Photo of Eric Alexander

When we have given talks on preemption, whether to law firm personnel, attendees of professional conferences, or new FDA employees, we have tended to start with the Supremacy Clause and then break up the types of preemption and the issues related to them into different buckets.  We have not had to name those buckets or make tough calls about what issues go into which buckets because the categories have been pretty clear for a long time.  There is express preemption, which exists when there is a preemption provision in the federal law.  Those provisions often flag state requirements that are “in addition to or different from” the federal requirements.  There is also implied preemption, which breaks up into field preemption and conflict preemption—which itself breaks up into “impossibility” preemption and “purposes and objectives” preemption.  Riegel and Lohr are express preemption cases and Buckman, Levine, and Albrecht are implied preemption cases.  The interplay of express and implied preemption in product liability litigation over PMA devices has led to the invention of the near-mythical “parallel claim” monster.  We have suffered not so silently countless times when we have seen courts’ confusion over the types of preemption lead to gaffes like applying a presumption against preemption where it does not belong.

This dynamic is not limited to drug and device litigation and the FDCA.  There has been plenty of litigation involving pesticides and FIFRA, and it has produced a bunch of preemption decisions.  The most famous of these is probably Bates v. Dow Agrosciences LLC, 544 U.S. 431 (2005), which involved an herbicide—FIFRA treats weeds as pests.  While we do not purport to be FIFRA experts, we have written a number of times about the preemption issues posed by the express preemption clause in FIFRA, which mirrors the one in the FDCA for PMA devices.  Over the last few years, litigation over glyphosate-containing products and whether they cause cancer has been almost inescapable.  At least the lawyer advertising has been.  And it had not escaped our attention, either.  In 2019, we noted the clear rejection by EPA—which is responsible for enforcing FIFRA—of a link between glyphosate and cancer and wondered what effect it would have on litigation.  Over the next two years, we reported on an intermediate California appellate court decision and a Ninth Circuit decision that went plaintiffs’ way on the lack of preemption, despite EPA’s position.  Part of what carried the day in those cases was the argument that EPA’s actions lacked “force of law” and, thus, were not eligible to preempt any state claims.

Although we do not always use the fancy phrase “force of law,” the concept plays out in many drug and device preemption fights.  An action urged by a product liability plaintiff, such as a change in the labeling or the design of the product, was only impossible for the manufacturer if FDA really required something to the contrary.  A non-binding guidance or a mere urging from a pre-NDA meeting will probably not be enough.  By contrast, an approval letter that says selling the product with any changes to its approved specifications or labeling may create liability under the FDCA for adulteration or misbranding should be enough to measure against plaintiff’s theory of liability.  Impossibility is not part of the analysis of express preemption.

In Carson v. Monsanto Co., __ F.4th __, 2023 WL 4418639 (11th Cir. July 10, 2023) (en banc), the full court re-examined a panel’s decision that warnings claims under Georgia law were not preempted by EPA’s actions and requirements because none of them had “the force of law.”  There was a spirited discussion in the majority opinion, a concurring opinion, and a dissenting opinion on whether there was a live controversy for the court to decide, but we will not address that here.  Instead, the court’s consideration of the legal issue of whether “force of law” is part of the analysis for express preemption is what got our attention.  Rather than decide the application to the case at hand—or allegedly at hand—the en banc court addressed the legal issue and remanded the application to the original panel.

As one might expect with a purely legal decision, this came down to basic principles.  Force of law is part of the inquiry for implied preemption because a conflict with “a federal agency rule that does not have force of law is not the type of conflict between state and federal legal obligations that the Supremacy Clause addresses.”  Id. at *4.  FIFRA, however, contains an express preemption provision and that, rather than a conflict, is what “triggers preemption.”  Id.  “Within the limits of its enumerated constitutional powers, see U.S. Const. art. I, §§ 8-9, Congress may define the body of law that an express provision preempts.”  Id.  That is that.  The evaluation of express preemption thus comes down to “the ordinary principles of statutory interpretation.”  Id.

The application, as to which the Carson court expressed no opinion, seems pretty straightforward to us based on something the court did not state but maybe assumed.  By requiring submission of labeling in connection with registering a pesticide, FIFRA presumably sets out penalties for the sale of a registered product that has non-compliant labeling.  That should make it easy to find that state law liability for a label that does not mention a purported cancer risk would entail a requirement that is “in addition to or different from” the label that EPA/FIFRA required.  For failure to warn claims about medical products with an applicable express preemption provision under the FDCA, it does not get more complicated.  Whether FDA would have accepted a change to the label, as through a CBE submission with a branded prescription drug, is not part of the express preemption analysis.  That is an implied preemption issue.  A different bucket.