More on the Upcoming Supreme Court Case of Wyeth v. Levine and the Preemption Temptation: Part Two of a Two-Part Series
|By ANTHONY J. SEBOK and BENJAMIN C. ZIPURSKY
|Tuesday, Oct. 07, 2008|
In Part One of this two-part series of columns, we argued that the upcoming Supreme Court case of Wyeth v. Levine represents a very weak tort suit. The case was based on the theory that the pharmaceutical company failed to provide adequate warnings of the risks attendant upon certain ways of administering its drug, but the record displays only a marginal difference between Levine's proposed warnings and Wyeth's actual warnings. Levine's failure to warn theory was far too slight to support a products liability claim. Thus, putting aside the possibility that we might be missing some factual or doctrinal peculiarities of the case, we think that Wyeth should have won in the Vermont courts.
For this reason, we believe the case is an unfortunate vehicle for the Court's examination of whether federal law preempts state law when it comes to warnings relating to pharmaceuticals. We are worried that the Justices might be unduly drawn to preemption arguments because the verdict against Wyeth seemed to be a product of sympathy, rather than plausible state tort law. In short, we are worried that a hard case will make bad law.
In this column, Part Two in the series, we will address the question we previously deferred: whether there is a sound argument for preemption here.
A Quick Recap of the Core Facts and Legal Proceedings
In the first column in this two-part series, we set forth the full facts of Wyeth v. Levine, a case that the United States Supreme Court will hear this Term. Briefly, the plaintiff received an IV injection of Wyeth's anti-nausea drug, Phenergan, for her migraine. Because the drug accidentally leaked into an artery, her arm became gangrenous and had to be amputated.
After settling her claim against the clinic that negligently injected the Phenergan, Levine went to trial against Wyeth, arguing that its packages' warnings of gangrene risks from IV injection were not strong enough. Levine won a multimillion-dollar verdict.
At the Vermont Supreme Court, Wyeth argued that since the package warnings that Wyeth provided were approved by the FDA, Levine's claim was preempted by federal law. The Vermont Supreme Court rejected the preemption argument and affirmed the jury's verdict. The United States Supreme Court then granted review.
Is There a Valid Case for Preemption Here? The Weaker Arguments Before the Court
In Part One, we deferred discussing the central question in Levine: whether there is a sound argument for federal preemption. To simplify our discussion, we will largely put to one side what we regard as Wyeth's more ambitious but less promising arguments, which advocate an across-the-board preemption position. For instance, its argument that a federal regulation (21 U.S.C. § 355 (a)-(c)) implicitly renders it impermissible for Wyeth to change its warning without FDA approval is not a fair reading of the text of that provision and the structure of the regulatory system, as the overwhelming majority of appellate courts have concluded.
Nor do we think it is likely that the Court will defer to the FDA's new, pro-defendant "preamble" of 2006 - which declares preemption to be the law. In the Riegel decision last term, Justice Scalia implicitly warned the FDA and the drug companies not to rely on the preamble argument.
We believe that the § 355 and preamble arguments should be viewed against the larger question at the core of the inquiry: Did Congress, in passing the Federal Food, Drug, and Cosmetic Act ("FDCA") or in amending it, mean to preempt injured persons from bringing lawsuits under state tort law against drug companies whose products injured them? We think none of the broad arguments brought by Wyeth and favored by the FDA cast any real doubt on the obvious answer: No.
The Better Pro-Preemption Arguments That Are Before the Court
This leaves a number of narrower, context-sensitive arguments for implied preemption, which we think are stronger. Wyeth's argument is as follows: The FDA has made a policy decision that Wyeth should make Phenergan available and that the availability should include IV-push, but that there should be warnings of a certain kind and stringency regarding the use of IV-push. Vermont - through its tort law, as interpreted by a jury - has made a different decision: that either IV-push should not be available at all, or that it should be available but with a prohibitively dire warning. The FDA directive about how Wyeth should behave and the Vermont tort directive are thus in conflict with one another in this context. Under the Constitution's Supremacy Clause, the Vermont directive must yield to the federal directive.
The argument comes in two forms - an "impossibility" version and a "frustration" version. Under the impossibility version, the essential problem is for Wyeth - it is put between a rock and a hard place, for it is allegedly impossible for it to comply with both federal and state law, and preemption doctrine is intended to protect legal actors from this sort of Catch-22. Under the frustration version, the problem is really for the federal agency, and more importantly for the public: The federal agency is trying to effectuate an important public policy, and Vermont tort law is frustrating its ability to do so.
We will focus here upon the frustration argument, because we suspect the impossibility argument is less likely to cover many cases, and because it is the weaker argument in Levine (if the Court reads plaintiff's complaint as a true failure-to-warn case).
The Argument that Vermont Is Frustrating Federal Policy
One of the most important aspects of the frustration argument is that it is supported, in theory and in law, by the Court's important opinion in Geier v. American Honda Motor Co.. Geier encourages courts making the preemption determination to think about two questions of institutional competence in an area of federal/state overlap: (a) Did the federal agency make a policy decision in its core area of competency about what sort of primary conduct it wished to incentivize?; and (b) Would the imposition of liability under state tort law, in the context at issue in the litigation, be likely to create an obstacle to or stand in the way of that intended incentivization, and thereby alter primary conduct away from what the competent federal regulators decided there ought to be?
To answer these questions in the Wyeth case, the Court will have to examine the record carefully to determine the FDA's policy with regard to the labeling of Phenergan.
The history of the FDA's supervision of Wyeth's labeling of Phenergan is long (beginning in 1955) and layered (there are many recorded episodes of correspondence between the FDA and Wyeth over the labeling). At first glance, it looks like the FDA actually made a judgment over whether IV push was, all things considered, safe, and whether it was important to leave it as an option for healthcare providers. But it would be a mistake for the Court to reach that conclusion.
Certainly, when Phenergan was first approved for sale by the FDA, a positive determination was made that it was "safe and effective." But no one would claim that this decision in 1955 preempts all subsequent products liability claims. Rather, the plaintiff claims that, starting in 1967, Wyeth knew about additional risks that made Phenergan unsafe unless accompanied by certain warnings concerning IV push. Wyeth's response here is that the FDA made a judgment that the warnings Wyeth proposed rendered the product safe and appropriately retained as an option for Phenergan delivery. However, this is not at all obvious from the record.
The record reveals instead that, while both Wyeth and the FDA proposed and counterproposed substantive changes concerning the warning that physicians should receive about the risk of interarterial leakage of Phenergan, the FDA never proposed restricting the method of delivery of Phenergan due to the risk of interarterial leakage. In fact, the warning language that was adopted on the label was, in substance, language recommended by Wyeth itself.
The true question here is this one: What should the U.S. Supreme Court make of the fact that there is nothing in the record to suggest that the FDA ever considered the possibility that the warning label for Phenergan should have warned against IV push except in an emergency? Can that silence be interpreted as actual endorsement of the use of IV push in circumstances other than an actual emergency?
To draw this conclusion would be to conflate two conjectured scenarios: one in which the FDA believed that Wyeth answered the risk/benefit trade-off presented by the clinical use of IV push correctly when it failed to require a stronger warning, and one in which the FDA simply did not form any opinion at all as to the whether FDA selected the right risk/benefit trade-off.
If the Court were to hold for Wyeth in the first scenario, the Court would be shielding Wyeth so as not to frustrate a real policy judgment that the FDA actually thought about and expressly made. But if it were to hold for Wyeth in the second scenario, it would be shielding Wyeth so as not to frustrate an imaginary policy judgment that was actually never made. That difference is all the difference in the world.
Given that preemption is an affirmative defense, we are inclined to think that the lack of a completed case on frustration should probably count in the plaintiff's favor. Realistically, however, the Court has a variety of intermediate options - including remand, which Prof. Cathy Sharkey has recommended.
The Most Difficult Question the Court Faces: Creating a Preemption Framework
The hard question for the Court here is not who wins this case: As should be clear by now, we think that although Wyeth should have won as a matter of tort law, Levine has the stronger case on preemption, absent remand or an extremely strong pro-defendant reading of the record. What is most important, however, is what kind of preemption framework the Court chooses to create.
The central questions in preemption cases like Levine are whether the FDA actually made a thoughtful, exhaustive, and well-informed decision about the warning or usage issue at the crux of the lawsuit, and whether the recognition of a state tort action would really frustrate a federal policy decision for consumer health protection that was made in conjunction with that decision. The line between preemption and no-preemption should be drawn by reference to the answers to these questions.
Two recent appellate decisions provide striking examples on either side of the line, and in each case, the appellate court correctly recognized on which side of the line the facts fell. In Colacicco v. Apotex, Inc., the plaintiff's estate claimed that there ought to have been an additional, explicit warning about the risk of increased suicidality in adult patients using SSRI anti-depressants, like Paxil and Zoloft. The Third Circuit held that the case was preempted because the FDA had "clearly and publicly" stated it position that "[t]here is no evidence that Paxil is associated with an increased risk of suicidal thinking in adults," and the FDA had thereby decided such warnings should not be given regarding adult usage.
By contrast, a careful decision by the New Jersey Appellate Division in McDarby v. Merck reached the opposite preemption conclusion regarding Merck's COX-II inhibitor Vioxx. There, it was clear that the FDA had not decided against the existence of cardiovascular risks of Vioxx, and the FDA had not made a policy decision that Vioxx should be marketed unencumbered by warnings of cardiovascular risk, even though it also had not yet required a warning on Vioxx. (The reality that has emerged is that during the time many of the Vioxx plaintiffs took the drug, the FDA and Merck were engaged in negotiations as to the content and prominence of a cardiovascular warning). Under these circumstances, preemption on frustration grounds would be pure fantasy.
The Elephant in the Room: Vioxx
Ultimately, the Vioxx debacle is really the elephant in the room in Levine. There is a wealth of publicly-available information - including Congressional testimony by a high-up FDA insider (Dr. David Graham) and internal documents from Merck now posted on the Internet - that indicate the shocking dysfunctionality of the regulatory system in the case of Vioxx. According to the leading medical journal in the U.K., failures by both Merck and the FDA may have cost tens of thousands of lives.
In light of this evidence, this hardly seems like the right time in the history of drug regulation to cut off a major source of the accountability of drug companies to individuals - tort litigation - notwithstanding the serious imperfections of the tort system. And, to reiterate, there is no evidence that Congress has ever intended such immunization of drug companies. While there may be a great deal to say regarding the differential expertise between FDA and juries, and about the social costs of pharmaceutical litigation under varying state-wide standards, it is the legislative, not the executive or judicial branch, that gets to speak on that topic.
Surely Levine is A Hard Case - But Will It Also Make Bad Law?
In sum, for all of the reasons indicated here and in Part One of this series, Levine is a very hard case indeed. Despite our many twists and turns, we have made essentially three points in Part One and this column: (1) Levine, while tragic, was a very weak case for plaintiff that Wyeth should have won as a matter of state tort law; (2) Although the Vermont tort claim was weak, the federal preemption argument is also quite weak; the argument that the FDA made a policy decision about IV-push is not adequately supported by the record, leaving the Court with the principal options of affirming Levine's victory or remanding; (3) More generally, although a place exists for some form of frustration preemption, the combination of federalism concerns, deference to Congress, and the continuing need for an accountability backstop for the FDA counsels strongly in favor of constructing only a narrow space for implied preemption in the pharmaceutical context.
Because of the weakness of the plaintiff's case, we predict that the Court will be tempted to construct a powerful preemption doctrine in order to protect companies like Wyeth against ill-founded pharmaceutical actions. However, for the reasons we have given, the temptation to craft broad pro-defendant preemption doctrine should be strongly resisted.