In a complicated set of opinions today (in In re Federal Bureau of Prisons’ Execution Protocol Cases), a divided D.C. Circuit panel denied relief to federal death-row inmates seeking an injunction against their execution. The case involves several legal issues (including some on which the panel was unanimous), but I’m going to focus here on the three-way divide among the panel members on the legal questions arising under the Federal Food, Drug & Cosmetic Act, or FDCA (not to be confused with the legal issue concerning the Federal Death Penalty Act, or FDPA).
1. Let’s start with the per curiam opinion, which only Judge Patricia Millett (an Obama appointee) joined in full—and of which she was presumably the lead author. That opinion, applying what it regarded as binding circuit precedent, concluded, first, that the FDPA gives the Food and Drug Administration jurisdiction over drugs used in lethal injections; that plaintiffs were entitled to sue under the Administrative Procedures Act to challenge unlawful agency action; and that the Federal Bureau of Prison’s protocol for execution was unlawful “to the extent that it allows the dispensation and administration of pentobarbital without a prescription.” (Pp. 21-24.)
But the per curiam opinion also concludes that the district court properly declined to award the plaintiffs a permanent injunction against the unprescribed use of pentobarbital. In a single paragraph, the opinion states that the plaintiffs could not demonstrate that they were likely to suffer irreparable harm “due to the unprescribed use of pentobarbital” because the lower court “specifically found” that the evidence did not support their contention that they were “likely to suffer flash pulmonary edema while still conscious.”
2. In a separate opinion, Judge Cornelia Pillard (also an Obama appointee) agreed that the Bureau of Prison’s use of pentobarbital without a prescription was unlawful. But in her judgment “[t]hat conclusion alone requires a stay of the pending executions until the government complies” with the FDCA. She argued that the government’s decision to ignore the prescription requirement “subjects those affected to substantial and unnecessary risks of bodily injury, illness, and suffering” and that exposure to those risks is irreparable injury.
3. In a separate opinion cutting in the other direction, Judge Neomi Rao (a Trump appointee) disputed the majority’s conclusion that binding circuit precedent governs the FDPA issue. One supposed circuit precedent, she argued, “merely assumed the applicability of the FDCA to lethal injection drugs” in the context of distinct legal question. And the other supposed circuit precedent conflicts with a later decision by the Supreme Court.
Addressing the issue afresh, Rao concluded that “applying the requirements of the FDCA to lethal injection drugs does not cohere with the text and structure of the whole statute.” In particular, “the prescription requirement is designed with the therapeutic benefit of the patient in mind” and should not be applied to drugs used for the purpose of lethal injection.
Rao also argued that the plaintiffs cannot, in any event, challenge the FDA’s decision not to enforce the FDCA against drugs used for lethal injection.
4. I haven’t followed this complicated litigation, so my initial take might well be unsound, but I confess that I’m puzzled by the per curiam opinion’s seemingly perfunctory conclusion that plaintiffs failed to show irreparable injury. If I’m understanding things properly (and perhaps I’m not), in Judge Millett’s view the irreparable harm that execution would inflict does not count as a harm attributable to violation of the prescription requirement because execution would ensue anyway if the prescription requirement had been satisfied. In the words of the district judge (also an Obama appointee), “Assuming the BOP finds a doctor to write a prescription, Plaintiffs will still be executed using pentobarbital.” But the same could be said for any suffering from flash pulmonary edema. Why doesn’t the possibility that the government would be unable to get a prescription factor into the calculus? Why, in other words, is it proper to assume that a prescription would be issued?