Summary
denying class certification where plaintiff had "yet to design" conjoint survey and had not yet chosen attributes that would be included in model
Summary of this case from Opperman v. Kong Techs., Inc.Opinion
CASE NO. 2:12-cv-9366-SVW (MANx)
12-18-2014
JENNIFER L. SAAVEDRA, DR. MELISSA STRAFFORD, CAROL JACQUEZ, and DAVID MATTHEWS, on behalf of themselves and all other persons similarly situated, Plaintiffs, v. ELI LILLY AND COMPANY, an Indiana corporation Defendant.
ORDER DENYING PLAINTIFFS' MOTIONS FOR CLASS CERTIFICATION PURSUANT TO FEDERAL RULES OF CIVIL PROCEDURE 23(b)(3) OR 23(c)(4) [73, 74]
I. INTRODUCTION
This is a putative class action arising from defendant Eli Lilly and Company's ("Lilly") alleged misrepresentations regarding its antidepressant, Cymbalta. Plaintiffs filed this action on October 31, 2012. In their corrected First Amended Complaint ("FAC"), Plaintiffs assert claims under four states' consumer protection laws. (Dkt. 44.)
Presently before the Court are Plaintiffs' alternative motions for class certification under Federal Rules of Civil Procedure 23(b)(3) and 23(c)(4). (Dkts. 73 & 74.) For the reasons discussed below, the Court DENIES both motions.
II. FACTUAL AND PROCEDURAL BACKGROUND
Lilly's antidepressant, Cymbalta, is available only by prescription. (Perahia Decl. ¶ 3.) Since the Food and Drug Administration approved Cymbalta in 2004, Cymbalta's United States Package Insert (called its "label") has included a warning about possible discontinuation symptoms. (Hoog Decl. ¶¶ 7, 10.) The warning states that withdrawal symptoms occurred "at a rate greater than or equal to 1% and at a significantly higher rate in duloxetine [Cymbalta's chemical name]-treated patients compared to those discontinuing from placebo." (Hoog Decl. ¶ 12.) This warning has undergone only minor revisions since 2004. (Hoog Decl. ¶ 10.) Plaintiffs Jennifer Saavedra, Melissa Strafford, Carol Jacquez, and David Matthews, Jr. (collectively, "Plaintiffs") claim that the risk of withdrawal symptoms following Cymbalta is in fact approximately 44%. (Corrected First Amended Complaint ("FAC") ¶ 30.) Plaintiffs thus claim that in marketing and advertising Cymbalta, Lilly misrepresented the risk of experiencing withdrawal symptoms upon its discontinuation.
There is some debate over whether the misrepresentations complained of appeared only in the Cymbalta label or in Lilly's nationwide advertising materials. See (Def's Opp. Ps' Motion Class Cert., at 20.)
Plaintiffs filed their FAC on January 10, 2013. (Dkt. 44.) In their FAC, Plaintiffs assert claims under: (1) California's Consumer Legal Remedies Act ("CLRA"), Cal. Civ Code §§ 1750, et seq.; (2) California's Unfair Competition Law ("UCL"), Cal. Bus. & Prof. Code §§ 17200, et seq.; (3) California's False Advertising Law ("FAL"), Cal Bus. & Prof. Code §§ 17500, et seq.; (4) Massachusetts's Consumer Protection Act, Mass. Gen. Laws Ch. 93A, §§ 1, et seq.; (5) Missouri's Merchandising Practices Act ("MPA"), Mo. Rev. Stat. §§ 407.010, et seq.; and (6) New York's Consumer Protection from Deceptive Acts and Practices Law, N.Y. Gen. Bus. Law §§ 349, et seq.
Plaintiffs also assert several individual claims not relevant to the issue of class certification.
On February 26, 2013, this Court granted Lilly's motion to dismiss Plaintiffs' claims for injunctive and declaratory relief for lack of standing. (Dkt 52.) The Court otherwise denied Lilly's motion to dismiss the complaint. (Id.)
On March 27, 2013, Lilly moved for summary judgment on the grounds that the learned intermediary doctrine barred relief and that the labels were not misleading to doctors. On June 13, 2013, the Court held that the learned intermediary doctrine applies to Plaintiffs' claims and that Plaintiffs were entitled to additional discovery before a motion for summary judgment would be heard. (Dkt. 72: Order, at 6.) The Court also directed Plaintiffs to move for class certification. (Id.) The Court directed the parties to assume for purposes of the motion that plaintiffs will prevail in showing that the warnings given to physicians were inadequate as alleged in the FAC. (Id. at 8.)
Plaintiffs now move to certify a class, including four subclasses, under Rule 23(b)(3). The class and subclasses are defined as:
All natural persons within the Commonwealth of Massachusetts and the States of Missouri, New York, and California who purchased and/or paid for Cymbalta manufactured, distributed, and/or marketed by Lilly from Cymbalta's August 2004 launch until the present, divided into the following four subclasses:
(Dkt. 73: Ps' Mot. Class Cert. Pursuant to Rule 23(b)(3), at 2-3.) Plaintiffs also filed an alternative motion to certify a similarly defined issue class and subclasses under Rule 23(c)(4). (Dkt. 74.) Plaintiffs propose certifying this issue class with respect to "the particular issue of whether Lilly's omissions regarding Cymbalta were materially misleading under the state laws of Missouri, New York, Massachusetts, and California." (Dkt. 74: Ps' Mot. Class Cert. Pursuant to Rule 23(c)(4), at 3-4.)(1) A California UCL/FAL/CLRA class of consumers who purchased and/or paid for Cymbalta in California between August 2004 and the present;
(2) A Missouri Merchandising Practices Act class of consumers who purchased and/or paid for Cymbalta in Missouri between August 2004 and the present;
(3) A New York General Business Law §§ 349-350 class of consumers who purchased and/or paid for Cymbalta in New York between August 2004 and the present;
(4) A Massachusetts General Law Chapter 93A class of consumers who purchased and/or paid for Cymbalta in Massachusetts between August 2004 and the present
III. PLAINTIFFS' MOTION FOR CLASS CERTIFICATION UNDER RULE 23(b)(3)
A. Legal Standard
A party seeking class certification must satisfy two requirements. See Zinser v. Accufix Research Inst., Inc., 253 F.3d 1180, 1186 (9th Cir. 2001), amended by 273 F.3d 1266 (9th Cir. 2001). First, the moving party must show that the proposed class meets four criteria: (1) the members of the proposed class must be so numerous that joinder of all claims would be impracticable ("numerosity"); (2) there must be questions of law and fact common to the class ("commonality"); (3) the claims or defenses of the representative parties must be typical of the claims or defenses of absent class members ("typicality"); and (4) the representative parties must fairly and adequately protect the interests of the class ("adequacy"). Fed. R. Civ. P. 23(a).
Second, the moving party must demonstrate that the class fulfills the conditions of at least one of the three subdivisions of Rule 23(b). "The party seeking certification bears the burden of showing that each of the four requirements of Rule 23(a) and at least one requirement of Rule 23(b) have been met." Zinser, 253 F.3d at 1186.
Plaintiffs assert that the class meets the requirements for Rule 23(b)(3). To qualify for certification under this subsection, a class must satisfy two conditions: (1) common questions of law or fact must "predominate over any questions affecting only individual members," and (2) class resolution must be "superior to other available methods for the fair and efficient adjudication of the controversy." Fed. R. Civ. P. 23(b)(3). The predominance requirement is satisfied where common questions comprise a significant portion of the case and can be resolved for all class members in one adjudication. See In re ConAgra Foods, Inc., — F.R.D —, No. CV 11-05379 MMM AGRX, 2014 WL 4104405, at *29 (C.D. Cal. Aug. 1, 2014). Rule 23(b)(3)'s predominance requirement also requires the moving party to show that "damages are capable of measurement on a classwide basis." Comcast Corp. v. Behrend, — U.S. —, 133 S. Ct. 1426, 1433 (2013). Specifically, this requires plaintiffs to tie their method of proving damages to their theory of liability. Id.
Nevertheless, class certification is not defeated solely by the requirement to engage in individualized damages calculations. Leyva v. Medline Indus. Inc., 716 F.3d 510, 514 (9th Cir. 2013) (allowing class certification where individualized damages could readily be calculated from computerized payroll records).
A party seeking to certify a class may not merely rest on his pleadings. Rather, "[a] party seeking class certification must affirmatively demonstrate his compliance with the Rule—that is, he must be prepared to prove that there are in fact sufficiently numerous parties, common questions of law or fact, etc." Wal-Mart Stores, Inc. v. Dukes, — U.S. —, 131 S.Ct. 2541, 2551 (2011) (emphasis added). Thus, a trial court is expected to engage in a "rigorous analysis" to determine if the moving party has discharged its burden. Dukes, 131 S.Ct. at 2551 (quoting Falcon, 457 U.S. at 161). This analysis often "will entail some overlap with the merits of the plaintiff's underlying claim." Id.
B. Application
Because Plaintiffs' motion for class certification raises more significant problems under Rule 23(b)(3) than under Rule 23(a), the Court first addresses Rule 23(b)(3).
1. Rule 23(b)(3) Requirements
a. Predominance
Much of the predominance inquiry hinges on Plaintiffs' unusual theory of injury and damages.
(1) Determining Classwide Damages
Plaintiffs rely on Dr. Joel W. Hay ("Dr. Hay") to establish their method of proving classwide damages.
(a) Legal Standard
Federal Rule of Evidence 702 governs the admissibility of expert testimony in the federal courts. Rule 702 provides that:
If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert . . . may testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case.Fed. R. Evid. 702.
In considering expert testimony presented in connection with a motion for class certification, a district court should act as a "gatekeeper" to exclude evidence that doesn't meet Rule 702's reliability standard. Daubert v. Merrill Dow Pharmaceuticals, Inc., 509 U.S. 579, 591 (1993); Ellis v. Costco Wholesale Corp., 657 F.3d 970, 982 (9th Cir. 2011). Moreover, in conducting its rigorous analysis of the motion for class certification, the court must go beyond assessing admissibility under Daubert; the court must also assess the persuasiveness of the evidence. Ellis, 657 F.3d at 982-84.
(b) Application
At the heart of Plaintiffs' case lies their novel theory of damages. Plaintiffs do not seek damages for personal injuries. Instead, Plaintiffs argue that class members were harmed because they purchased a product that was represented to have a roughly 1% risk of withdrawal side effects but that actually had an approximately 44% risk of withdrawal side effects. Thus, Plaintiffs claim they were injured because the drug as received was worth less than the drug as represented. However, Plaintiffs do not assert that class members were harmed by being overcharged or by being induced to purchase something that they would not have otherwise purchased. Instead, Plaintiffs argue that the harm was in receiving a product that had less value than the value of the product as class members expected to receive it. (Suppl. Br. Supp. Ps' Mot. Class Cert., at 4-7.)
Plaintiffs use the term "value" to mean consumer utility—a concept distinct from price. (Id. at 5) (describing method of calculating damages and stating that the analysis will "measure the benefit that consumers were deprived of by Lilly's deception" rather than price). According to Plaintiffs, consumer value (also called "utility") "is the measure of the benefit that consumers believe they will obtain by using or owning a product." It thus appears that consumer value is a subjective concept distinct from the fair market value concept commonly used when calculating benefit-of-the-bargain damages. In contrast to consumer value, fair market value is the "price that a seller is willing to accept and a buyer is willing to pay on the open market and in an arm's-length transaction; the point at which supply and demand intersect." VALUE, Black's Law Dictionary (9th ed. 2009); Schwab v. C.I.R., 715 F.3d 1169, 1178 (9th Cir. 2013) (quoting Black's Law Dictionary (9th ed. 2009).
For clarity's sake, "consumer value" is used to refer to the concept of value that Plaintiffs use.
The Court rejects Plaintiffs' assertion that price is determined only by the seller. (Pls.' Resp. to Def.'s Supp. Submission, at 18.)
Plaintiffs' theory of injury is distinct from the typical benefit-of-the bargain claim because it focuses only on the demand side of the equation, rather than on the intersection of supply and demand. In other words, Plaintiffs seek to prove injury by proving that each class member received a drug that the average consumer subjectively values less than the average consumer subjectively values the drug he expected to purchase. In contrast, the typical benefit-of-the bargain claim relies on a difference in fair market value (i.e. the amount that a willing buyer and willing seller would both accept) between the product as represented and the product actually received. As discussed below, this twist on the usual argument causes significant problems. See Apple, Inc. v. Samsung Electronics Co., No. 11-CV-01846-LHK, 2014 WL 976898, at *11 (N.D. Cal. Mar. 6, 2014) (rejecting a damages model that failed to provide a way to compare "willingness to pay metrics—which relate only to demand for the patented feature—to the market price of the infringing devices, which reflects the real-world interaction of supply and demand for infringing and noninfringing devices").
Dr. Hay proposes calculating class members' lost consumer value using conjoint analysis. Conjoint analysis is a statistical technique capable of using survey data to determine how consumers value a product's individual attributes—often called the market's willingness to pay. (Dkt. 83: Decl. of Dr. Joel W. Hay ("First Hay Decl."), ¶ 16). Conjoint analysis "predicts how, on average, all consumers value a particular attribute." (First Hay Decl., ¶ 18). Dr. Hay claims that he would be able to use conjoint analysis to determine the relative value that "consumers place on a drug with a withdrawal risk of 'greater than or equal to 1%' compared to a drug whose risk is 'at least 44%.'" (Dkt. 131: Decl. of Dr. Joel W. Hay ("Third Hay Decl."), ¶ 24). This raises the question of how to turn the relative valuation ascertained via conjoint analysis into an absolute valuation to be awarded as damages. Dr. Hay would use the relative value to determine a "refund ratio," which he would then apply to each class member's out-of-pocket costs to determine her damages. (Third Hay Decl., ¶ 25). Thus, if the value of a drug with a 1% withdrawal risk is 30% higher than a drug with a 44% withdrawal risk, then a class member's damages would be equal to 30% of her out-of-pocket costs. (Id.).
In his earlier declarations, Dr. Hay also mentions measuring damages as the dollar amount of consumers' willingness to pay for a drug with a 1% withdrawal risk rather than a 44% withdrawal risk, up to the amount of the class member's out-of-pocket costs. See (Second Hays Decl., ¶ 19). Because neither party focuses on this measure and because it was seemingly abandoned in Dr. Hay's third declaration, the Court declines to address it.
Even assuming that conjoint analysis can be used to compute the relative value that consumers place on a drug having a lower withdrawal risk (which Lilly disputes), Dr. Hay's proposed measure of damages is highly flawed.
First, as discussed above, Dr. Hay's model looks only to the demand side of the market equation. By looking only to consumer demand while ignoring supply, Dr. Hay's method of computing damages converts the lost-expectation theory from an objective evaluation of relative fair market values to a seemingly subjective inquiry of what an average consumer wants. The Court has found no case holding that a consumer may recover based on consumers' willingness to pay irrespective of what would happen in a functioning market (i.e. what could be called sellers' willingness to sell).
Plaintiffs advance Plubell v. Merck & Co., Inc., 289 S.W.3d 707 (Mo. Ct. App. 2009) in support of this proposition. Plubell considered consumers' claim that a prescription drug manufacturer violated Missouri's consumer protection statute and held that "because Plaintiffs alleged Vioxx was worth less than the product as represented, they stated an objectively ascertainable loss under the MMPA using the benefit-of-the-bargain rule." Id. at 715. However, Plubell did not make this statement while considering whether a loss in consumer value (rather than fair market value) could serve as the measure of damages. Moreover, Plubell made this statement while considering class certification under a state procedural rule that did not allow courts to conduct even a preliminary inquiry into the merits of plaintiffs' claims. See id. at 712. Thus, the court declined to address defendant's argument that pharmaceutical pricing didn't vary with risk, and stated that whether plaintiffs could prove their theory of liability was irrelevant. Id. at 715, 715 n.6. This is at odds with the Rule 23 inquiry, which is strict and often overlaps with the merits of plaintiffs' claims. See Dukes, 131 S.Ct. at 2551.
Second, as Dr. Hay readily admits, the prescription drug market is not an efficiently functioning market. (Third Hay Decl., ¶ 9). Unlike markets for ordinary consumer goods, the prescription drug market is heavily regulated and restricted. (Id.). The market is further complicated by insurance plans' (or their absense's) determinative effect on the price that an individual pays. (Suppl. Brief in Supp. Def.'s Opp. to Ps' Motions for Class Cert., at 14.) This price, in turn, relies on prices set by a complex array of contracts between such entities as health plan sponsors, third-party payers, pharmacy benefit managers, retail pharmacy chains, and the drug manufacturer. (Id.) Thus, depending on her insurance plan, an individual might pay nothing, a percentage of a "full price" determined by a contract between her insurance provider and another entity, a flat co-payment, or some other "full" price.
In an ordinary market, price is a proxy for value. (Third Hay Decl., ¶ 12.) Thus, the price paid for a good that was misrepresented to have a given characteristic can serve as a proxy for the value of a product with the misstated characteristic. Therefore, applying Dr. Hay's refund ratio to the price paid by consumers in such a market would yield a valid approximation of the value lost due to the misrepresentation. Although the refund ratio determined via conjoint analysis still looks only to the demand side of the equation, applying this ratio to the market price at least tethers it to a functioning market and thus to the product's fair market value.
In contrast, the numerous complicating factors in the prescription drug market sever the relationship between price and value. See In re POM Wonderful LLC, No. ML 10-02199 DDP RZX, 2014 WL 1225184, at *4 (C.D. Cal. Mar. 25, 2014) (stating that in contrast to an efficient market, in an inefficient market some information is not reflected in an item's price). In other words, a consumer's out-of-pocket cost for a drug is not a proxy for the drug's value to the consumer. Id.; (Dkt. 96: Suppl. Decl. of Dr. Joel W. Hay ("Second Hay Decl."), ¶ 13). Thus, class members' out-of-pocket costs are not a proxy for the value of Cymbalta as represented. Therefore, applying the refund ratio to class members' out-of-pocket costs fails to tether the consumers' relative valuations of product features to Cymbalta's fair market value. Instead, it yields an arbitrary amount that is unrelated to the amount of harm incurred by individual class members.
Dr. Hay's attempt to overcome this problem fails. Dr. Hay argues that under basic economic principles:
[A]ny money spent by the consumer, whether it is the full price or as part of a co-pay, was spent based on an overall valuation of the medication. Since [in this case] that valuation was inflated by a certain percent, i.e. the refund ratio, it makes sense to refund that percent to the consumer.(Third Hay Decl., ¶ 57c.) While Dr. Hay is correct that a rational consumer would not pay more for the drug than she believes it is worth, he forgets that a rational consumer would surely pay less than she believes the drug is worth. Thus, it does not follow that a consumer who pays a $20 co-payment believes that the drug is only worth $20. Therefore applying the refund ratio to that consumer's co-payment does not yield an accurate approximation of the difference between the consumer's subjective valuation of the drug as represented and the drug as actually received.
Admittedly, the same could be said with respect to a product's price in a functioning market—that some consumers value the product more highly than the market and are happy to pay less for the product than they perceive it to be worth. However, this is unproblematic in the typical benefit-of-the-bargain analysis. Such a theory of liability relies on differences in fair market value rather than differences in individuals' subjective valuations.
Additionally, Dr. Hay's model suffers from serious methodological flaws. He proposes conducting a survey in 2014 (or later) to estimate consumers' valuation of Cymbalta as allegedly represented and as allegedly received. He plans to apply this estimate to harms incurred by class members from 2004 to 2014. Plaintiffs assert that "absent evidence suggesting that consumers have changed their valuation of withdrawal risk itself, changes in the antidepressant marketplace over the last decade—even changes that affect the demand for Cymbalta itself—will not affect the validity of Dr. Hay's proposed study." (Supp. Brief in Supp. of Ps' Motions Class Cert., at 26). Lilly points to several changes that it asserts show that consumers have changed their valuation of withdrawal risk, including: new competitive entries, increased awareness of discontinuation side effects, consumer sensitivity to price, changes in the mix of the Cymbalta patient base, and the recent entry of six generic substitutes for Cymbalta. (Yoram Decl., ¶ 32). Dr. Hay refutes these contentions with bare statements of disagreement. See, e.g., (Third Hay Decl., ¶ 60b) ("[S]imply because consumers are more aware that withdrawal risk exists, there is no reason to assume that their valuation of that risk is any different."). Dr. Hay's bald assertions are unpersuasive in light of Lilly's evidence of events that common sense tells are potentially significant to consumer valuation.
Finally, and perhaps most importantly, Dr. Hay has yet to design the survey and method he will use in his conjoint analysis. Dr. Hay has not yet collected any data from which he will determine the refund ratio. He has not decided which attributes will be included in his model or whether he will analyze data from each of the four subclass members' states together or separately. See (Third Hay Decl., ¶¶ 15-24.) Dr. Hay admits that only relevant attributes should be included in a conjoint analysis survey. (Third Hay Decl. ¶ 18). However, he has not yet determined that the risk of withdrawal side effects is a relevant and thus appropriate attribute for inclusion in a conjoint analysis of Cymbalta. (Third Hay Decl. ¶ 55) ("As of right now, we cannot state the outcome of this first step [determining the appropriate attributes] of my proposed conjoint analysis, but there is simply no evidence that withdrawal risk is inappropriate as an attribute.") Thus, Plaintiffs have done worse than not even advancing a reliable method of calculating classwide damages—they have advanced "no damages model at all." In re ConAgra Foods, Inc., 2014 WL 4104405, at *6-7 (refusing to consider expert's conjoint analysis-based damages model in putative consumer class action because expert failed to identify variables to be included in his model or data in his possession to which the model could be applied).
Moreover, given the disconnect between the price paid by a given individual and that individual's valuation of the product, the Court finds that Dr. Hay's method is inadequate to calculate damages even on an individual basis. Accordingly, Plaintiffs have failed to show that damages could be "feasibly and efficiently calculated" once liability issues common to the class are decided. Rahman v. Mott's LLP, No. 13-CV-03482-SI, 2014 WL 6815779, at *8 (N.D. Cal. Dec. 3, 2014); accord Lilly v. Jamba Juice Co., No. 13-CV-02998-JST, 2014 WL 4652283, at *9-10 (N.D. Cal. Sept. 18, 2014) (same).
For all of these reasons, Plaintiffs' have failed to present a method of calculating damages that is tied to their theory of liability. The Court therefore DECLINES Plaintiff's motion to certify a damages class under Rule 23(b)(3).
(2) Predominance of Other Common Issues
"Considering whether 'questions of law or fact common to class members predominate' begins, of course, with the elements of the underlying cause of action." Erica P. John Fund, Inc. v. Halliburton Co., — U.S. —, 131 S. Ct. 2179, 2184 (2011) (quoting Rule 23(b)(3)). Plaintiffs and Lilly dispute the precise requirements that Plaintiffs must show to prove each of their asserted claims. Much of their dispute centers around the relevant standard of causation that must be satisfied for each claim—specifically whether proof of "but for" causation is required. Nevertheless, the parties apparently agree that each of Plaintiffs' claims requires proof of some form of causation (which Plaintiffs contend is satisfied by proof of materiality) and of some type of injury. Assuming arguendo that each of the relevant laws allows for classwide proof of causation and injury, Plaintiffs still fail to show that classwide proof of causation and injury is appropriate in this case.
The causation inquiry is complicated by Plaintiffs' unique, subjective theory of injury. In the ordinary consumer protection case allowing classwide proof of causation, the theory is: (1) a misrepresentation was made to all class members; (2) a reasonable person would find the misrepresentation important; (3) this important misrepresentation either induced individual class members to buy something they wouldn't have otherwise bought or was absorbed into the market and artificially increased the product's price; and thus, (4) class members were harmed by either buying something they would not have bought but for the misrepresentation or by overpaying for the product due to the artificially increased demand. See, e.g., Werdebaugh v. Blue Diamond Growers, No. 12-CV-2724-LHK, 2014 WL 2191901, at *1, *12-14 *18-21 (N.D. Cal. May 23, 2014) (allowing classwide proof of materiality and reliance for a class of consumers who purchased a product with an allegedly misleading label and who claimed they would not have purchased the product but for the purported misstatements); Brown v. Hain Celestial Grp., Inc., No. C 11-03082 LB, 2014 WL 6483216, at *15-18 (N.D. Cal. Nov. 18, 2014) (allowing classwide proof of materiality and reliance for class of consumers who purchased a product with an allegedly misleading label and who claimed the label caused them to overpay).
Under either of these scenarios, classwide proof of causation makes sense because each class member claims that an objectively material misstatement was made to each class member and caused them each to incur an objectively measurable harm—either paying an inflated fair market price or paying any price at all. In other words, classwide proof of causation is feasible in these circumstances because the same misstatement generally acted in a similar manner on each class member to produce a similar effect.
In stark contrast to this typical pattern, Plaintiffs argue that they each purchased something that they expected would be worth more to the average consumer than the thing they actually received—and that this difference in value exists irrespective of price. See (Supp. Br. in Support of Pls.' Mot. Class Cert. 5-6) (stating that the amount that Plaintiffs paid for Cymbalta beyond what they claim they should have paid represents "the incrementally greater value that Plaintiffs placed on Cymbalta [as a result of the alleged misrepresentation]-it does not represent a higher price") (emphasis added). As discussed above, Plaintiffs' theory of damages (and thus of injury) is essentially subjective. Because Plaintiffs divorce their injury from price, their asserted injury is akin to subjective disappointment with the product received. Assuming arguendo that subjective disappointment is a cognizable harm, the existence and degree of Plaintiffs' claimed injury will differ based on each individual's (or the individual's physician's) consideration of Cymbalta's withdrawal symptom risk relative to Cymbalta's other attributes. In other words, the causal effect of Lilly's alleged misstatements will differ widely between individuals. It is thus inappropriate to assume that because the alleged misstatement would be important to a reasonable person that it must have induced Plaintiffs to purchase Cymbalta or to otherwise incur some objectively measurable harm.
Physicians' roles in prescribing Cymbalta further attenuates the causation and materiality inquiries. The importance of the risk of withdrawal side effects to a reasonable physician may differ depending on the severity of the patient's depression or other symptoms treated with Cymbalta. In other words, the reasonable physician can defined only in relation to a patient—an elusive inquiry at best. Even ignoring the learned intermediary doctrine, the extent of a given class member's reliance on her doctor's advice—possibly to the exclusion of all other considerations—will differ depending on the person and her regard and trust in her physician.
This case is distinguishable from Krueger v. Wyeth, Inc., No. 03CV2496 JAH AJB, 2011 WL 8971449 (S.D. Cal. Mar. 30, 2011). In Krueger there was no less risky alternative product and plaintiff sought a refund of the entire purchase price (thus obviating the need for "individualized assessment of the drug's value to each class member [or] an individualized assessment of the proper comparator drug for each class member."). Id. at *6. In contrast, there are generic substitutes available for Cymbalta, as well as other antidepressants. Also, Plaintiffs seek to recover an individualized assessment of Cymbalta's allegedvalue rather than their entire purchase price.
Plaintiffs attempt to transmute their subjective injury into an objective one by expressing their theory of injury in terms of value to the average consumer. While this measure has some superficial appeal, it is nonetheless inapt. Aside from pointing to the inability to calculate fair market value in the inefficient prescription drug market, Plaintiffs fail to justify using the average consumer's willingness to pay to define injuries classwide. It is unclear to the Court why any individual is harmed when she purchases a product that the average person (but not necessarily the purchaser) subjectively overvalues because of a misrepresentation. It is also not readily apparent that an objectively important misstatement would cause Plaintiffs to pay more than Cymbalta was subjectively worth to them. This argument is akin to relying on proof of the personal injuries incurred by the average car accident victim to show that a particular car accident caused that same amount of harm to a particular victim. Neither argument rests on a sound causal nexus.
The Court thus finds that classwide proof of reliance and causation is not appropriate under Plaintiffs' theory of the case. Cf. In re ConAgra Foods, Inc., 2014 WL 4104405, at *30 (quoting Stearns v. Ticketmaster Corp., 655 F.3d 1013, 1022-23 (9th Cir. 2011)) (internal quotation marks omitted) ("Citing California law, the Ninth Circuit has held that if a misrepresentation is not material as to all class members, the issue of reliance var[ies] from consumer to consumer and no classwide inference arises."); Jimenez v. Allstate Ins. Co., 765 F.3d 1161, 1169 (9th Cir. 2014) (finding the use of statistical sampling among class members to determine liability permissible where the district court accepted a model "capable of leading to a fair determination" of the defendant's liability). For similar reasons, the Court finds that classwide proof of injury is not appropriate under Plaintiffs' theory of their case.
Finally, these concerns regarding common proof of causation and injury are compounded by the problems with Plaintiffs' damages model discussed above. Given the Court's finding that no classwide inference of causation is available in this case, Plaintiffs would need to offer some other proof that causation can be established classwide. They claim to do this through Dr. Hay's proposed model. However, the deficiencies in this model render it incapable of showing that causation may be proven classwide. Additionally, the Court is not convinced that causation could be established classwide via experts or class representatives' testimony.
For the aforementioned reasons, Plaintiffs failed to establish how causation could be proven on a classwide basis. Thus, they do not satisfy their burden of showing that common questions predominate. See In re ConAgra Foods, Inc., 2014 WL 4104405, at *30 (holding that plaintiffs failed to satisfy predominance requirement where plaintiffs did not demonstrate that reliance and causation were provable on a classwide basis).
b. Superiority
In determining whether a class action is superior to other available methods of adjudication, courts consider: (1) class members' interest in "individually controlling the prosecution or defense of separate actions"; (2) the extent and nature of any litigation concerning the controversy already begun by or against class members; (3) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and (4) the likely difficulties in managing a class action. Rule 23(b)(3)(A)-(D); Zinser, 253 F.3d at 1190.
Where damages suffered by each class member are not large, the first factor weighs in favor of class certification. Id. Here, the damages are not large. Plaintiffs seek only a percentage of their out-of-pocket costs for Cymbalta. For example, Plaintiff Saavedra seeks only a portion of the $455 of her own money that she alleges she spent on Cymbalta. (Second Hay Decl. ¶ 21). The second factor is neutral: Other individual lawsuits have been filed against Lilly regarding Cymbalta. However, those suits primarily involve personal injuries. The Court is unaware of another consumer protection action regarding Cymbalta that is similar to the case at bar. The third factor favors class treatment. Given the small amount of damages claimed by each putative class member it is desirable to consolidate all claims before one forum. See In re ConAgra Foods, Inc., 2014 WL 4104405, at *32.
However, the fourth factor—the likely difficulties in managing a class action—strongly counsels against certification. As discussed above, Plaintiffs failed to put forth a reliable and feasible method for calculating classwide damages. See Leyva, 716 F.3d at 515 (finding superiority requirement met where plaintiffs showed that individual damages could feasibly be calculated). Given the inability to calculate classwide damages, the need for individualized proof of causation and damages, and the fact-intensive individualized inquiry that will likely be required to show damages, a class action would be unmanageable. These issues are compounded by the enormous size of the putative subclasses. Plaintiffs estimate that there are over 1 million members in the California subclass, over 576,000 members in the New York subclass, over 192,000 class members in the Massachusetts subclass, and over 174,000 class members in the Missouri subclass. It would be completely unfeasible to decide each putative class member's damages (and possibly causation) in one proceeding. Thus, a class action is not the superior method for adjudicating plaintiffs' claims.
For the foregoing reasons, The Court DENIES Plaintiffs' motion to certify a class under Rule 23(b)(3).
IV. PLAINTIFFS' MOTION FOR CLASS CERTIFICATION UNDER RULE 23(C)(4)
In the alternative, Plaintiffs seek certification of an issue class under Federal Rule of Civil Procedure 23(c)(4) with respect to the "issue of whether Lilly's omissions regarding Cymbalta were materially misleading under the . . . [relevant] state laws."
Federal Rule of Civil Procedure 23(c)(4) states that "[w]hen appropriate, an action may be brought or maintained as a class action with respect to particular issues." Federal Rule of Civil Procedure 23(c)(4). The Ninth Circuit has approved the use of issue classes where certification under Rule 23(b)(3) is not proper because common questions do not predominate. See Valentino v. Carter-Wallace, Inc., 97 F.3d 1227, 1234 (9th Cir. 1996). However, neither the Ninth Circuit nor the Supreme Court has established when certification of an issue class is appropriate. One problem posed by the certification of issue classes is that this could be used as an end-run around Rule 23(b)(3)'s predominance requirement. See Castano v. Am. Tobacco Co., 84 F.3d 734, 745 n.21 (5th Cir. 1996). On the other hand, refusing to certify any issue class under Rule 23(c)(4) unless the predominance requirement was met would render Rule 23(c)(4) a nullity. See In re Nassau County Strip Search Cases, 461 F.3d 219, 226 (2d Cir. 2006).
Mindful of these competing concerns, the Court finds that certification of an issue class is not proper here. Though materiality is often evaluated on a classwide basis, as discussed above, it is not appropriate to do so here. Moreover, Plaintiffs fail to show that damages can be determined even on an individual basis once liability is decided. Thus certification of an issue class would not advance the resolution of this litigation. See In re ConAgra Foods, Inc., 2014 WL 4104405, at *33 (declining to certify a liability issue class because it was unclear that certification of the issue class would fundamentally advance the resolution of the litigation).
In Rahman v. Mott's LLP, the court refused to certify an issue class under Rule 23(c)(4). The court found that plaintiff failed to show that damages could be calculated on a class-wide basis. Rahman, 2014 WL 6815779, at *8. The court acknowledged that the Ninth Circuit seems to have implicitly endorsed other Circuits' approach of allowing the certification of liability-only classes where the plaintiff failed to establish predominance on the damages issue. Id. (citing Jimenez, 765 F.3d at 1168). The court also distinguished Lilly v. Jamba Juice Company. According to Rahman, Jamba Juice certified a liability class where the plaintiff failed to present any evidence regarding class-wide calculation of damages because "[s]ome of the difficulties in determining individual damages may fall away after liability is determined[.]" Id. at *9 (alteration in original) (quoting Jamba Juice, 2014 WL 4652283, at *11). In contrast to Jamba Juice, Rahman's claims were already refined by the court's summary judgment order and Rahman had engaged in sufficient discovery to produce the evidence required to show that damages could be proved classwide. Id.
As in Rahman and unlike in Jamba Juice, Plaintiffs have already engaged in extensive litigation. Their claims were refined by several of this Court's Orders (particularly by those regarding Defendant's motion to dismiss and motion for summary judgment). (Dkts. 52, 72.) The parties have already engaged in substantial discovery, including expert disclosures. Plaintiffs had "ample opportunity to produce evidence necessary to satisfy the requisites of Comcast and certify a class as to both liability and damages." Rahman, 2014 WL 6815779, at *9. Plaintiffs nevertheless failed to do so.
For the aforementioned reasons, the Court finds that certifying a liability-only issue class under Rule 23(c)(4) will not materially advance this case's resolution. The Court therefore DECLINES Plaintiffs' motion to certify an issue class under Rule 23(c)(4).
In light of the Court's conclusions regarding the Rule 23(b)(3) requirements and Rule 23(c)(4), the Court finds it unnecessary to address the other requirements for class certification.
V. ORDER
1. For the aforementioned reasons, the Court DENIES Plaintiffs' motion to certify a class under Rule 23(b)(3). /// /// /// /// ///
2. For the aforementioned reasons, the Court DENIES Plaintiffs' motion to certify a class under Rule 23(c)(4).
IT IS SO ORDERED. Dated: December 18, 2014
/s/_________
STEPHEN V. WILSON
United States District Judge