Photo of Rachel B. Weil

We just returned from our annual “girl trip” to a lovely spa in the mountains of Pennsylvania. And we stared our decrepitude full in the face. For the first time, we set off enthusiastically on a bike ride (cool “fat tire” bikes we’d never seen before) and found ourselves falling behind on hills and struggling to catch up with the 25-year-olds who composed the rest of the group. Don’t get us wrong – we rowed the hell out of the rowing machines and acquitted ourselves nicely in boxing class. But we were forced to admit that the legs and knees, challenged by tough terrain and compared to those belonging to others with much fresher faces, just didn’t work the way they were supposed to anymore.

So we are heartened to present a case in which implied conflict preemption—“warnings preemption” under Wyeth v. Levine—worked just the way it was supposed to. In Maze v. Bayer Health Care Pharm., Inc., 2019 WL 1062387 (E.D. Tenn. Mar. 6, 2019), the plaintiff alleged that the defendant’s contraceptive pill caused her to suffer a stroke. She asserted the usual warnings claims, alleging that the defendant’s warnings of the risk of thromboembolism were inadequate, though admitting in the same breath that the defendant “devoted substantial space to warning of the risk of venous thromboembolism.”   Maze, 2019 WL 1062387 at *1. The defendant moved to dismiss, arguing that all of the plaintiff’s claims were preempted.

The court explained that, under Levine, state law can still hold a manufacturer liable for failing to update a prescription drug label with “newly acquired [risk] information or even a new analysis of previously submitted data.” Id. (citation to Levine omitted). But changes “not based on new information—and thus necessarily based on the same information already considered by the FDA—require prior FDA approval.” Id. (citation omitted). That is what initial FDA approval is all about. So federal law preempts any claim that the manufacturer should unilaterally have changed a label based on information the FDA had at the time it approved the label. As the court explained, this “lets the FDA be the exclusive judge of safety and efficacy based on information available at the commencement of marketing, while allowing states to reach contrary conclusions when new information not considered by the FDA develops.” Id., quoting In re Celexa & Lexapro Mktg. & Sales Practices Litig., 779 F.2d 34, 41 (1st Cir. 2015).

Under that reasoning, the court continued, “[i]t follows . . . that any claim asserted by [the plaintiff] and based on information known to the FDA as of April 2012—when the label at issue here was approved—is plainly preempted by federal law.” In her complaint, the plaintiff alleged that the defendant was aware, at the time the label at issue was created, that the medication increased users’ stroke risk by 500% (the defendant disputed this fact), and did not set forth “any newly acquired information or even new analyses of previously submitted data,” id. at *3 (internal punctuation and citation omitted). She argued, in her motion, that such information existed, but, the court held, her “blanket statements [were] merely conclusory allegations and [did] not count as a sufficient showing that the plaintiff is entitled to relief.” Id. (internal punctuation and citation omitted).

The court continued,

Finally, and what seals the deal here, is that the label in effect during the entire relevant time repeatedly warned about the risk of stroke. Thus, even assuming the science marshaled by [the plaintiff] is newly acquired and says anything at all about the risk of stroke (as stated above, it is not and does not), nothing indicates that the risk is any higher than what is reflected in the . . . label that the FDA approved in 2012.

Id. In Levine, in contrast, the “main question [was] whether a nonexistent warning should be included in the label at all,” not whether “an FDA-approved, repeated stroke warning that the . . . label has included all along” is adequate. Id. “The former,” the court stated, “is heads-or-tails, the latter involves matter-of-degree questioning with respect to something that was clearly known and considered by the FDA in 2012.” The court concluded, “Without some indication that the science available now is somehow different, the Court will not second guess the adequacy of [the] warning label with state tort law.” Id.   And because the manufacturer could not unilaterally change the label’s stroke warnings, any claim the plaintiff might have had under state law was preempted by federal law.   Case dismissed.

This is how preemption is supposed to work, in the hands of the still-too-rare judge who is not afraid to dismiss a tort suit on preemption grounds. We like this decision, and we will continue to keep you posted on the good, the bad, and the very bad in the world of preemption jurisprudence.