Attempts to certify classes in pharmaceutical and medical device cases tend to fail more often than they succeed, particularly in the wake of the Supreme Court's Amchem and Ortiz decisions. Amchem Products, Inc. v. Windsor, 521 U.S. 591 (1997); Ortiz v. Fibreboard Corp., 527 U.S. 815 (1999). Complex issues of personal injury causation, for example, normally make class treatment impermissible. None of that is to say, however, that pharmaceutical and medical device class actions are gone forever. To the contrary, new theories of injury and causation have emerged to make class cases appear more manageable. Further, some courts continue to certify even personal injury claims.
This article briefly considers recent trends in pharmaceutical and medical device class litigation. The first part summarizes the traditional analysis that leads most often to denials of class certification. The second part discusses recent attempts by plaintiffs to streamline such cases, forsaking personal injury relief and pursuing only limited economic damages, in the hope of avoiding individual issues that preclude class treatment. The third and final part analyzes three recent published decisions where personal injury claims were certified as class actions.
I. Overview Of The Class Certification Analysis
Decisions denying certification in the pharmaceutical and medical device contexts reflect concerns expressed in the 1966 amendments to Rule 23 itself, where mass torts are deemed "ordinarily not appropriate" for class treatment. Such putative class actions tend to fail due to the interplay of the commonality, typicality, and predominance requirements of most class actions statutes, together with related concerns about manageability and judicial efficiency. State law variations introduce still further complications.
Historically, challenges to pharmaceuticals and medical devices have involved significant issues of personal injury that would make a class trial unmanageable. Any such trial would involve inquiries into a claimant's family medical history, his or her own preexisting medical conditions, as well as his or her age, gender, life style, drug or alcohol use and, in the case of pharmaceuticals, the amount of the challenged compound consumed and whether it was consumed in conjunction with other drugs. The unsuitability of giving class treatment to such claims is further underscored in circumstances where the same plaintiff seeks to represent individuals with personal injury claims as well as asymptomatic plaintiffs, raising significant conflict of interest issues.
But personal injury issues are not the only impediments to class treatment. Determining what warnings an individual received about a drug introduce further complications. Juries must determine whether class members actually received warnings, as well as read, understood, and adhered to them. These are inherently individual determinations, only compounded because different putative class members often receive differing forms of warnings over time. And if the product was obtained through an intermediary, such as a pharmacist or physician, juries must also determine whether the intermediary knew about the associated risks at issue and properly prescribed the product with those risks in mind. Medical device litigation introduces additional individual issues such as variations in product models and varying uses under each putative class member's individual circumstances.
II. Abandoning Personal Injury To Avoid Individual Issues
The question, then, is how the plaintiffs' bar is responding to these significant impediments to class certification? Many plaintiffs, it turns out, now simply disavow personal injury claims altogether. They instead pursue medical monitoring, statutory consumer fraud, or unjust enrichment claims based merely on an "enhanced health risk," together with seeking limited economic loss damages such as the return of their product purchase price. The theory of liability normally is that the product has a latent propensity to increase one's risk of harm, that the alleged risk has been concealed from the public through campaigns of false or deceptive marketing, and that the plaintiffs have, in effect, been denied the benefit of their bargain because no one would have purchased a potentially dangerous product if they had "known the truth."
This trend has been expressed in several drug cases over the past two years. For example, plaintiffs in the Rezulin litigation from the Southern District of New York, while maintaining that Rezulin, a diabetes medication, caused liver damage and death, couched their claims in terms of "enhanced health risk" and limited their prayer for relief to return of purchase price. Similarly, plaintiffs who challenged the anti-inflammatory drug Duract, while alleging that the product causes liver failure and death, limited their claim to return of purchase price because, in view of the purported product risks, they had been denied "the benefit of the bargain" when they purchased and used Duract.
This kind of no-personal-injury theory also has powerful implications for future class litigation involving clinical trials, wherein plaintiffs have already employed aggressive arguments challenging the sufficiency of informed consent. Although older than the rest of the cases discussed in this article, such implications can be seen in the Diaz v. Hillsborough County Hosp. Auth. decision, where a class was certified of all pregnant women who participated in trials of a drug designed to improve lung development of fetuses at risk of premature birth. There, no evidence existed that any fetus was harmed. Indeed, some may have benefited the drug. Despite that, plaintiffs nonetheless claimed that the defendants' disclosure of risks had been inadequate, thus depriving them of their "constitutional right to be treated with dignity."
These "no injury" tort theories pose significant risks to defendants. If accepted by a court, such claims effectively relieve the plaintiff from any burden of showing that anyone ever was actually harmed by a challenged product. Indeed, in many cases, evidence shows that few, if any, consumers really have been harmed by the challenged product, others benefited from its use, and yet the plaintiff still purports to assert a claim and have it aggregated by class treatment on behalf of all users anyway. And whether something merely "enhances a risk" has the sound of a common issue and seems to avoid any complicated questions of individual injury causation. Indeed, claims of enhanced health risk are typically subject to great scientific disagreement, but the defendant is caught in the trap of not being able to respond to potentially speculative allegations without appearing to concede that the defect really can be addressed on a common, classwide basis. The defect claim also can be susceptible to adverse media attention, because "dangerous" products – particularly ones that supposedly are dangerous to children – are "newsworthy" (and, conversely, the actual safety of a challenged product is not always particularly interesting).
All that said, this "enhanced health risk" tactic suffers from a number of critical problems that should prevent class certification.
To begin with, the seemingly narrow issue of enhanced risk really brings with it the same issues of exposure, dose, drug interaction, and personal medical history as does the traditional personal injury analysis. The challenges of quantifying consumption and dose levels highlight how individual issues normally will predominate. Different individuals will have taken a challenged drug at different times, for different periods, and in different amounts. Where the drug is obtained by prescription, physician or pharmacy records may assist in answering some of these questions. But even that minimal assistance is unavailable where the drug at issue is sold over-the-counter. Most people do not keep comprehensive records or tabulations of their over-the-counter drug purchases and usage. In an attempt to manage this problem on a class basis, some plaintiffs have proposed to quantify individual class member drug use by employing detailed affidavits. Most courts, however, have been unwilling to accept such affidavits out of concern for preserving the defendant's right of cross-examination. Of course, allowing such cross-examination of all class members would be unmanageable, thus leading courts to deny class certification. That result ought to be the same whether the claim is enhanced health risk or actual personal injury.
In addition, whether one is entitled to a return of purchase price normally requires evidence that the value the purchaser received was less than bargained for. That raises basically the same issues of injury and causation as any personal injury suit: Did the plaintiff benefit from use of the product? Did he suffer any complications? The strength of the claim will vary depending on the answers to such questions. As the Southern District found in the Rezulin case, "to obtain restitution of the purchase price of Rezulin, plaintiffs and class members would be obliged, at least in many jurisdictions, to prove some kind of harm…. [T]he question of whether an individual class member got his or her money's worth is inherently individual. Indeed, it involves very much the same question as would a claim for money damages for personal injury."
Moreover, it is far from clear whether an "enhanced health risk" claim, disavowing personal injury relief, is a truly viable tort theory. Some courts may question whether such a claim actually raises a justiciable case or controversy. In reviewing the Duract challenge, for example, the Fifth Circuit pointedly concluded that lost "expectations" or lost "benefit of the bargain" claims arise in contract, not in tort. The latter requires injury and, since the plaintiffs could not claim they suffered any of the alleged ill-effects of Duract, the court found that no Article III standing existed to prosecute a product liability action. Indeed, the notion of an "enhanced health risk" comes perilously close to a theory of unmanifested defect, which has been rejected in most jurisdictions. And, to the extent such claims sound in breach of the UCC's implied warranty of merchantability, it is far from clear whether "product safety" is a valid UCC – as opposed to traditional tort – issue.
A final issue is the impact of claim-splitting on adequacy of representation. Disavowing certain avenues of relief, like personal injury damages, to facilitate class certification raises serious due process concerns. All damages arising from or relating to a single act, or the same transaction or occurrence, must be claimed and recovered in one action or not at all. This is true in both the class setting and in individual actions. Personal injury claims arise from the same transaction that gives rise to claims of enhanced health risk and lost benefit of the bargain – the purchase and/or use of a given drug or medical device. Upon a final judgment, unasserted personal injury damages could be forever barred by operation of res judicata – for all class members. Failing to seek full recovery by splitting out personal injury claims thus creates a significant conflict of interest that may destroy adequacy of representation.
Notably, with these considerations in mind, of four recent decisions granting certification of pharmaceutical or medical device classes, the plaintiffs in only one of the cases attempted this enhanced risk, no personal injury approach. Even there, personal injury still was at issue to some extent. These decisions are addressed below.
III. Three Recent Published Decisions Granting Class Certification
A. State Court Cases
1. West Virginia
In In re West Virginia Rezulin Litigation, plaintiffs sought to certify a class of all persons who either consumed Rezulin in West Virginia or consumed the drug after having it prescribed or sold to them in West Virginia. The trial court denied certification, but the West Virginia Supreme Court of Appeals reversed and remanded.
The case has many of the hallmarks of the enhanced health risk approach outlined above. Plaintiffs challenged an "aggressive," and allegedly fraudulent marketing campaign for Rezulin that concealed what plaintiffs claimed to be an increased risk of liver-related complications. Plaintiffs asserted claims for statutory consumer fraud and medical monitoring. Defendants countered that plaintiffs could not identify anyone else in the state "who suffered a Rezulin-related injury." Defendants argued that there could be no class cohesion under such circumstances, and thus no commonality, typicality, adequacy or predominance of common issues. But the court rejected all such arguments as being improper attacks on the merits.
The court instead found class action status appropriate based on a number of inter-related conclusions. First, it concluded that medical monitoring claims only require significantly increased risks of some harmful effect, and thus complicated issues of proving an actual physical injury, and causation thereof, need not be considered in the certification analysis. Second, it applied a benefit of the bargain analysis to conclude that plaintiffs' consumer fraud claims could be resolved on a class basis: "If the consumer proves that he or she has purchased an item that is different from or inferior to that for which he bargained, the ‘ascertainable loss' requirement [of the West Virginia statute] is satisfied." Third, while the degree to which actual personal injury claims were at issue appears to be limited, the court dismissed such issues either as not at play or as merely "individual damage issues" that do not ordinarily preclude certification.
The West Virginia court's conclusions are subject to dispute and challenge for a number of reasons. For example, the answer to whether a risk (if any) is "significantly increased" obviously will vary depending on the amounts of the drug actually consumed. And the existence of a mere risk, without injury, raises the standing and justiciability issues discussed above. The answer to whether someone received "the benefit of the bargain" – ignoring the fact that this is a contract concept inapplicable to tort law – will vary depending not only on that risk equation, but also on whether he or she derived any corresponding benefit from use of the product. Further, dismissing the significance of personal injury claims as merely issues of damages is dubious at best. Personal injury is the hallmark of a traditional challenge to the safety of a pharmaceutical. Finally, the court appeared to base its conclusions in no small measure on all the purportedly common issues plaintiffs framed within their prima facie theory of the case. The court gave scant, if any, attention to the individual evidence that defendants would use to rebut plaintiffs' case. Failing to consider the defendants' evidence is inconsistent with the court's required certification analysis to look at all the facts, with an eye to how the claims and defenses would be tried to a jury. That failure puts defendants' due process rights at risk.
2. Louisiana
In Davis v. American Home Products Corporation, the Louisiana Court of Appeals held that a class of approximately 1800 Louisiana women implanted with the Norplant contraceptive device and experiencing any one of fifteen physical symptoms was appropriate for certification. Plaintiffs claimed that the Norplant device was defectively designed because the time-release mechanism for the drug distributed highly concentrated amounts of the drug throughout a user's body during the first twelve to eighteen months of use.
In the court's view, at least four factors militated in favor of certification. Unlike prescription or over-the-counter drugs, the Norplant device is implanted in a patient's body only once, constitutes one product model, contains the same prescription strength and dosage across the board, and is manufactured by one company. The court found that these factors left the ultimate issue in the case open to resolution on a class-wide basis: Was the Norplant device operationally defective? And resolution of that question, the court determined, would be aided by two additional factors. First, the court noted that the case involved only personal injury plaintiffs, as opposed to a mixture of personal injury and asymptomatic plaintiffs, and thus no potential inter-class conflicts were raised. Second, the class was restricted to female residents of Louisiana, instead of women spread across a variety of states that might raise problems of variations in applicable law.
The rational behind Davis is suspect for several reasons. Unlike the issue of enhanced risk in West Virginia Rezulin, the Davis case involved issues of causation as to actual, and extremely far-flung, physical ailments. For example, maladies like headaches, nervousness, nausea, dizziness, acne, change of appetite, weight gain, and hair loss were attributed to the Norplant device. It is difficult to imagine circumstances in which highly individual issues of causation would not predominate under those circumstances. Such individual issues would only be compounded by further issues as to whether physicians transmitted adequate warnings and properly prescribed the treatment. The court dismissed such issues as ones of damages, but they clearly go to the fundamental liability issues of injury and causation. Like West Virginia Rezulin, the court also based its conclusions almost exclusively on plaintiffs' theory of the case, without considering defendants' rebuttal evidence.
B. The One Federal Court Case: St. Jude
One perhaps surprising decision, because it issued from a federal court, certified a personal injury class in a medical device challenge. Plaintiffs in In re St. Jude Medical, Inc. Silzone Heart Valves Products Liability Litigation challenged the safety of a heart valve that allegedly caused leaks. Plaintiffs asserted causes of action for strict liability, breach of implied and express warranties, negligence, medical monitoring, and statutory consumer fraud. They sought to represent two classes of individuals implanted with Silzone heart valves: The first being of asymptomatic individuals across the United States seeking medical monitoring for side effects associated with the valve; and the second consisting of all individuals in the United States manifesting actual physical injuries allegedly caused by the valve.
The district court certified both classes, finding that personal injury issues were minimized because the case involved a single product, made by a single manufacturer, containing a single defect, with no individual issues of exposure with which to contend. It found that individual issues of causation were not "overarching," noting that the valve had been recalled due to its alleged propensity to leak, and that the British government, as well as various scientific articles and studies, likewise had identified a greater likelihood of complications from the valve.
As for the medical monitoring issues, the court based its certification order on the proposition that persons with a proven increased risk of harm, even if unmanifested, would be entitled to medical monitoring. The court found that such a class posed no Article III concerns despite the fact that some jurisdictions do not recognize an increased risk of injury to be an injury-in-fact. On this point, the court concluded that plaintiffs' expert testimony showed a plausible link between increased risk of paravalvular leaks and use of the Silzone valve. This alone was enough for the court, even though the court knew many putative class members had injuries that would never manifest themselves. Finally, the court recognized that although the number of states recognizing medical monitoring claims is increasing, a sizeable group still do not allow the claim at all. The court nonetheless certified the monitoring class, conditioned on plaintiffs proposing manageable subclasses to address any state law variations.
In January 2004, the St. Jude court partially reversed itself, but for reasons having little if anything to do with individual personal injury issues. Rather, it decertified the personal injury class because of variations in state law, ruling that the plaintiffs' reliance on six liability theories with substantial differences in interpretation across jurisdictions militated against certification. The court refused to supervise what it predicted could amount to over twenty-five subclasses. The court made only passing reference to individual fact issues: "Given the significant differences in state law and the number of plaintiffs' claims," the court wrote, "combined with potential individual differences among the plaintiffs, the Court can no longer find that a class action would be the superior method of adjudication of these disputes." The court, in turn, limited the medical monitoring class to subclasses of states that have expressly recognized such claims with various categories of requirements.
A certified personal injury class in federal court is very unusual, although it may be explained at least to some extent by the unique facts of the case. Attributing causation of a leaking heart valve to the vale itself would seem considerably easier and more manageable than proving causation in the average personal injury claim. The trouble, of course, is that the court really does not explain its rationale for concluding that common questions of fact predominate; nor does it elaborate on the defendants' basis for arguing for the opposite conclusion. One is thus left to wonder what were the "potential individual differences among the plaintiffs" to which the court referred in finally decertifying the personal injury class. Given that uncertainty, the ability of future plaintiffs to use St. Jude to expand federal court class jurisprudence further into the personal injury realm likely will be constrained.
IV. Conclusion
Based on the foregoing analysis, it is safe to say that the well-established trend of denying class certification is not likely to shift course any time soon. Putative pharmaceutical and medical device class actions requiring individual assessments of drug consumption and dosage, product variation, and warnings received, still make poor certification candidates. As the West Virginia Rezulin, Davis and St. Jude cases clearly show, however, exceptions to these rules exist and will continue to provide openings for new class certification battles in the future.