Mifepristone, preemption, and public health federalism

Abstract

On June 24, 2022, the Supreme Court issued an opinion in which five justices voted to overturn Roe v Wade. Even before the final opinion issued, scholars and advocates had begun to consider legal strategies that might mitigate the decision’s anticipated harmful consequences. One such strategy involves challenging state restrictions on Food and Drug Administration (FDA)-approved pregnancy termination drugs on preemption grounds. This article begins by exploring how these challenges might fare—considering both drug-specific restrictions and complete bans on abortion—arguing that there are compelling legal grounds on which courts should conclude that many state restrictions are preempted. Importantly, although these state restrictions have arisen within a larger debate about reproductive health care, this is far from the only area in which states seek to regulate prescription drugs. States have long regulated drugs in ways that diverge from FDA, arguably increasingly so in recent years. Accordingly, the article investigates the implications that preemption challenges in the abortion context may have for other areas of state drug regulation, making the case that the benefits of public health federalism need not be undermined by successful preemption challenges in the abortion arena.

I. INTRODUCTION

On June 24, 2022, the Supreme Court issued an opinion in Dobbs v Jackson Women’s Health Organization in which five justices voted to overturn Roe v Wade and Planned Parenthood v Casey, eliminating a right to abortion grounded in the US Constitution.¹ The decision has had immediate, harmful consequences, which are disproportionately falling on marginalized people.² Emergency rooms in states newly prohibiting abortion care are denying or delaying life-saving care for people experiencing miscarriages or ectopic obstructions.³ In July, a 10-year-old rape survivor was denied an abortion in Ohio, requiring her to cross state lines for health care.⁴ Many states are denying millions of people the ability to make decisions to protect their health and well-being,⁵ with greater restrictions and other harms looming on the horizon.⁶

One legal strategy that has gained attention as a promising avenue to mitigate Dobbs’s consequences involves challenging state restrictions or bans on Food and Drug Administration (FDA)-approved pregnancy termination drugs as preempted by the Federal Food, Drug, and Cosmetic Act (FDCA).⁷ The same day that the Dobbs opinion issued, Attorney General Merrick Garland issued a statement in which he said ‘[T]he FDA has approved the use of the medication Mifepristone. States may not ban Mifepristone based on disagreement with the FDA’s expert judgment about its safety and efficacy’.⁸ Preemption also appears to be one focus of the ‘reproductive rights task force’ that the Department of Justice announced in July 2022.⁹ Even before the Dobbs opinion came out (and before a draft of the majority opinion was leaked in May 2022), GenBioPro, a manufacturer of a generic mifepristone product, had challenged Mississippi’s restrictions on the drug,¹⁰ and scholars—including two authors of this article—had considered preemption challenges to states’ various pre-Dobbs restrictions on mifepristone as a possible means to improve access to reproductive health care.¹¹

Against this background, this article considers the fate of potential preemption challenges to the new state restrictions on abortion care arising in the wake of Dobbs, including both drug-specific restrictions and complete bans on abortion care.¹² The courts will be the final arbiters. However, there are compelling legal arguments that support courts concluding many state laws limiting or banning access to mifepristone are preempted by FDA regulation.

Critically, any preemption challenges to state medication abortion laws will not happen in a vacuum. Although state mifepristone laws have arisen within the larger debate about reproductive health care, abortion is far from the only area in which states regulate prescription drugs. States have long regulated pharmaceuticals in ways distinct from FDA’s approach,¹³ and federal, state, and local variation may serve important public health goals.¹⁴ For instance, public health scholars have long viewed state tort law as supplementing FDA oversight by ‘providing an added safeguard in the case of drugs in the market’, thus ‘serving an important public health function’.¹⁵

There are valid concerns about the impacts of preemption challenges to state medication abortion laws on other areas of state regulation. If medication abortion preemption challenges prove unsuccessful, preemption defenses in tort lawsuits may be weakened, but doors may be opened to state bans on other FDA-approved products—for example, states may be interested in banning COVID-19 vaccines or drugs for HIV pre-exposure prophylaxis (PrEP).¹⁶ Successful preemption challenges, by contrast, may be used as fodder to strengthen preemption defenses in tort lawsuits, while limiting states’ ability to regulate the use of drugs in areas in which states historically have regulated quite extensively, as with the use of opioid analgesics. Ultimately, we argue that, although these potential for broader effects of preemption challenges to state laws governing abortion drugs are important to consider, preemption challenges need not be understood as risking beneficial public health federalism because FDA and state regulation of mifepristone may be distinguishable from other areas of intersecting federal and state pharmaceutical regulation.

To develop these arguments, the article first discusses current medical knowledge regarding mifepristone and explains how preserving access to mifepristone is one important step that can help mitigate Dobbs’s consequences. Next, the article explains that while FDA tightly regulates all prescription drugs, mifepristone has been subject to more federal scrutiny than most. The article then demonstrates why courts could—and should—understand federal law as preempting much state regulation of mifepristone, considering as examples state restrictions on access to mifepristone that fall short of a ban, a medication ban, and a total abortion ban of the kind that has followed the Dobbs decision in some states. Finally, the article considers the implications that preemption challenges to state mifepristone restrictions may have for the future of public health federalism beyond the context of abortion.

II. THE PROMISE OF MIFEPRISTONE ACCESS

In an Executive Order issued on July 8, 2022, President Biden identified access to medication abortion as one of five priorities in his administration’s plan to protect access to reproductive health care after Dobbs.¹⁷ Preserving access to medication abortion, especially where access to procedural abortion care is limited, holds promise and is rightfully a priority because mifepristone¹⁸—a drug approved and used for medication abortion—is safe, effective, and relatively easy for health care professionals and patients to use.¹⁹

Mifepristone is a progestin antagonist that FDA initially approved for medication abortion over 20 years ago.²⁰ It is used in combination with another drug, misoprostol, to induce abortion.²¹ This regimen is highly efficacious.²² In the two trials supporting approval of the drug, 97 per cent (n = 16,794) and 96 per cent (n = 18,425) of participants taking mifepristone and misoprostol had a complete medication abortion (with efficacy decreasing slightly over the course early pregnancy, from 98 per cent at 7 weeks to 93 per cent in the 10th week).²³ Mifepristone’s effectiveness—how well a drug works in the real world—is also well documented. One prospective observational study comparing patients who took 200 mg of mifepristone followed by 800 micrograms of misoprostol 24–48 hours later at 57–63 days versus 64–70 days of gestational age found that the effectiveness was 94 and 93 per cent, respectively.²⁴ Another study that analyzed 13,373 case reports between 2006 and 2011 for women who took the same regimen up to 63 days gestational age found an overall success rate of 98 per cent.²⁵

Mifepristone has also proven extremely safe. Although some women who take the drug experience minor side effects such as nausea, headache, and dizziness,²⁶ serious adverse reactions are rare. FDA reported that 0.03–0.5 per cent of women taking mifepristone, in the US clinical studies, required a transfusion, 0.2 per cent had sepsis, and 0.04–0.6 per cent had a related hospitalization. The frequency of these events was similar in non-US clinical studies: 0–0.1 per cent of women required a transfusion, <0.01 per cent had sepsis, and 0–0.7 per cent had a related hospitalization.²⁷ In mifepristone’s first 16 years on the US market (2000–16), only 19 deaths were reported to FDA out of more than 3 million women who had taken the drug, making the drug’s associated mortality rate 0.00063 per cent or 14 times lower than the background risk of pregnancy-related death.²⁸ Although FDA has long restricted the prescribing and dispensing of mifepristone, mifepristone use in countries without such restrictions has proven equally safe.²⁹

Beyond mifepristone’s safety and effectiveness, its use may be harder for states to police than procedural abortion and thus even more important for retaining access to necessary health care in the post-Dobbs era. As Professor Greer Donley explained, ‘Historically, abortion was done by procedures, which meant that, if you could control doctors, you could really control abortion provision. But now that abortion pills exist, a state’s abortion ban is not going to have the same effect as it used to’.³⁰ Even before Dobbs, mifepristone was key to reproductive health care. Many people preferred the privacy of medication abortion, and unlike with procedural abortion, non-physician clinicians can sometimes provide medication abortion.³¹ Already, medication abortion makes up over half of all abortions in the USA, and this proportion will almost certainly increase post-Dobbs.³²

III. EXTENSIVE FDA REGULATION

Since approving mifepristone in 2000, FDA has subjected mifepristone to more scrutiny and regulation than most other prescription drugs.³³ A safety program, established as a condition of mifepristone’s approval under Subpart H of 21 CFR Part 314,³⁴ restricted dispensing to physicians who attested to the ability to assess the duration of pregnancy, diagnose ectopic pregnancies, and provide surgical intervention or access to medical facilities able to provide blood transfusions in cases of incomplete abortion or severe bleeding.³⁵ The program further required patients to receive a medication guide and patients and physicians to sign agreement forms. These measures were intended to ensure awareness of possible rare adverse events from taking mifepristone and safe use conditions.³⁶ Finally, the program required in-person ingestion of the drug in the presence of the physician.

The federal approach to regulating mifepristone has evolved over the years. Mifepristone is now subject to a REMS with elements to assure safe use (ETASU). FDAAA amended the FDCA to give FDA the statutory power to require that manufacturers institute REMS for prescription drugs if necessary to ensure that the benefits of a drug outweigh its risks.³⁷ REMS can have various components, ranging from a medication guide and communication plan explaining risks and safe use practices to patients and physicians, respectively, to more complex ETASU, such as laboratory testing requirements, patient enrollment in a registry, or restricted distribution of the drug through specialty pharmacies.³⁸

In 2008, FDA issued a notice listing drugs, including mifepristone, that were deemed to have in effect an approved REMS, under section 909(b) of FDAAA, as a result of existing safe use conditions.³⁹ Danco Laboratories then submitted a proposed REMS for mifepristone, which FDA approved in 2011.⁴⁰ The initial REMS incorporated the original safe use conditions and also required three patient visits to receive mifepristone: the first to receive mifepristone from the physician, the second (on day 3) for the physician to determine whether the termination was complete and to provide misoprostol if it was not, and the third (on day 14) to confirm whether complete termination of the pregnancy occurred.⁴¹

In 2016, FDA approved a revised indication for mifepristone, extending its approved use from up to 7 weeks of pregnancy to up to10 weeks of pregnancy and changing the dosing regimen, in part to allow for greater flexibility for the timing of misoprostol.⁴² While reviewing the data provided to support the expanded indication—including clinical trials, observational studies, and systematic reviews—FDA also conducted a review of the REMS and authorized several changes.⁴³ First, the Medication Guide was removed as a component of the program.⁴⁴ Second, the requirement that mifepristone be dispensed by a physician who attested to certain qualifications was expanded to include non-physician clinicians (eg nurse practitioners).⁴⁵ Third, clinicians were no longer required to report all serious adverse events associated with the drug, such as hospitalizations and transfusions, but instead only deaths.⁴⁶ Finally, the number of required office visits was reduced to one. During this visit, patients received mifepristone, instructions to take misoprostol in the next 24–48 hours, and advice to follow-up between 7 and 14 days after treatment to confirm the complete termination of pregnancy.⁴⁷

In April 2019, nearly 20 years after mifepristone was first marketed in the USA, FDA approved a generic version of the drug manufactured by GenBioPro.⁴⁸ A single, shared REMS program was approved to cover both Mifeprex, Danco’s brand product, and generic products.⁴⁹ This shared system included no substantial changes to the REMS elements.

While the COVID-19 pandemic prompted FDA to relax in-person dispensing requirements for other drugs subject to REMS, FDA initially continued to enforce the in-person visit requirement for mifepristone.⁵⁰ The American College of Obstetricians and Gynecologists and several other parties filed a lawsuit in May 2020 to contest this policy, arguing that the in-person REMS requirement posed an undue burden to patients, substantially limiting access without providing patients with a significant benefit.⁵¹ In July 2020, the court preliminarily enjoined enforcement of the in-person requirement.⁵² Although the Supreme Court stayed the injunction,⁵³ in April 2021, FDA announced that it would continue to decline to enforce the in-person requirement during the pandemic based on its review of the safety of mifepristone use during the injunction.⁵⁴

In addition to the lawsuit that challenged the in-person requirement, in 2017, the American Civil Liberties Union (ACLU) and ACLU of Hawaii filed suit on behalf of a physician and several medical organizations, arguing that the entire REMS was unnecessary and that it violated substantive due process, equal protection rights, and the Administrative Procedure Act.⁵⁵ As a result of the ACLU lawsuit, in May 2021, FDA announced that it would review the entire REMS (not just the in-person requirement).⁵⁶ FDA subsequently announced, in December 2021, that the mifepristone REMS will no longer require that the drug be dispensed in-person through clinics, medical offices, and hospitals. Once the modifications to the REMS are approved, it will include only a certification requirement for dispensing pharmacies and prescribing clinicians and a requirement that patients receive counseling. President Biden’s Executive Order calling on HHS to take additional steps to protect access to abortion care may be cause for hope that FDA will consider additional reductions in the mifepristone REMS requirements or, eventually, a release of the REMS altogether.⁵⁷

IV. THE CASE FOR PREEMPTION OF STATE MIFEPRISTONE RESTRICTIONS

Because of mifepristone’s potential to help preserve access to safe and effective reproductive health care post-Dobbs, legal strategies to increase access to mifepristone are critical. This Part explores one such strategy that flows from the extensive FDA regulation of the drug: challenging state restrictions on mifepristone as preempted by FDA regulation.⁵⁸ Although states generally may regulate concurrently with the federal government, even in areas that are extensively federally regulated, as with pharmaceuticals,⁵⁹ the Supremacy Clause of the US Constitution provides that federal law ‘shall be the supreme Law of the land…’.⁶⁰ Thus, where federal law and state law conflict, federal law preempts state law—state law is ‘without effect’.⁶¹

To determine whether a federal law preempts a state law, ‘the purpose of Congress is the ultimate touchstone’.⁶² Courts may rely on several theories to conclude that Congress intended federal law to displace state law. The two theories most relevant to the question of whether FDA regulation of mifepristone preempts state restrictions are impossibility preemption, in which parties cannot comply with both state and federal law, and obstacle preemption, when state law is an obstacle ‘to the accomplishment and execution of the full purposes and objectives of Congress’.⁶³

In this Part, we investigate how impossibility and obstacle preemption apply to three examples of state restrictions on mifepristone—restrictions on medication abortion that fall short of a total ban, a ban on abortion drugs (or mifepristone), and a ban on all abortion care, whether pharmaceutical or procedural.⁶⁴ We strive to articulate the strongest versions of these preemption arguments, showing that courts would be on firm legal ground concluding that state efforts to regulate mifepristone more stringently than FDA does or to ban the drug are preempted and that there might be actions that FDA could take to bolster such preemption claims. A preemption challenge to a total ban on abortion care presents a more difficult case but nevertheless might still succeed.

We start, however, with a note of caution. Preemption determinations are ultimately made by courts, and even the strongest legal arguments that can be made in favor of the FDCA preempting state mifepristone regulation are not without counterarguments, including that courts ‘start with the assumption that the historic police powers of the States were not to be superseded by the Federal Act unless that was the clear and manifest purpose of Congress’.⁶⁵ Given the politics surrounding abortion, there may be strong practical reasons to doubt that this Supreme Court—and some federal judges in appellate and district courts—would be inclined to agree with even the most compelling legal arguments in favor of preemption.⁶⁶ Nevertheless, because of the gravity of the public health harms resulting from Dobbs, and the merits of the legal arguments supporting preemption challenges, such challenges are worth pursuing.

IV.A. Mifepristone Restrictions Falling Short of a Ban

Before Dobbs, certain states hostile to abortion care, such as Mississippi, implemented regulatory schemes for mifepristone that restricted the use of the drug in ways that go beyond what FDA has required under its REMS authorities.⁶⁷ Such restrictions—although likely less appealing to states than direct bans in the post-Dobbs era—may persist at least for some time or may be pursued in states where the legislatures are unable to enact bans.⁶⁸ We use Mississippi’s regulatory scheme as an example to demonstrate how preemption challenges might succeed for such restrictions. We consider obstacle preemption arguments most apt. For prescribing and dispensing restrictions that fall short of an outright ban on the drug’s sale, impossibility preemption may be less applicable because the state and federal laws act on different parties—thus it remains technically possible for the drug’s manufacturer to comply with the federally required REMS while doctors and pharmacists comply with the state’s restrictions,.⁶⁹

In 2013, the Mississippi Women’s Health Defense Act went into effect. This law limited the circumstances in which mifepristone could be prescribed or dispensed.⁷⁰ The law’s requirements included a physical examination by a physician prior to prescribing, a 24-hour waiting period before dispensing,⁷¹ ingestion of the drug at an ‘abortion facility’ in the prescribing physician’s presence,⁷² and a follow-up appointment 14 days after dispensing.⁷³

FDA could have mandated the same set of restrictions had it deemed them necessary to ensure that mifepristone’s benefits outweighed its risks. In fact, it did mandate many of these requirements at one point—including in-person dispensing, in-person ingestion, and in-person follow-up—later removing these requirements after considering the evidence.⁷⁴ FDA also could have required that mifepristone only be dispensed with documentation that the patient underwent a 24-hour waiting period.⁷⁵ Yet FDA did not do so.⁷⁶ In 2021, after years of experience—and extensive review, including in litigation—FDA determined that all in-person requirements, including the measures Mississippi imposed, were still not necessary for safe use of the drug.⁷⁷

This tension between Mississippi’s regulatory scheme and FDA’s determinations under its REMS authority give a court good reason to conclude that Mississippi’s scheme is preempted as an obstacle to the purposes of federal law. When deciding to require a REMS for a prescription drug, FDA is statutorily required to weigh the benefits and risks of the drug.⁷⁸ And when FDA requires a restrictive REMS with ‘elements to assure safe use’ like the mifepristone REMS, Congress has required FDA to consider not only the risks and benefits of the drug but also other factors related to drug access and burdens on the health care system.⁷⁹ For example, FDA must consider how implementation of a REMS affects drug access and minimize ‘delays or interruptions’ in care.⁸⁰ By imposing additional restrictions on mifepristone beyond those in the REMS, Mississippi upsets this balancing of benefits, risks, accessibility, and burdens that Congress charged FDA with undertaking in developing the mifepristone REMS.⁸¹ Although, generally, an agency failing to impose a requirement, or failing to ‘act’, makes a weaker preemption argument, by requiring a REMS, FDA has acted and determined based on its complex balancing of benefits, risks, accessibility, and burdens that no additional safeguards are necessary to ensure mifepristone’s safety and effectiveness, or are possible without being ‘unduly burdensome’ on patient access.⁸² For example, FDA has explained that its December 2021 decision to remove in-person requirements from the mifepristone REMS was because ‘data support modification of the REMS to reduce burden on patient access and the health care delivery system and to ensure the benefits of the product outweigh the risks’.⁸³ In Mississippi, the state’s restrictions did hinder patient access, prompting GenBioPro, the manufacturer of a generic mifepristone product, to file suit challenging the state’s restrictions—a suit it later voluntarily dismissed without prejudice.⁸⁴

Should FDA decide to eliminate the mifepristone REMS altogether, as many reproductive health care experts recommend,⁸⁵ state restrictions on the drug would be similarly or possibly even more vulnerable to obstacle preemption challenges.⁸⁶ FDA may only release a REMS upon a determination that it is ‘no longer necessary to ensure a medication’s benefits outweigh its risks’.⁸⁷ A decision to remove a REMS reflects an active determination by FDA that a REMS is not needed for a drug to be safe and effective. Thus, as with requirements that go beyond those in an existing REMS, state restrictions when FDA has released a REMS undermine the uniform drug regulatory scheme that Congress sought to create through enacting the FDCA⁸⁸ and frustrate FDA’s ability to do the complex balancing of safety, effectiveness, health care system burdens, and patient access that Congress has required of the agency in the REMS provisions of the FDCA.⁸⁹

Although these arguments in favor of obstacle preemption are strong, it is important to note that such claims are generally considered to be weaker than impossibility preemption claims⁹⁰ and that there exist counterarguments. For example, in Wyeth v Levine, the Supreme Court explained that implied preemption arguments are ‘particularly weak where Congress has indicated its awareness of the operation of state law in a field of federal interest, and has nonetheless decided to stand by both concepts and to tolerate whatever tension there [is] between them’.⁹¹ The Mississippi law was enacted after Congress added REMS authorities to the FDCA, but various restrictions on abortion care have been in place for many years, and it is difficult to imagine that Congress was not aware that states were legislating in this area—which might indicate that Congress intended some level of state and federal regulation to coexist. Nevertheless, as we have highlighted, there are compelling arguments that support courts’ finding in favor of obstacle preemption challenges to restrictions like those imposed by Mississippi.

Additionally, although courts may not defer to FDA’s own views on whether its REMS authorities preempt similar state restrictions on a drug’s use,⁹² there may be steps that FDA could take to strengthen obstacle preemption challenges, should it wish to do so. For example, although FDA does post information about REMS on its website, FDA could—as applicable—more clearly, publicly, and formally document its reasoning for requiring certain elements and not others in a REMS, removing REMS elements, and releasing a REMS altogether. Such explanations might bolster arguments that state restrictions that go beyond those required in a REMS, or remain after a REMS is released, thwart the balancing of considerations that Congress charged FDA with undertaking in its REMS authorities.

IV.B. Bans on Mifepristone

Even before Dobbs—in fact from the time mifepristone was first approved in 2000⁹³—states have shown interest in banning mifepristone or other drugs used for medication abortion.⁹⁴ This interest has only grown following the ruling.⁹⁵ However, when premised on a state legislature’s determinations about the safety and effectiveness of mifepristone, such bans are precisely the state laws that Attorney General Garland identified as particularly vulnerable to preemption challenges.⁹⁶

Texas’s Senate Bill (S.B.) 4, enacted in 2021, provides an instructive example.⁹⁷ S.B. 4 established various reporting requirements and protocols for drugs ‘known to have abortion-inducing properties’, specifically naming Mifeprex—the brand name for mifepristone.⁹⁸ Among other provisions and relevant for a discussion of bans on mifepristone,⁹⁹ S.B. 4 requires physicians to ensure that patients are not beyond a gestational stage of 7 weeks before they are dispensed abortion inducing drugs. Anyone who ‘intentionally, knowingly, or recklessly’ violates the law is subject to possible criminal penalties.¹⁰⁰ The state legislature explained that it was passing the law in part because it had determined that ‘the failure rate and risk of complications [of drug-induced abortion] increases with advancing gestational age’.

FDA, however, has made a different determination about the safety and effectiveness of mifepristone. Although initially approving the drug for use through 7 weeks of gestation and limiting dispensing to physicians,¹⁰¹ FDA determined, in 2016, that mifepristone could be safely and effectively used through 10 weeks of gestation and dispensed by non-physician clinicians—and approved a revised indication.¹⁰² S.B. 4 directly contradicts this relatively recent FDA determination.¹⁰³

This contradiction, in turn, gives rise to an impossibility preemption argument. S.B. 4 effectively bans manufacturers from selling their drug for use in patients from 8 through 10 weeks of pregnancy despite FDA having approved the drug for that specific use. Manufacturers technically could comply with both state and federal law by declining to sell their product for use in patients more than 7 weeks pregnant. However, in 2013, the Supreme Court specifically rejected this kind of ‘stop-selling’ rationale as saving state tort law from impossibility preemption in Mutual Pharmaceutical v Bartlett. In that case, a plaintiff injured by the defendant’s drug asserted design-defect claims under state law regarding the drug’s labeling and composition. The defendant drug manufacturer, Mutual, argued that it was impossible to comply with both state and federal law because, under the FDCA, Mutual could not have made the plaintiff’s proposed changes to the drug’s labeling or composition without first obtaining FDA approval for those changes. The Court agreed, further explaining,

The Court of Appeals’ solution—that Mutual should simply have pulled [its drug] from the market in order to comply with both state and federal law—is no solution. Rather, adopting the Court of Appeals’ stop-selling rationale would render impossibility pre-emption a dead letter and work a revolution in this Court's pre-emption case law.¹⁰⁴

Laws like Texas’s S.B. 4 banning certain federally approved sales that are grounded in the state’s safety and effectiveness judgments could likewise be understood as preempted as they might make impossible compliance with certain post-marketing obligations, such as adverse event reporting in the mifepristone REMS, given that reporting could be used to support criminal prosecution under state law.¹⁰⁵

Such an impossibility argument is even more powerful for bans on all federally-approved sales of a drug. Just as was the case for the defendant drug manufacturer in Mutual Pharmaceutical v Bartlett, a law that effectively bans manufacturers from selling their drug for its approved use for pregnancy termination, or that effectively bans manufacturers from selling the substance mifepristone, would mean that manufacturers must obtain FDA approval of a new indication or a new composition of their drug before selling in the state.¹⁰⁶ In other words, it would be impossible for the manufacturer to comply with both state and federal law without pulling its drug from the state’s market—which is precisely the solution that the Supreme Court rejected in Mutual Pharmaceutical v Bartlett.¹⁰⁷

Obstacle preemption challenges are also worth consideration.¹⁰⁸ As noted above, obstacle preemption is seemingly disfavored by some courts as a general matter.¹⁰⁹ This disfavor might be amplified with respect to state prescription drug bans that contradict FDA approval decisions, specifically, rather than decisions about REMS requirements. Uncodified language included in the 1962 Amendments to the FDCA—the amendments that created the modern approval standard—specified that state authority was preserved except where it ‘direct[ly] and positive[ly] conflict[s]’ with those amendments.¹¹⁰ This, in turn, might lend support to arguments that only impossibility theories should be relevant where the conflict between state and federal law concerns the drug approval standard (rather than, for example, the REMS provisions of the FDCA added in separate 2007 amendments).¹¹¹

However, there are strong obstacle preemption arguments. One way that state mifepristone bans may pose an obstacle to the purposes of the FDCA is by eliminating national uniformity in the drug market. As Professor Lars Noah has argued, ‘to the extent that Congress intended for the FDA to make definitive and nationally uniform judgments about the safety and effectiveness of pharmaceutical products, state efforts to second-guess the agency’s determinations certainly would threaten to frustrate those … purposes’.¹¹² Although there is no express provision in the FDCA describing national uniformity in the prescription drug market akin to what exists for non-prescription drugs,¹¹³ the structure of the FDCA, and the history of its enactment, support the view that Congress intended the law to create a nationally uniform market for prescription drugs. For example, scholars have explained that industry sought federal regulation of food and drugs because, in the early 20th century, ‘inconsistencies in applicable state laws made operating on a national scale increasing difficult’.¹¹⁴ Similarly, the House Report accompanying the Pure Food and Drug Act of 1906—the predecessor to the FDCA—explained ‘[o]ne of the hoped-for good results of a national law … is the bringing about of a uniformity of laws and regulations on the part of the States within their own several borders’,¹¹⁵ and the goal of national uniformity remained a focus in Congress’s deliberations leading to the 1938 enactment of the FDCA.¹¹⁶ Moreover, the modern drug approval structure—wherein drug manufacturers must conduct time-consuming, expensive clinical trials to demonstrate safety and effectiveness sufficient to obtain FDA approval—is premised on the promise of a national market being available to those manufacturers who do prove their drug safe and effective.¹¹⁷

Another reason might arise from FDA’s congressionally defined mission, which includes both ‘protect[ing] the public health’ ‘by ensuring that … drugs are safe and effective’ as well as ‘promot[ing] the public health’ by ‘taking appropriate action on the marketing of regulated products in a timely manner’.¹¹⁸ In Zogenix, Inc. v Patrick, the manufacturer of an FDA-approved, extended-release single-ingredient hydrocodone products challenged the Massachusetts ban on the prescribing and dispensing of the drug, and a federal judge pointed to this mission in her decision to enjoin the ban, reasoning that Massachusetts ‘interposed its own conclusion about [the medication’s] safety and effectiveness’ which in turn ‘undermine[d] the FDA’s ability to make drugs available to promote and protect the public health’.¹¹⁹ To be sure, FDA cannot ‘guarantee’ any person access to any given drug—for instance, FDA does not regulate prescription drug pricing (and cannot assure patients will be able to afford a drug), nor can the agency compel a manufacturer to continue to make and sell a product if the manufacturer chooses to cease production for business (or any) reasons. Nevertheless, as the judge in Zogenix observed, FDA’s public health mission, as described in section 1003 of the FDCA,¹²⁰ includes aspects of drug availability.¹²¹

Considering these purposes of the FDCA, obstacle preemption arguments apply in a relatively straightforward way to state laws that completely ban mifepristone, analogous to Massachusetts’s total ban on prescribing and dispensing single-ingredient hydrocodone products. Importantly, this holds true even if the state laws are framed as regulating the medical practices of prescribing and dispensing mifepristone—areas conventionally understood to be generally outside FDA’s direct authority and within states’ authority—rather than as laws regulating the drug itself.¹²² In at least two cases—Zogenix and a case involving a successful ‘field preemption’ challenge to Maine’s attempt to revise its practice of pharmacy laws to allow the importation of unapproved drugs—federal judges have concluded that states cannot escape preemption challenges simply by framing their efforts to regulate prescription drugs as part of their oversight of medical practice.¹²³

The arguments should likewise apply to any partial bans on mifepristone, such as Texas’s S.B. 4. After Massachusetts’s complete ban on prescribing and dispensing single-ingredient hydrocodone products was enjoined, Massachusetts revised its regulations to require prescribers ‘take certain steps before prescribing Zohydro [the only approved extended-release single-ingredient hydrocodone product at the time]’.¹²⁴ In subsequent litigation, the judge declined to dismiss the obstacle preemption challenge to these new regulations because the drug manufacturer’s complaint alleged that the regulations were affecting the drug’s availability, and ‘if the allegations are proven’, the manufacturer would be ‘entitled to relief’.¹²⁵ S.B. 4, and laws like it, clearly restrict the availability of mifepristone—and following the rationale of the subsequent Zogenix decision, may therefore be vulnerable to obstacle preemption arguments. FDA’s relatively recent 2016 decision to extend mifepristone’s approved indication from use through 7 weeks, to use through 10 weeks of pregnancy, only strengthens this argument.¹²⁶ To approve the new indication, the agency reviewed the safety and effectiveness data and determined that the benefits of mifepristone outweigh its risks in women seven to 10 weeks pregnant. Texas’s S.B. 4, according to the text of the law itself, likewise is premised on a determination about the safety of mifepristone but with the Texas legislature reaching a different conclusion than FDA did. Accordingly, a court could reasonably conclude that S.B. 4 contradicts, and upsets, the complex balancing of risks and benefits that Congress has charged FDA with conducting in its approval process, undermining the uniform drug regulatory scheme created by the FDCA, the FDA’s mission to promote public health through helping to ensure the availability of safe and effective drugs, or both.

Thus far, we have considered state mifepristone bans premised on state legislatures’ judgments about the drug’s safety and effectiveness. A key question, especially post-Dobbs, is whether a state mifepristone ban based on reasoning other than safety and effectiveness concerns would likewise be vulnerable to an impossibility or an obstacle preemption challenge. For example, in February 2022, ‘The Alabama Chemical Abortion Prohibition Act’ was introduced into Alabama’s House, which would make it ‘unlawful for any person or entity to manufacture, distribute, prescribe, dispense, sell, or transfer the “abortion pill,” otherwise known as RU-486, Mifepristone, Mifegyne, or Mifeprex, or any substantially similar generic or non-generic abortifacient drug’ in the state.¹²⁷ The bill includes several proposed findings of the state legislature, one of which is safety-related,¹²⁸ but many of which are not, including explaining that ‘Alabama is committed to the sanctity of human life, from conception to natural death’ and ‘in this state [] abortion is not health care’.¹²⁹ In a post-Dobbs world, states may do away with the pretense of including in laws findings on mifepristone’s safety and effectiveness that differ from FDA’s and instead rely on findings framed as social or moral concerns regarding fetal life. At the same time, FDA is not generally authorized to consider the social or moral questions as part of its consideration of a product’s safety and effectiveness,¹³⁰ an interpretation that the agency itself has put forward.¹³¹

The oft-invoked presumption against preemption¹³²—which Professor Elizabeth McCuskey has described as ‘plac[ing] a thumb on the intent scale’ against finding Congressional intent to displace state law¹³³—may be the aspect of preemption doctrine that provides the best evidence that states have at least some latitude to regulate in ways that contradict federal law when states are considering factors outside the purview of the relevant federal regulator. This is, perhaps, why Attorney General Garland specifically asserted that states ‘may not ban Mifepristone based on disagreement with the FDA’s expert judgment about its safety and efficacy’.¹³⁴

Yet, logically, an impossibility challenge should not turn on the stated purpose of state law—if parties cannot comply with both state and federal law, they are in a bind regardless of whether state law reflects a direct disagreement with the federal government or some other considerations. Consider state tort law. Although, at a high level, the FDCA and state tort law share some similar product-safety aspects, it is also the case that tort law’s purposes are distinct from the FDCA’s purposes. Unlike FDA regulation, which does not provide a private right of action, tort law serves to offer injured people the opportunity for compensation, for example.¹³⁵ Those differing purposes have not stopped courts from concluding that the FDCA preempts state tort law in certain circumstances in which compliance with both is impossible. And just as permitting a stop-selling argument to prevail in tort lawsuits ‘would render impossibility pre-emption a dead letter’, in the Court’s words, so too would allowing states to escape impossibility preemption merely by claiming to be considering some factor outside the federal government’s purview in banning a drug.¹³⁶

Similarly, a state’s purpose in banning mifepristone should not be understood as relevant to obstacle preemption challenges. Because FDA disclaims authority to consider ‘moral, religious, or ethical issues’ regarding the products it regulates, a court, as a practical matter, might be tempted to conclude a state government can do so without frustrating the purposes of FDA oversight.¹³⁷ Contrary to this intuition, however, courts should understand FDA’s lack of authority to consider ‘moral, religious, or ethical issues’ as strengthening preemption challenges to morally-based mifepristone bans that thwart the FDCA’s creation of a nationally uniform drug market. That is, in enacting the FDCA, Congress envisioned a nationally uniform drug market based on technical determinations about drug safety and effectiveness—excluding moral, religious, and ethical considerations from determining the parameters of that uniform market. Likewise, regardless of the state’s purpose, bans on FDA-approved drugs, by limiting their availability, arguably thwart FDA’s mission to promote public health through helping to ensure the availability of safe and effective drugs.¹³⁸

IV.C. Complete Bans on Abortion Care

In addition to retaining or enacting drug-specific restrictions or bans on mifepristone, post-Dobbs states will also, enact, and in some cases already have enacted, complete or near-complete bans on all abortion care. For example, Missouri law now provides ‘no abortion shall be performed or induced upon a woman, except in cases of medical emergency’,¹³⁹ while Ohio prohibits any person from ‘knowingly and purposefully perform[ing] or induc[ing] an abortion’ after a fetal heartbeat has been detected unless the abortion ‘is designed or intended to prevent the death of the pregnant woman or to prevent a serious risk of the substantial and irreversible impairment of a major bodily function of the pregnant woman’.¹⁴⁰

Although these laws do not specifically target an FDA-approved prescription drug, as mifepristone bans do, they nevertheless have the same impact as a mifepristone-specific ban, effectively prohibiting drug distribution.¹⁴¹ Nevertheless, challenges to state bans on all abortion care on the ground that such laws are preempted by the FDCA, at least at first glance, seem less likely to succeed than challenges brought against drug-specific restrictions or bans. As Professor David Cohen, Professor Greer Donley, and Dean Rachel Rebouche have argued, it is relatively easy to imagine courts applying ‘the concept that Congress does not hide huge, politically-relevant policy decisions in the interstices of a statute’ (the ‘no-elephants-in-mouseholes doctrine’) to preemption arguments.¹⁴² In other words, by enacting the FDCA, which does not ‘explicitly’ mention abortion, Congress did not ‘implicitly’ give FDA the power to decide the social question of whether abortion is permissible.¹⁴³

More broadly, the less state laws appear to be replicating, and thwarting, the careful balancing of benefits and risks that Congress has tasked FDA with for drug approval and REMS decisions, and the more state laws appear to be doing ‘something else’, the weaker courts may perceive preemption arguments to be as a practical matter—particularly when states are regulating medical practice, an area conventionally understood to be within states’ purview.¹⁴⁴ Perhaps for this reason, abortion opponents have offered medical aid in dying (MAID) as an instructive example.¹⁴⁵ Ed Whelan, for example, argued ‘[a]ssume that the FDA approved a drug for use in physician-assisted suicide. Why would anyone imagine that FDA approval overrode state laws barring physician-assisted suicide? Why should it be any different [state abortion laws]?’¹⁴⁶

Notwithstanding this practical intuition, as a legal matter, it may not be the case that states are free to regulate medical practical in ways that effectively ban FDA-approved drugs.¹⁴⁷ Complete or near-complete state bans on abortion care effectively prohibit the distribution of mifepristone for its FDA-approved use, requiring the drug manufacturer to seek FDA approval of a new indication for its product or to stop selling its product within the state. If the Court’s reasoning in Mutual Pharmaceutical v Bartlett is to be taken seriously—that a manufacturer’s ability to stop selling a product within a state’s borders is no counter to an impossibility preemption theory¹⁴⁸—then, total bans on abortion care are potentially vulnerable to arguments that they make it impossible for drug manufacturers to comply with state and federal laws.¹⁴⁹ Likewise, the obstacle preemption questions raised by state mifepristone bans may be raised by state bans on all abortion care. State bans on all abortion care will limit the availability of mifepristone, in turn undermining the uniform drug regulatory scheme that Congress envisioned in creating the FDCA as well as FDA’s mission to promote public health through helping to ensure the availability of safe and effective drugs. That is, courts could conclude that, while states retain significant power to regulate the practice of medicine, they do not have the power to regulate medical practice in ways that make compliance with FDA requirements impossible, or obstruct the purposes of the FDCA.

V. BEYOND ABORTION

As the above analysis demonstrates, preemption challenges to state mifepristone regulation—including both drug-specific restrictions and complete bans on abortion care—hold promise for mitigating the consequences of Dobbs. Their success would have clear public health benefits in the context of abortion. States, however, also regulate the uses of prescription drugs for many other indications and have done so since before FDA’s inception.¹⁵⁰ This regulation might come in the form of positive law, as when Massachusetts attempted to ban the prescribing and dispensing of FDA-approved, extended-release single-ingredient hydrocodone products or California’s 2004 ‘track and trace’ law that sought to impose stricter requirements on the drug supply chain than FDA did.¹⁵¹ This regulation also might come in the form of tort liability in which patients harmed by a prescription drug seek compensation for their injuries from drug manufacturers under state law.

Such overlapping state and federal oversight can have public health benefits. For example, Professors Stephanie David and Sara Rosenbaum explained,

“[S]tates also engage in extensive health and welfare regulation, often in ways that overlap with federal law, frequently supplementing federal safeguards. Thus, for example, federal law establishes standards for preventing unsafe products from entering the market, while state laws may add remedies for people who are injured by unsafe products…”¹⁵²

They go on to explain that tort law, in particular, benefits public health by protecting consumers through producing additional information about drugs against the reality that ‘FDA’s resources to monitor drugs on the market are limited, particularly with regard to post-marketing surveillance of approved drugs’.¹⁵³ In this Part, we investigate the effects that preemption challenges to state mifepristone restrictions—whether they fail or succeed—may have for other areas of intersecting state and FDA drug regulation.¹⁵⁴

V.A. What If Challenges Fail?

As our analysis in Part IV acknowledges, preemption challenges to state mifepristone restrictions may ultimately prove unsuccessful for various reasons. Depending on a court’s reasoning, such an outcome may weaken preemption defenses in certain tort lawsuits—for example, if a Supreme Court opinion, notwithstanding Mutual Pharmaceutical, endorsed a stop-selling rationale as an effective rebuttal to an impossibility preemption argument—which may be viewed as a public health win.¹⁵⁵ However, opinions in these cases may also be written narrowly, distinguishing tort law from the abortion context, avoiding this result.

Likely more problematic for public health is if failed preemption challenges to state mifepristone laws pave the way for state laws banning or imposing more stringent restrictions than FDA’s on other prescription drugs. This may be a particular risk for drugs for other stigmatized indications or that are otherwise controversial. Interest in limiting or prohibiting access to—or otherwise shaping use of—such drugs is not merely hypothetical. The same advocates that pushed Texas to adopt S.B. 8, which permits private lawsuits against any person who assists a person getting abortion after 6 weeks of pregnancy, are reportedly now seeking to limit insurance coverage for drugs indicated for HIV PrEP.¹⁵⁶ A bill was introduced into Ohio’s house in 2019 that would ban insurance coverage for hormonal contraceptives or contraceptive devices.¹⁵⁷ Florida declined to order COVID-19 vaccines for children under 5 years of age, and declined to recommend their administration in that population, in June 2022.¹⁵⁸ Concerns have also been raised about the possibility that state lawmakers will attempt to restrict access to selective serotonin reuptake inhibitors (SSRIs) and other psychiatric drugs.¹⁵⁹ Although the previous examples represent measures pursued by conservative activists and lawmakers, interest in such efforts is not limited to one political party. In fall 2020, amidst concerns that FDA bowed to political pressure from President Trump’s Administration to authorize certain COVID-19 therapeutics on questionable scientific evidence, New York established a process by which the state would independently review any FDA-authorized COVID-19 vaccines.¹⁶⁰ Somewhat similarly, under Governor Deval Patrick, Massachusetts attempted in 2014 to ban the prescribing and dispensing of extended-release single-ingredient hydrocodone products, as discussed in Part IV.

Although these examples—excluding Massachusetts—did not constitute outright bans, one can easily imagine how the same political forces that drove these measures could pursue more severe restrictions or bans on these and other prescription drugs. A patchwork of state judgments diverging from FDA’s would undermine public health by limiting patient access to safe and effective products and by making the market for prescription drugs no longer uniform.¹⁶¹ Yet, the possibility that preemption challenges to state mifepristone laws will fail does not militate against pursuing such challenges for several reasons. As explained in Part II, mifepristone access is critical to preserve needed health care. Additionally, there may be ways to distinguish state mifepristone bans and restrictions from state efforts to regulate other prescription drugs. For example, if preemption challenges to state mifepristone bans are unsuccessful because courts find persuasive the idea that states are free to regulate abortion drugs on social policy grounds that FDA does not consider, state efforts to regulate at least certain other prescription products—like potential bans on COVID-19 vaccines or therapies or SSRIs—might not be as easily characterized as grounded in something other than public health or scientific determinations.¹⁶² But, perhaps most persuasively, states are clearly pursuing restrictions on other prescription drugs¹⁶³ whether mifepristone laws are challenged or not; it, therefore, makes little sense to hold up important public health efforts to increase access to mifepristone for fear states will do something they are already doing.¹⁶⁴

V.B. What If Challenges Succeed?

Having considered the possible implications of courts rejecting preemption challenges to state mifepristone laws, we now consider what the effects of courts endorsing these arguments might be. One possibility is that preemption defenses in failure-to-warn or other tort lawsuits will be strengthened. For instance, in Wyeth v Levine, a majority of the Supreme Court rejected a preemption defense in a failure-to-warn case brought against the manufacturer of a brand-name drug product but based on facts that happened before Congress added the REMS provisions to the FDCA. Prominent food and drug law attorneys have argued that ‘a preemption defense premised on the presence of a REMS is perfectly intuitive’ for many of the reasons discussed in Part IV.A.¹⁶⁵ It may be that successful challenges to state mifepristone restrictions, that at least partly rely on FDA’s REMS authorities, will be used by courts as reason to accept preemption defenses in lawsuits involving drugs subject to REMS.¹⁶⁶

Such an outcome would be troubling from a public health policy perspective. State tort law is an important supplement to FDA oversight of marketed drugs. It is an avenue for learning vital information about drug safety, serves a role separate from FDA’s by compensating injured patients, and can be a catalyst to change industry and government regulators’ behaviors.¹⁶⁷

Here too, however, there may, be ways to distinguish preemption analyses in the context of state mifepristone restrictions and state tort law. For example, the Ninth Circuit recently declined to extend the Supreme Court’s holding in Pliva v Mensing—in which the Court concluded that a failure-to-warn claim against a manufacturer of a generic drug was preempted on an impossibility theory—to drug distributor defendants in a much more similar scenario than is presented by state mifepristone laws.¹⁶⁸ Moreover, given Pliva and a similar holding in Mutual Pharmaceutical v Bartlett, claims against manufacturers of generic drugs, which comprise roughly 90 per cent of all drugs dispensed in the USA,¹⁶⁹ are already often preempted. Against that background, the possibility that preemption challenges to state mifepristone restrictions will further expand preemption defenses in tort lawsuits might not outweigh the potential benefits for reproductive health care access that such challenges hold.

Beyond tort law, successful preemption arguments regarding state mifepristone laws may be extended to other areas in which states often regulate prescribing and dispensing. An obvious example is state regulation of the prescribing and dispensing of opioid analgesics when that regulation comprises requirements that FDA could also impose under the FDCA.¹⁷⁰ As with state mifepristone restrictions, states heavily regulate the prescribing and dispensing of opioids even though FDA has long required REMS for various classes of opioid analgesics.¹⁷¹ Some requirements imposed by states have clearly been considered by FDA. For example, some states, such as Ohio, require that opioid treatment agreements be used with certain patients.¹⁷² These agreements, also known as pain contracts, opioid contracts, or pain agreements, go beyond informed consent documents, outlining the steps a patient must take to receive their medications. Although FDA has declined to require such an agreement as part of the opioid analgesic REMS (eg by requiring that such an agreement be documented as a condition of safe use), FDA’s Safe Use Initiative developed a model agreement in 2012, which remains available on FDA’s website, suggesting the agency has given thorough consideration to such agreements.¹⁷³ In a scenario like this, where state requirements go beyond those that FDA has chosen to implement in a REMS, opioid restrictions may seem vulnerable to preemption challenges in much the same way mifepristone restrictions may be.¹⁷⁴

More generally, concluding that FDA’s REMS authorities, in particular, preempt any and all state restrictions that FDA could impose might give some public health experts pause. Public health scholars have critiqued FDA’s use of its REMS authorities, and empirical studies have cast doubt on how well at least certain REMS are helping to mitigate drugs’ serious risks.¹⁷⁵ The Department of Health and Human Service Office of Inspector General has also critiqued FDA’s implementation of its REMS authorities, both as a general matter and with respect to opioid analgesics specifically.¹⁷⁶ Regulation supplemental to FDA’s, accordingly, may be beneficial in some circumstances even when FDA has required a REMS.

At the same time, these possible implications of successful preemption challenges to state mifepristone laws do not seem inevitable, or perhaps not even uniformly negative.¹⁷⁷ As demonstrated in Part III, mifepristone is not only a drug for which FDA has required a REMS, but even among the subset of drugs for which a REMS is required, it is also a drug for which there is unusually extensive documentation (and litigation regarding) the reasons why FDA has required and released specific elements within the REMS. This particularly extensive documentation and debate within the agency, and between the agency and external reproductive health care advocates, may present a path forward for distinguishing state mifepristone laws from other state laws regarding even drugs subject to REMS.

Even if state opioid restrictions were indistinguishable from state mifepristone restrictions, it might still not be worth delaying the potential reproductive health care benefits of challenging state mifepristone laws on preemption grounds. Reproductive health care experts and organizations widely, if not uniformly, view state restrictions on abortion care as unnecessary and harmful.¹⁷⁸ By contrast, there is greater disagreement about the public health value of state opioid restrictions. Some of these restrictions, such as requirements for opioid treatment agreements, have been widely criticized as unnecessary, stigmatizing, and ineffective at reducing the risks of opioids.¹⁷⁹ However, those restrictions are still often supported by some experts or organizations.¹⁸⁰ Additionally, there is likely to be less agreement overall as to whether the risk that such state restrictions would be displaced under preemption theories would harm, or benefit, public health—and much less agreement about what restrictions on opioid analgesics are necessary and sufficient to ensure the safe use of those drugs.

Finally, when courts conclude state restrictions are preempted, state efforts to regulate prescription drugs nevertheless can influence federal policy or decision-making.¹⁸¹ For example, when Massachusetts attempted to ban the FDA-approved, extended-release hydrocodone product, the state’s major concern was that FDA approved the drug without an abuse-resistant formulation, meaning the drug could be ‘crushed and inhaled or injected, making the full dose of hydrocodone available immediately’.¹⁸² As discussed above, a federal judge concluded, in April 2014, the Massachusetts ban was preempted by the FDCA. Yet, by January 2015, the drug company had developed an abuse-deterrent formulation.¹⁸³ Perhaps the drug manufacturer would have developed, and FDA would have approved, such a formulation anyway, but Massachusetts’s efforts to ban the formulation that lacked abuse-deterrent may have hastened such efforts.¹⁸⁴ Even when courts ultimately conclude state efforts to regulate prescription drugs are preempted by the FDCA, those state efforts might still be impactful, as a practical matter.

VI. CONCLUSION

Reproductive rights advocates and scholars are rightly pursuing a comprehensive strategy to address the harmful consequences of Dobbs and of the state restrictions on abortion care that have followed and will continue to follow the decision.¹⁸⁵ However, the legal arguments advanced to protect reproductive health care may also have implications for other areas of public health.¹⁸⁶ These broader consequences are important to try to understand, particularly at this moment when public health authorities seem generally under attack. In this article, we have shown that one approach, challenging state mifepristone laws as preempted by FDA oversight, need not necessarily lead to an expanded FDA preemption footprint into areas of important complementary state public health regulation, like tort law, nor undermine the future of beneficial public health federalism.

FUNDING

Dr Sarpatwari’s work is supported in part by Arnold Ventures.

DISCLOSURES

Prof. Zettler reports consulting for the World Health Organization, US FDA, and Hopewell Fund and serving as an expert witness retained by the Direct Purchaser Class Plaintiffs in In re Suboxone Antitrust Litigation, No. 2:13-MD-2445 (E.D. Pa) and the Direct Purchaser Class, End Payor Class, and Retailer Plaintiffs in In re Opana Antitrust Litigation, No. 14cv-10150 (N.D. Ill.). Dr. Sarpatwari reports having served as an expert witness retained by the American Civil Liberties Union in American College of Obstetricians and Gynecologists v. United States Food and Drug Administration, No. TDC-20-1320 (D. Md). Ms. Beckmeyer and Ms. Brown report no conflicts.