October 2019, Volume XXXIII, No 7

Medicine and the Law

The Supreme Court’s Albrecht decision

Unclear implications for adverse reaction warnings

en years ago, the U.S. Supreme Court held that a drug manufacturer would not be liable under a state-based failure-to-warn theory if “clear evidence” showed that the FDA would not have approved the patient-plaintiff’s preferred product warning label. But in the years since that decision came down, lower courts have struggled to consistently determine what meets the requisite “clear evidence” threshold. So last summer, many in the drug and device industry cheered when the High Court agreed to review the Third Circuit’s controversial In re Fosamax (Alendronate Sodium) Prod. Liab. Litig. decision, as the case provided a rare opportunity for the Court to clarify its prior pronouncement and reshape a heavily litigated area of the law. Unfortunately, however, the Court’s resultant ruling in Merck Sharp & Dohme Corp. v. Doris Albrecht, et al. (May 2019) failed to provide the degree of clarity that most industry observers had hoped for.


The Albrecht litigation involves Fosamax, a drug widely used to treat osteoporosis in postmenopausal women. Fosamax combats the effects of osteoporosis—the development of weak or brittle bones due to a progressive loss of bone cells—by slowing a patient’s rate of bone cell loss, allowing the body time to replace lost bone cells, thereby decreasing the risk of osteoporotic fractures. But like all medications, taking Fosamax is not without risk. Indeed, one potential risk—and the one at issue in Albrecht—is that long-term Fosamax users run the risk of developing an atypical femoral fracture. Though present-day prescribers are (hopefully) aware of this risk, such was not always the case.

The drug’s label did not warn of the then-speculative risk of an atypical femoral fracture.

FDA first approved Fosamax in 1995. At the time of its approval, the drug’s label did not warn of the then-speculative risk of an atypical femoral fracture. But once on the market, evidence began to accumulate that some long-term Fosamax users were experiencing such fractures. As a result, in 2008, Merck applied to FDA—the federal agency charged with regulating drug labels—for preapproval to add certain language to the Adverse Reactions and Precautions sections of the Fosamax label.

Specifically, Merck sought to include reference to the risk of a “low-energy femoral shaft fracture” in the Adverse Reactions section and a separate discussion concerning the risk of “stress fractures” in the Precautions section. FDA approved Merck’s proposed Adverse Reactions language but rejected the proposed Precautions language, asserting that Merck’s justification for the proposed Precautions language was “inadequate.” In fact, the case record showed that FDA representatives told Merck to “hold off” on the Precautions language and that the agency “would then work with . . . Merck to decide on language for a [Precautions] atypical fracture language, if it is warranted.” After receiving this feedback, Merck added the requested language to Fosamax’s Adverse Reactions section in 2010 but made no corresponding changes to the label’s Precautions section. The following year, after further communications with FDA, Merck revised its Precautions section to include language mentioning Atypical Subtrochanteric and Diaphyseal Femoral Fractures.

The plaintiffs (more than 500 individuals who took Fosamax and suffered atypical femoral fractures between 1999 and 2010) sued Merck, asserting that the company had failed to warn them about the risk of atypical femoral fractures. Merck, relying on its prior communications with FDA, sought dismissal of such claims, arguing that “clear evidence” demonstrated that prior to 2011 FDA would not—and did not—approve of such a warning. The implication being that the company should be immunized from such a claim because it was legally impossible for Merck to comply with state tort law without violating the Constitution’s Supremacy Clause, which holds that in the event of a direct conflict between federal and state law, federal law takes precedence.

Though the District Court agreed with Merck, the Third Circuit did not, notably holding that: (1) to satisfy the “clear evidence” standard, Merck needed—but failed—to show that it was “highly probable that the FDA would not have approved a change to the drug’s label”; and (2) whether FDA would have rejected Merck’s proposed label change was a fact question for the jury. Merck’s appeal thereafter followed.

To Albrecht and beyond!

Many people have the mistaken view that all Supreme Court decisions are contentious and narrowly decided by a razor thin margin, but the reality is that in most cases the justices widely agree on the outcome. Albrecht is a good example. Though the justices did not uniformly join the majority opinion, all nine of them agreed to vacate the Third Circuit’s decision. In so doing, the Court rejected the Third Circuit’s attempt to characterize or define “clear evidence” in terms of evidentiary standards and further rejected the contention that a layperson jury—as opposed to a judge—was best situated to determine whether FDA would (or did) reject a proposed label change. Interestingly, the Court chose not to decide the ultimate question of whether Merck was liable, electing instead to remand the matter for further proceedings.

Taken together, the Albrecht decision amounts to a mixed bag. One the one hand, the decision may result in a modest decrease in existing product liability suits because, by vacating the Third Circuit’s decision, the High Court erased a key authority used by plaintiffs to survive dismissal. Moreover, the decision should force otherwise reluctant judges to decide a potentially dispositive, purely legal issue as opposed to punting the matter to a jury. But on the other hand, the decision amounts to a missed opportunity.

The decision amounts to a missed opportunity.

For years, lower courts have struggled to consistently determine the meaning of “clear evidence.” And unfortunately, with the exception that the Court rejected the Third Circuit’s attempt to equate “clear evidence” with an existing evidentiary standard (e.g. “preponderance of the evidence” or “clear and convincing evidence”), the opinion made no effort to define an otherwise undefined term. Instead, the Court’s guidance essentially amounted to characterizing “clear evidence” in “I know it when I see it” terms, which is akin to no guidance at all.

As a result, as was the case before Albrecht, lower courts will likely need to determine what satisfies the “clear evidence” standard on a case-by-case basis. And doing so, as one lower court has remarked, will likely require courts to evaluate several factors, including but not limited to “the regulatory history of the drug or drug class at issue, temporal gaps between FDA action and accrual of a plaintiff’s claims, citizen petition submissions and rejections, available scientific data, and whether the FDA has reviewed the particular harm at issue and the consistency of any resulting conclusions.” Needless to say, such an individualized, multifactorial analysis will inherently involve some degree of subjectivity, which may yield inconsistent or haphazard results.

Two additional points bear mentioning. First, Albrecht continues to underscore the difficulty drug manufacturers face in asserting that a state-based failure-to-warn claim is precluded where the FDA may have considered—but did not adopt—the proposed label change. Indeed, the Court pointedly noted that such claims are only barred if the drug manufacturer, who “bears responsibility for the content of its label at all times,” is prohibited by federal law from adding the requested warning, and that under the FDA’s Changes Being Effected (CBE) regulation a drug manufacturer may, without prior approval, unilaterally change a label “’to reflect newly acquired information’ if the changes ‘add or strengthen a ... warning’ for which there is ‘evidence of a causal association[.]’” The implication being that a failure-to-warn claim may only be barred if a drug manufacturer tries to make a label change pursuant to the CBE regulation, but is then thwarted from doing so by the FDA.

Second, though the issue was neither before the Court nor decided, the opinion appears to suggest that only formal FDA actions, or similar agency actions carrying the force of law, matter to this analysis. That is to say, informal communications (e.g., calls or emails) between FDA and a drug manufacturer regarding a proposed label change may not be of significance. Should that indeed be the case, it would be quite notable, not only due to its potentially chilling impact to Albrecht on remand, but also because it would contradict some existing lower court views. It may very well be the case that this issue—the “question of disapproval ‘method’”—will mark the next chapter of this ongoing saga.

Elie Biel, JD, is an attorney at Faegre Baker Daniels LLP, where his practice is devoted to drug and device litigation and the defense of health care systems and professionals in professional liability matters. 


PO Box 6674, Minneapolis, MN 55406

(612) 728-8600


follow us

© Minnesota Physician Publishing · All Rights Reserved. 2019


Elie Biel, JD, is an attorney at Faegre Baker Daniels LLP, where his practice is devoted to drug and device litigation and the defense of health care systems and professionals in professional liability matters.