American Journal of Law & Medicine

Drug preemption and the need to reform the FDA consultation process.


The scope of product liability for prescription drug injuries is a high-stakes matter. As the recent Vioxx litigation illustrates, drug manufacturers can face numerous lawsuits when users are injured by risks discovered only after the drug reaches the market. (1) The extent to which Food and Drug Administration (FDA) decisions preempt tort liability has emerged as a major issue. The agency itself has issued a policy that favors preemption when a drug manufacturer consults with the agency about a new risk. (2) The Supreme Court has found liability preempted for medical devices based on an express preemption provisions (3) and is expected to make an important ruling on drug preemption this term in Wyeth v. Levine. (4) Some believe that the Court will rule in favor of preemption even though there is no express statutory provision for drugs. (5)

The appropriateness of drug preemption has received extensive attention and divergent assessments in the academic scholarship with a mixture of some support (6) and criticism. (7) The scholarship has examined how federal courts have accepted the preemption defense to a greater degree than state courts, (8) and supported an agency reference model in court eases to identify the agency's position on preemption. (9) Other scholars have emphasized the need for a lower level of judicial deference, absent rulemaking, for agency preemption decisions that seek to displace the traditional ability of the states to protect their citizens through the availability of liability suits. (10) Another article maintains that industry needs to provide the FDA with "the kinds of information within its control that are the logical predicates for a preemption-worthy federal policy." (11) This paper will not directly address the general issues about preemption, but rather will emphasize the importance of the consultations between the agency and a drug sponsor about new risks in deciding on the appropriateness of preemption.

The most important and difficult preemption question is the liability for risks that are discovered by use after the drug enters the market. Wyeth concerned a serious risk posed by a particular method of administration of a drug, which was discovered during use. (12) The ease illustrates the drawbacks of the FDA's informal process for determining the need for revised warnings. One ground for preemption raised by the company was that it had proposed a revision and the response of the agency reviewer was to "[r]etain verbiage in current label," but the reason for the agency's response and the company's proposal was not stated. (13) The federal government now maintains in its brief in Wyeth that "if preemption turned on the details of the agency's deliberations," it could devolve into "an intrusive, and potentially inconclusive second-guessing of the agency's decisional process." (14) However, the important point is not the agency's internal deliberations, but the need to have an adequate explanation of the basis for the agency's decision at the time it is made if it leads to preemption.

This article maintains that reform in the agency's present consultation process is necessary. Change is needed to ensure that drug companies identify the new risk information that has led to a consultation, and that the agency issues an accessible decision containing its reasons for whether a change is needed. (15) This will promote better decision making. Further, if the Wyeth decision recognizes a preemptive effect, the need to improve the process in the future will be even more critical. If preemption is to apply to past consultations, the agency decision needs to identify the new risk information and reflect that the agency gave considered attention to the core points.

This article also examines the weaknesses of the FDA's new preemption policy with a special focus on its applicability to risks found after a drug is approved. A basic flaw with the agency's present preemption policy is the assertion that it is "squarely and solely" for the FDA to determine whether warnings should be added to the label. (16) The FDA's interpretation is at odds with the text and policy behind the agency's existing Changes Being Effected (CBE) rule, (17) which gives a manufacturer the ability and responsibility to add or strengthen warnings about drug risks without the agency's prior approval. Under the agency's preemption policy, preemption may also occur through inaction when a company has proposed the "substance" of a warning to the agency and the agency has not required it. (18) At a minimum, some form of disclosure about pending consultations regarding risk signals should be provided to doctors if the consultation affects preemption. (19) The CBE rule and the agency's policy should be reexamined to address these tensions.

Developing adequate disclosures about post-market risks presents special challenges. The initial testing for drugs is not sufficient to detect the risks that can occur when the drug is used by a larger population over a longer period of time. (20) As the FDA has long recognized, the post-market period is a continuation of the experimental testing on a drug. (21) The prospect of preemption increases the importance of implementing the recommendations in the study of drug safety by the Institute of Medicine (IOM) Report. (22) The IOM Report advocates a comprehensive agency review of the risks from new molecular entities within two to five years of approval and the need for disclosures to doctors and in consumer advertisements about the potential for added risks from these newly-approved drugs. (23) A considered reform of the FDA's consultation process should entail further examination of these recommendations.

An improved consultation process about emerging risks is still necessary, even though Congress has strengthened the agency's authority in the Food and Drug Administration Amendments Act (FDAAA). (24) In addition, the limits on the agency's resources to evaluate and respond to post-market risk must to be taken into account in assessing which circumstances may warrant preemption. (25)

To develop the basis for the recommendations urged here, Part II of this article describes the FDA regulation that allows companies to change the drug label and the agency's preemption policy that endorses preemption based on informal consultations with the manufacturer. Part III notes the recent United States Supreme Court decisions that impact drug preemption and focuses on the Wyeth case to identify its special issues and examines its illustration of the consultation process in practice. Part IV identifies reasons why new risks are found after a drug is approved and discusses the lessons from the IOM Report. It also addresses the experience with Vioxx about information and disclosures that may be needed regarding these new risks. (26) Part V examines the drawbacks of the FDA's existing preemption policy in light of this background and examines the reforms that ought to be considered in the consultation process as a policy matter and as a potential basis for preemption. Part VI concludes by acknowledging the drawbacks of preemption and the strengths of relying on tort relief. However, it also takes account of the limits of torts and the value of a process that facilitates an informed agency position on the need for new warnings.


This Part describes the key features of drug approval provisions and the CBE regulation that allows drug manufacturers to add and strengthen warnings. It also discusses an interpretation of that rule in the agency's preemption policy. This Part of the article provides important background for the following sections on the key Supreme Court preemption cases, the challenges of assessing new risks, and the preemption policy.


FDA's approval is needed to market a new drug and to change the labeling. (27) Manufacturers are also under a separate obligation not to have labeling that is misleading including with respect to materially misleading omissions. (28)

The FDA's regulation allows a drug manufacturer to "add or strengthen a warning" without prior FDA approval when the agency receives a supplemental application from the manufacturer providing for the labeling change. (29) In addition, under another provision in the agency's regulations, "[i]n accordance with [section] 314.70 ... the labeling must be revised to include a warning about a clinically significant hazard as soon as there is reasonable evidence of a causal association" but "a causal relationship need not have been definitely established." (30) In Wyeth, the state supreme court relied on the CBE regulation, which makes the pending case a test of its relevance for preemption. (31)


In 2006, the FDA published a policy statement on preemption that limited the significance of the CBE process. (32) According to the agency, a tort lawsuit "encourages and in fact requires lay judges and juries to second-guess the assessment of benefits versus risks of a specific drug." (33) The FDA's preemption statement maintained that warnings could also affect the choices available to doctors since the prospect of adverse jury verdicts could lead manufacturers to propose defensive labeling, "which if implemented could result in ... underutilization of beneficial treatments." (34) To avoid liability, manufacturers might add warnings that overstate the risk and discourage beneficial uses. (35) In the agency's view, "the determination whether labeling revisions are necessary is, in the end, squarely and solely FDA's under the act." (36) Moreover, federal law establishes both a floor and a ceiling for any additional statements. While manufacturers "may" make changes, "in practice [they] typically consult with FDA" before using the CBE procedure "to avoid making changes the agency might disagree with which could lead to enforcement action" on the grounds that the change was misleading. (37) The agency maintains that liability claims based on a failure to warn should be preempted, including those in which "the substance" of the warning had been "proposed" to FDA for inclusion in the labeling and was not required by FDA at the relevant time. (38)


There is controversy over the level of judicial deference that the agency's policy statement should receive. (39) The FDA proposed the rule to codify its position with the apparent aim of ensuring that its preemption policy receives greater deference. (40) The proposed rule also maintains that it confirms the agency's longstanding interpretation that the CBE rule applies only to "newly discovered" risks. (41) However, the CBE rule refers not only to adding, but also to strengthening a warning. (42) This indicates that it covers not only a new type of risk, but also a newly found increase in the prevalence of a risk already warned about in the label. …

Log in to your account to read this article – and millions more.