I saw the sign…and the answer is no—FDA-approved labeling apparently is not enough under state failure-to-warn laws, according to certain courts. While it has been long established that FDA-approved or mandated labeling preempts state failure-to-warn claims, some courts have decided that sometimes labeling just isn’t enough, never mind what FDA thinks. FDA approves (or sometimes dictates depending on the product and regulatory pathway to market) the content of drug product labeling to set forth the conditions of use for which the benefits of the product outweigh the risks. While the approval of that labeling is squarely on the shoulders of FDA, state failure-to-warn laws are used by plaintiffs to ask courts to pass judgment on the adequacy of labeling notwithstanding the fact that the scientific experts at the FDA have signed off on the labeling. This begs the question of whether these courts, and other courts looking at other state law claims, are actually undermining FDA’s authority to determine whether a drug can be used safely and effectively under its proposed conditions of use. After all, the labeling is intended to do exactly that: it informs patients and prescribers of the risks to ensure the product is used effectively and in a manner that mitigates risks, according to FDA’s understanding of the relevant science.
Brand drugs, generic drugs, and medical devices alike have all been the target of state failure-to-warn litigation; in a recent case, OTC acetaminophen is the target. For OTC products, two pathways to marketing exist: through a full New Drug Application or Abbreviated New Drug Application, requiring FDA approval (including the approval of the label), or under an OTC monograph, which basically serves as an instruction manual for making and labeling certain OTC drugs. Relevant to our conversation, OTC monographs establish conditions—such as active ingredients and indications for use, and product labeling—under which an OTC drug is generally recognized as safe and effective and does not require a product-specific approval. Most OTC drugs, particularly drugs that have been around for a while, are primarily marketed pursuant to OTC monographs. As explained in a 2020 CRS Report and a 2020 GAO Report, until the 2020 passage of the CARES Act (see our blog post here), amending a monograph was challenging because “the agency’s ability to update and finalize monographs in response to safety issues and to reflect new scientific information was limited by the rulemaking process the agency was required to follow, as well as insufficient resources. Agency officials estimated that it took at least 6 years to complete the required rulemaking process.” More than a few have taken far longer—literally decades. The GAO Report further explained that the agency did not have the resources to regulate the estimated 100,000 OTC drugs marketed through the monograph process. (Recall that the funds provided under the Prescription Drug User Fee Acts (PDUFA) are not available to support the agency’s efforts related to monograph drugs.) FDA did, however, try to identify safety issues through the medical literature and (after 2007 when mandatory safety reporting for OTC human drug products marketed without an approved application became effective) through the adverse event reporting system. Further, FDA has utilized a variety of strategies to call attention to safety issues through Dear Doctor letters or other forms of communication, even if directing changes to labeling is difficult. The passage of the OTC monograph provisions of the 2020 CARES Act was intended to address the lack of resources, and user fees are being collected, but, as expected, implementing all the other aspects of the new law is a long process.
FDA’s jurisdiction over OTC monograph drug labels is clear, but some courts seem to believe that compliance with the label requirements set forth in the relevant monograph is not only not sufficient, but also is discretionary, implicitly chipping away at the authority of the agency. A recent state law failure-to-warn case in the SDNY makes that very point. There, Plaintiffs (a mother individually and on behalf of her child) alleged that Defendants Johnson & Johnson and Walmart failed to include a warning on acetaminophen products they sold pursuant to the relevant OTC monograph, as well as under NDAs, that use of acetaminophen while pregnant causes children to be born with autism spectrum disorder and ADHD. Plaintiffs also alleged strict liability for design defect due to inadequate warnings and precautions and negligence and breach of consumer protection laws. Plaintiffs alleged that Defendants had a duty under state law to warn consumers of a (potential) risk of prenatal exposure to acetaminophen. Interestingly, the Plaintiffs alleged that this duty existed when the mother was pregnant in 2015 even though the studies relied upon in the litigation are from 2021. But we digress.
Defendants moved to dismiss on grounds that state failure-to-warn claims are preempted, as FDA approved the NDA-product labeling and dictated the labeling for OTC monograph products, but the Court disagreed. It’s important to note here that the warnings and precautions listed in the acetaminophen labeling complied with FDA’s Pregnancy Rule (which applies to all OTC systemic products), stating “if pregnant or breast-feeding ask a health professional before use,” and with the acetaminophen OTC Monograph and the relevant NDAs. Thus, Defendants argued that FDA already required the warning for pregnancy, rendering the decision to use it during pregnancy to a healthcare professional, and that FDA had considered the risks of in utero exposure and rejected the addition of those risks to the labeling. Nevertheless, despite the immense hurdles to changing a warning in a monograph (as specifically noted in the GAO Report), despite that FDA expressly rejected the change to the monograph at the time the Plaintiff-mother was pregnant in 2015, and despite the clear and specific language of the Pregnancy Warning Rule, the Court denied the Motion to Dismiss, holding that the failure-to-warn claims were not preempted.
Though the case addressed both NDA and monograph OTC products, it is notable as one of the first to address the issue of preemption with respect to the Pregnancy Warning Rule or to OTC Monographs. On these issues, the Court took the position that neither expressly forbade including additional labeling, and thus, under both of these FDA schemes, the presumption is that sponsors can add stronger warnings to product labeling without FDA concurrence; consequently, the Court concluded it was not impossible to follow federal law and provide appropriate warnings under state protection laws. Specifically, the Court explained, because “[t]he Pregnancy Warning Regulation simply does not speak to whether a further warning related to a drug’s use during pregnancy can be added to the general Pregnancy Warning on a drug label, whether added by the FDA or added by a manufacturer,” the ”regulation does not, therefore, preempt state law . . . .” Similarly, the Court stated, “FDA’s silence regarding acetaminophen-specific warning does not preempt state law.” Thus, the Court presumed, there is no reason that the sponsors could not add stronger pregnancy warnings.
And, as the Supreme Court previously explained, the Court noted, if the Defendants could have unilaterally strengthened warnings on labeling without prior approval from FDA, state failure-to-warn claims are not preempted. Here, the Court presumed that Defendants could do just that: absent an attempt by Defendants to change the labeling and subsequent FDA rejection, the Court decided that Defendants could have added truthful warnings about the risk of in utero exposure without violating federal law. Indeed, the Court explained, it could hardly be otherwise, because if a sponsor is “not free to add truthful warnings related to the use of acetaminophen during pregnancy, then that would mean that all manufacturers of OTC monograph-approved drugs intended for systemic absorption would be similarly barred from adding to their labels truthful warnings about the use of their own drugs during pregnancy.” But what the Court didn’t consider is that, if sponsors of OTC Monograph drugs could change their labeling to include more warnings without FDA buy-in, identical products could have very different labeling, which is not contemplated under the OTC Monograph scheme. While OTC Monographs may no longer include exact wording requirements for some information, this is only because, as the Court reminds us, “there can be various ways of accurately stating the same thing;” the problem arises when there are substantive changes to the labeling, like the addition of new warnings, such that all OTC Monograph products for a given drug don’t actually state the same thing even when there is no actual difference in risk between the products. This can lead to consumer confusion with consumers believing that there are actual differences. This is precisely the type of confusion FDA strives to avoid.
The Court’s determination that, even though FDA fully dictated the labeling for OTC Monograph products, Defendants could lawfully change the labeling undermines both FDA’s attempt to ensure consistency in the OTC drug market, as well as FDA’s decision that a healthcare professional should be consulted before use of acetaminophen in pregnancy; instead, it places the onus on manufacturers to basically beg FDA to modify monographs. This is because the presumption that manufacturers can unilaterally make substantive changes to the labeling of an OTC Monograph drug product is false. Just like it was over a decade ago in Pliva v. Mensing in the case of generic drugs, where the Supreme Court held that failure-to-warn claims for generic drugs were preempted precisely because the Reference Listed Drug holder—and FDA—controlled whether the generic sponsors could add new warnings to the labeling; in other words, generic drug sponsors could not just change labeling to add new safety warnings. And like the Court said in Pliva v. Mensing, “even if [the sponsors] had fulfilled [a] federal duty to ask for FDA assistance, they would not have satisfied the requirements of state law.” That requirement would only be fulfilled if FDA agreed to such a change. And indeed, after Pliva v. Mensing, there was much discussion about adding authority to the FDC Act to allow generics to make safety changes to product labeling without FDA approval. FDA went so far as to issue a proposed rule (see our blog post here), that ultimately went nowhere. And that such a proposal ultimately fizzled out is not surprising considering the issues with overwarning and the relatively limited profits made by generic companies when compared to the additional liability burden created by the obligation to make safety changes separately from the Reference Listed Drug product. The same considerations ring true for OTC drugs.
FDA, for years, has been concerned with overwarning. As FDA explained in a 2008 Rulemaking, FDA’s existing policies are “intended to ensure that scientifically valid and appropriately worded warnings will be provided in the approved labeling for medical products, and to prevent overwarning, which may deter appropriate use of medical products, or overshadow more important warnings.” In the Proposed Rule, FDA explained further that “[e]xaggeration of risk, or inclusion of speculative or hypothetical risks, could discourage appropriate use of a beneficial drug, biologic, or medical device or decrease the usefulness and accessibility of important information by diluting or obscuring it. As FDA has stated, labeling that includes theoretical hazards not well-grounded in scientific evidence can cause meaningful risk information to lose its significance.” In a 2017 Request for Information, the agency similarly noted that overwarning, or the concept that “individuals are exposed to so many warnings in the course of daily life that they are less likely to pay attention to any one particular warning,” “may lead consumers to discount all risks, or miss the most important risk information.” OPDP therefore initiated a study of “how repetition and overwarning apply to the presentation of risks in promotional prescription drug print pieces.”
If those marketing OTC drugs under the monograph were required to add new warnings because some literature noted a risk, the chances of overwarning are high, rendering the existing risk information potentially less impactful. It also raises questions about how much medical literature is required before a new warning is required, does it matter where it comes from (reddit or NEJM), or what types or quality of evidence are presented, and what if other medical literature disagrees? And, as in the acetaminophen case, what if FDA has already considered and rejected the need for a specific warning?
Notably, FDA had even addressed the strength of the evidence for acetaminophen warnings. Rather than adding in every potential risk in the case of acetaminophen, in that 2015 Drug Safety Communication, FDA specifically recommended that pregnant women consult their doctors before taking acetaminophen:
Pregnant women should always consult with their health care professional before taking any prescription or OTC medicine. Women taking pain medicines who are considering becoming pregnant should also consult with their health care professionals to discuss the risks and benefits of pain medicine use. Health care professionals should continue to follow the recommendations in the drug labels when prescribing pain medicines to pregnant patients.
FDA, in other words, directed, and continues to direct, pregnant women to a learned intermediary. According to FDA, therefore, use in pregnant women is neither condoned nor directed by the labeling—that’s a question for them to discuss with their doctors. The Court acknowledged all this, but nonetheless found that the failure-to-warn claims were not preempted because, essentially, the Defendants should have at least tried to change the labeling.
Unlike prescription drugs with approved NDAs, OTC monograph drugs generally are not blockbuster drugs protected by patents for which sponsors expect huge returns. FDA has estimated that there are more than 100,000 OTC drugs marketed through the monograph process, with 800 active ingredients for over 1,400 uses as of December 2019. With all that competition, particularly with respect to drugs as well-known as acetaminophen, it’s unlikely that there’s a lot of money to be made by an individual company; while the acetaminophen market may be large, there are just so many market participants. If every company that brings an OTC Monograph drug to market is hit with state failure-to-warn claims and forced to litigate such claims to completion, the market is sure to shrink. For many brands, the potential return is too small when compared to the potential for immense products liability claims for labeling that the sponsor cannot even control. Liability is precisely why the vaccine market shrunk so significantly in the 1980s, going from 26 manufacturers to 4, after lawsuits claiming $3.5 billion in damages were brought between 1980 and 1986. Things got so bad that Congress acted to shield vaccine manufacturers from liability in order to ensure sufficient availability of vaccines. There’s no reason to believe that continued state liability for FDA-mandated labeling would result in any different outcome where there is unfettered opportunity to bring state law claims against OTC monograph sponsors for labeling products in accordance with FDA requirements.
Nevertheless, court decisions that contrast with FDA’s positions are imposed on drug manufacturers of all sizes and types. Take, for example, sunscreens. In 2023, a Northern District of California Court determined that state law claims alleging violations of unfair competition and other consumer protection provisions could stand notwithstanding the fact that cosmetic foundation with sunscreen complied with FDA’s labeling requirements for sunscreens. There, the Court let proceed Plaintiffs’ suit alleging that foundation including sunscreen, advertised with a 24- or 25-hour duration, misleadingly suggested that sun protection could last for 24 or 25 hours, even though the directions stated clearly and unambiguously—and compliant with FDA’s sunscreen labeling rules—that the product must be reapplied every 2 hours for sunscreen coverage. Similarly, the District Court of Connecticut found that allegations of various state consumer protection laws for failure to list benzene as an ingredient in Banana Boat sunscreen were not preempted when nothing in FDA regulations suggests that benzene, as an impurity, is an “ingredient;” additional consumer protection claims about benzene also were not preempted because the Court found that nothing in the OTC sunscreen monograph precluded a warning about benzene.
To be fair, we note the many examples of cases where courts have found such claims preempted. But cases like the ones described here create precedent for the next OTC monograph case. In turn, each of these cases triggers concerns about how far courts are willing to go.
