation occurred), Dr. Hartman compared IMS data to actual TPP reimbursement figures provided by GE Group Life Insurance (“GE”) and Cigna for bellwether drugs (Lipitor, Pavix, Prevacid, and Wellbutrin) that were used by Dr. Willig to determine TPP mitigation. Dr. Hartman found a close relationship between the IMS data and the TPP data with respect to the market inflation in reimbursement amounts. He also states: “There is no evidence of systematic mitigation by Cigna over all drugs that the IMS data obscures or neglects over the period for which I have data.” (Docket No. 378 (“Hartman IMS Deck”) ¶ 15(d)). Instead, he argues that there is “evidence of a uniformly inflated mark-up over two years. There is no mathematical or economic evidence of pushback or mitigation of the inflation.” (Hartman Deck ¶¶ 32-33).
Moreover, according to Dr. Hartman, IMS data “is one of the most, if not the most, frequently used sources of data summarizing a variety of business transactions, strategic behavior and corporate activity of pharmaceutical manufacturers.” (Hartman IMS Deck ¶ 2). Peer-reviewed journals have relied on IMS NPA data. (Id.) Many courts have relied on IMS data in litigation involving the pharmaceutical markets. See, e.g., In re Cardizem CD Antitrust Litig., 218 F.R.D. 508, 526 (E.D.Mich.2003) (holding, for settlement purposes, the Plaintiffs used IMS data to “make an informed judgment of ... the potential damages arising” from the alleged antitrust violation); see also In re Neurontin Marketing and Sale Practices Litigation, 244 F.R.D. 89, 110-11 (D.Mass.2007). In Hartman’s words, IMS data have been considered the “gold standard for reasonable measurement of reimbursement by end pay-ors, including TPPs.” (Hartman IMS Deck ¶ 3).
With respect to McKesson’s argument that the reliance on the IMS data leads to an overstatement of aggregate damages, Dr. Hartman states that one can carve out third party payor data in the IMS data set from Medicaid and uninsured data. (Hartman IMS Deck ¶4 & n. 6). He points out that not all TPPs use PBMs and that some PBMs are claims administrators that simply pass through reimbursements to retailers.
This debate is largely beside the point once the class is redefined to include only the period of time when the typical PBM and TPP were not renegotiating to “clawback” against the inflation caused by the Scheme. Even if the IMS data do not exactly capture that payments by TPPs for drugs, Hartman can reasonably account for mitigation by discounts, dispensing fees, and rebates in his aggregate methodology. While it is true that there may be other ways in which TPPs may have begun to mitigate in the redefined class, McKesson has not pointed to any evidence of significant pushback prior to 2004.
For RICO claims, “Where injury is established, damages need not be demonstrated with precision.” In re Zyprexa Prods. Liab. Litig., 493 F.Supp.2d 571, 578 (E.D.N.Y.2007) (quoting Schwab v. Philip Morris, Inc., 449 F.Supp.2d 992, 1065 (E.D.N.Y.2006)). This is also true of antitrust actions that involve, as in this case, “the difficult task of ‘quantifying the difference between what actually happened and what would have happened in a hypothetical free market.’ ” See Coastal Fuels, Inc. v. Caribbean Petroleum Corp., 175 F.3d 18, 33 (1st Cir.1999) (quoting Fishman v. Estate of Wirtz, 807 F.2d 520, 550 (7th Cir.1986)); see also Story Parchment Co. v. Paterson Parchment Paper Co., 282 U.S. 555, 563, 51 S.Ct. 248, 75 L.Ed. 544 (1931) (holding that in antitrust cases “Where the tort itself is of such a nature as to preclude the ascertainment of the amount of damages with certainty,” that “it will be enough if the evidence show the extent of the damages as a matter of just and reasonable inference, although the result be only approximate.”). With respect to aggregate damage calculations in the RICO context, “Courts have not required absolute precision ... and have allowed damages to be proven by reference to the class as a whole, rather than by reference to each individual class member.” Neurontin, 244 F.R.D. at 112 (quoting 3 Herbert B. Newberg & Alba Conte, Newberg on Class Actions § 10.5 (4th ed.2002)).
After a rigorous review of the expert analysis for purposes of class certification, Plaintiffs have demonstrated that the IMS data is a reasonable proxy for what TPPs paid at retail during the redefined class period. It is also a reasonable proxy for assessing aggregate damages for the entire consumer class period.
With respect to the Co-Pay Class, McKes-son urges the Court to certify for liability and equitable relief only and to decertify the damage class. However, damage distribution can be readily accomplished. As one treatise notes:
If the only issue is to determine the amount of damages which class members are entitled to receive and this determination can be accomplished almost mechanically, simple proofs similar to those used for summary judgment are often appropriate, or otherwise very simple, informal requirements and plain-English forms will do, especially when individual claims are small or relatively modest.
See 3 Alba Conte & Herbert B. Newberg, Newberg on Class Actions § 9:64, at 457-58 (4th ed.2002). Members of the Co-Pay Class purchased the Marked Up Drugs based upon a point of sale determination by a PBM, and there is no evidence that co-pays were affected by post-transaction adjustments such as rebates or other form of pushback. (Docket No. 380 (“McDonough Suppl. Decl.”) ¶¶ 2-4; Hoscheit Decl. ¶ 4). Thus, it is feasible to calculate individual damages as to the CoPay Class by determining the price paid and the point-of-sale formula that determined their co-pay. Accordingly, any individual issues posed by Co-Pay Class damage allocation poses no predominance or manageability problems.
Damage allocation with respect to TPPs is a pricklier issue because some are large insurers while others are small Tafh-Hartley
Therefore, the Court will certify the TPP Class for damage purposes through December 2003. With respect to the remaining years, the Court will certify for liability and equitable relief only.
ORDER
Plaintiffs’ motion to certify the TPP Class (Docket No. 178) is ALLOWED pursuant to Fed.R.Civ.P. 23(b)(3) for damages for a period covering August 1, 2001 to December 31, 2003. The Court certifies the TPP Class through May 15, 2005 for liability and equitable relief only pursuant to Fed.R.Civ.P. 23(b)(2). Pursuant to Fed.R.Civ.P. 23(b)(3), the Court will certify the Co-Pay Class for the entire proposed class period through May 15, 2005. The Court DENIES the motion for further discovery, and defers the decision of whether to bifurcate the trials of liability and damages.
Accordingly, the Court certifies the following class for a period beginning August 1, 2001 and ending on May 15, 2005 for all purposes:
Class 1, Consumer Purchasers: All individual persons who paid, or incurred a debt enforceable at the time of judgment in this case to pay, a percentage co-payment for the Marked Up Drugs during the Class Period based on AWP, pursuant to a plan, which in turn reimbursed the cost of brand-name pharmaceutical drugs based on AWP. The Marked Up Drugs include all of the drugs identified in Exhibit A to the Third Amended Complaint and consist of certain brand-name drugs only.
The Court also certifies the following class for a period beginning August 1, 2001 and ending on December 31, 2003 for the purpose of damages, and for a period beginning August 1, 2001 and ending on May 15, 2005 for purposes of liability and equitable relief:
Class 2, Third-Party Payors: All third-party payors (1) the pharmaceutical payments of which were based on AWP during the Class Period; (2) that made reimbursements for drugs based on an AWP that was marked up from 20 to 25% during the term of its contract with its PBM or with another entity involved in drug reimbursement; and (3) that used First DataBank or Medispan for determining the AWP of the marked up drugs. The Marked Up Drugs are all drugs identified in Exhibit A and consist of brand-name drugs only.
Excluded from both classes are (a) each defendant and any entity in which any defendant has a controlling interest, and their legal representatives, officers, directors, assignees and successors; (b) any co-conspirators; and (c) any governmental entities that purchased such drugs during the class period.
. Defendant First Databank has entered into settlement negotiations with the Plaintiffs. On Jan-
. Plaintiffs have also filed their Third Amended Complaint to include a third class of consumers who paid the "usual and customary” price for the Marked Up Drugs ("Class 3,” or the "U & C Class”). (Docket No. 359). Plaintiffs have not yet moved to certify the U & C Class, and on January 2, 2008, the Court issued an order on a briefing schedule for certification of the U & C Class.
. Plaintiffs' expert Hartman is the Director and President of Greylock McKinnon Associates, an economic consulting and litigation support firm located in Cambridge, Massachusetts. He specializes in healthcare economics, and he has
. Defendant’s expert Willig is a Professor of Economics and Public Affairs at the Woodrow Wilson School and the Economics Department at Princeton University.
. These TPPs are BCBS of Montana, DC-37, Harvard Pilgrim Health Plan, Humana, Philadelphia Federation of Teachers Health and Welfare Fund (PFTHW), Pirelli Armstrong Retirees Medical Benefits Plan, Select Health, The Teamsters Health and Welfare Fund of Philadelphia, and ConnectiCare. (Docket No. 329 Ex. A ("Hartman Deck”) ¶ 50).
. McKesson relies on the deposition testimony of William Kiefer, Vice President and General Manager of the Commercial Division of ESI, who testified that "hundreds and hundreds of clients I'm sure, if not thousands" received the letter. (See Docket No. 327 app. A). Dr. Hartman argues that only 26 TPPs actually received the ESI letter. (Hartman Deck ¶ 46).
. However, as McKesson points out, Blue Cross of California did renegotiate its mail order agreement with ESI in December 2002 that "gave [Blue Cross] a slightly deeper discount.” How
. In fact, during the fairness hearing for the proposed First Databank settlement, which proposed a unilateral reduction of AWP for approximately eight thousand NDCs, many trade groups for small pharmacies, PMBs, and at least one TPP (Blue Cross Blue Shield Michigan) objected to the settlement in part because of the difficulties and costs in renegotiating contract terms.
. McKesson submitted the details of one TPP plan that determines the co-pay "based on the [TPP]’s guaranteed price, which is the lesser of U & C, MAC plus the dispensing fee, or the guaranteed AWP plus the dispensing fee.” (Docket No. 372, Ex. 6, at MHS A_0000236).
. Dr. Hartman concludes that co-insurance accounts on average for 25% of the reimbursement (Ex. C. ¶ 58).
. Plaintiffs agree that PBMs earn spreads on some drugs, but claim they are minimal or sometimes non-existent for brand name drugs. In light of the dense record and expert declarations that sail past each other like ships in the night, it is not clear to me how many drugs involve a PBM-pharmacy spread. In fact, under McKes-son's squeeze theory. Plaintiffs argue that IMS data may actually understate damages to TPPs unless there was mitigation to "clawback” some of the profits the PBM squeezed from pharmacies.
. He explains how he calculates the increased rebates and McKesson does not appear to challenge this methodology.