October 12, 2001

Comments Urging FDA to Deny Drug Industry Liability Disclaimers

Dockets Management Branch
Mailstop HFA-305
Food and Drug Administration
5630 Fishers Lane, Room 1061
Rockville, MD 20852

Re: Withdrawal of PPA, Docket No. 01N-0196: Comment

To Whom It May Concern:

On August 14, 2001, the FDA issued a notice of a Proposal to Withdraw Approval of New Drug Applications and Abbreviated New Drug Applications for products containing phenylpropanolamine (PPA) because of safety concerns documented in the Yale Hemorrhagic Stroke Project (HSP). [1],[2] That notice gave interested parties the right to request a hearing on the proposed withdrawal of PPA-containing products. In response, three manufacturers of PPA-containing products, Novartis Consumer Health (Novartis), American Home Products Corporation (AHPC), and Schering-Plough Health Care Products (Schering), submitted letters stating that they did not request a hearing. The letters did request, however, that the FDA include in its final rule withdrawing approval of products containing PPA a disclaimer regarding industry liability. [3],[4],[5] The requested disclaimer would state that the withdrawal of PPA-containing drugs does not mean that the drugs were marketed negligently; this disclaimer is intended to gain some sort of protection for the companies from product liability suits.

We urge the FDA to reject the companies request. The disclaimer sought by Novartis, AHPC, and Schering runs counter to the FDA s longstanding position that the agency should remain neutral in state-law liability matters. Moreover, the requests for a disclaimer distort prior FDA action and the facts concerning the dangers of PPA-containing products. For these reasons, as discussed in more detail below, the companies disclaimer request should be denied.

Notably absent from the Novartis, AHPC, and Schering letters is any support for the proposition that the FDA has authority to make or influence state-law liability determinations. In fact, the FDA has no such authority. When Congress was considering the legislation that ultimately became the 1938 Food, Drug, and Cosmetic Act (FDCA), it specifically rejected a proposal to include in the Act a private right of action for damages caused by unsafe or faulty drugs.[6] This decision was based on Congress s recognition that the states liability laws were adequate to protect consumers.[7] As a result, the FDCA which is the sole source of the FDA s authority to approve or withdraw approval of a drug such as PPA does not give the FDA any role in connection with state product liability law. The FDA simply has no power to issue disclaimers for the purpose of influencing liability determinations in state-law damages actions.

Consistent with the FDCA and its legislative history, the FDA has previously rejected industry requests that the agency make statements intended to influence the outcome of product liability cases. For example, in 1998, the FDA rejected a pharmaceutical industry plea that the FDA make a statement intended to eliminate liability with regard to consumer medication guides because tort liability can not be a major consideration for FDA which must be guided by the basic principles and requirements of the act in its regulatory activities.[8] This statement echoed one made almost two decades before: It is not the intent of the FDA to influence the civil tort liability of the [drug] manufacturer.[9]

In their letters to the FDA, Novartis, AHPC, and Schering refer to statements the FDA made in a 1999 Federal Register commentary and assert that the same disclaimer language that they request now was adopted in that commentary.[10] However, the statement to which Novartis, AHPC, and Schering refer was not a disclaimer at all. In 1999, the FDA issued a final list of drugs not suitable for pharmacy compounding because their marketing approval had previously been withdrawn. The language cited by Novartis, AHPC, and Schering was part of the agency response to industry comments seeking a disclaimer, much as the companies are seeking here. That response, taken solely in isolation, could be read to support the companies position. However, in language the companies studiously avoid, the FDA in fact rejected the request to adopt language regarding the effect of its action on civil litigation:

This list [of pharmaceuticals not suitable for compounding] is not intended to be used as evidence in a product liability suit, and the addition of language designed to minimize the potential effect of the list in litigation is unnecessary to fulfill its intended purpose.[11]

Thus, read in context, the 1999 rule on which Novartis, AHPC, and Schering rely did not adopt a disclaimer to protect pharmaceutical companies from product liability suits. Rather, the FDA expressed, albeit inartfully, its longstanding position that it does not become involved in liability issues arising in private tort litigation. 

Moreover, Novartis wrongly asserts that the pharmacy-compounding statement was made part of the final action listing another drug product,[12] creating the mis-impression that the 1999 statement concerned a single drug or class of drugs. In fact, the statement was made in response to industry comments on a rule prohibiting pharmaceutical compounding of 59 previously withdrawn drugs. Whereas the FDA may be able to assess whether the manufacturers of a single drug class, such as PPA-containing drugs, have been negligent, it cannot be expected to investigate manufacturer conduct regarding several dozen different drug types withdrawn over a period of many years. Thus, the companies attempt to adopt language from the pharmaceutical compounding proceeding for use with PPA is inappropriate for this reason as well.

Novartis and Schering also err when they discuss the evidence regarding the dangers of PPA. Both companies assert that prior to the Yale [Hemorrhagic Stroke Project], there was no scientifically reliable evidence of an association between PPA and hemorrhagic stroke.[13],[14] These statements, if not outright false, are highly misleading. To begin with, as early as 1983, Public Citizen s Health Research Group expressed concern over the safety of PPA-containing products.[15] Moreover, the companies statements ignore epidemiological data available as early as 1991. In that year, the FDA published an epidemiological study of drug-related stroke case reports (from 1977 to January 1991) showing that PPA-containing drugs were the most common suspect in all spontaneous reports of cerebrovascular accidents in women aged 10-59 years.[16] That study also found that [s]troke followed the first dose in half of PPA-diet pill cases and in over three-fourths of PPA-cough/cold cases, suggesting a causal relationship.[17] The FDA solicited comments on its 1991 study from Steven Kittner, M.D., M.P.H., of the University of Maryland School of Medicine, Janet R. Daling, Ph.D., of the Fred Hutchinson Cancer Research Center at the University of Washington, and Jack P. Whisnant, M.D., of the Mayo Clinic. Dr. Kittner confirmed that the observation that most reported stroke cases follow the first dose [is] consistent with the hypothesis of a causal relationship between PPA and hemorrhagic stroke.[18] Furthermore, after reviewing the comments of Whisnant and Daling, Dr. Kittner concluded that they all agree[d] that the data does not permit the conclusion that PPA is safe.[19]

Accordingly, the industry cannot be absolved of liability on the style=ground that it had no reason to know of PPA s danger prior to the Yale HSP study. Where data linking PPA and stroke has existed for more than a decade, the question whether the industry s failure to act on this information by continuing to market PPA-containing drugs was negligent is a matter for the courts. The FDA should not inject itself into this legal question by issuing the disclaimer sought by Novartis, AHPC, and Schering.

Finally, the means by which Novartis, AHPC, and Schering are seeking a disclaimer is unlawful. The companies want a disclaimer because they believe it will have a substantive effect favorable to them in product liability litigation. Even if the agency had the authority to issue a rule with substantive effect on state product liability rules, it could only do so through notice-and-comment rulemaking or an adjudicatory hearing, in which the public was informed that the agency was considering such a rule and had the right to fully participate in the rulemaking or adjudicatory process.[20] For the FDA to promulgate a new substantive rule adopting a product liability disclaimer in response to a comment on a proposed final action for withdrawal of a class of drug products from the market would circumvent Congress s carefully constructed procedures for affording public participation in rulemaking and adjudication.

For all of these reasons, we urge the FDA to reject the requests by Novartis, AHPC, and Schering for a product liability disclaimer.

Sincerely,

Brian Wolfman
Attorney
Public Citizen s Litigation Group

Sidney Wolfe, M.D.
Director
Public Citizen s Health Research Group

Peter Lurie, M.D., M.P.H.
Deputy Director
Public Citizen s Health Research Group

Amer Ardati
Research Associate
Public Citizen s Health Research Group