AHA Files 340B Motion for Temporary Restraining Order Against HHS, HRSA

Introduction

Ten months ago, this Court enjoined Defendants’ attempt to introduce an unprecedented rebate mechanism into the 340B Program. Unchastened, Defendants have returned with a “340B Rebate Model Pilot Program” (Rebate Program) that is nearly twice as large. But just like before, Defendants “failed to follow the APA’s basic blueprint.” Am. Hosp. Ass’n v. Kennedy, 820 F. Supp. 3d 30, 47 (D. Me. 2025) (AHA I). Immediate judicial intervention is necessary to protect the “financial survival” of 340B hospitals and the health of their patients. Am. Hosp. Ass’n v. Kennedy, 164 F.4th 28, 33–34 (1st Cir. 2026) (AHA II).

Defendants have again violated the APA in multiple ways. Like last year, they “failed to consider an important aspect of the problem”—“the costs and benefits associated with the regulation.” AHA I, 820 F. Supp. 3d at 44. They err on both sides of the cost-benefit equation by “unreasonably understat[ing] the costs of the [Rebate Program] while unreasonably overstat[ing] the benefits.” Chamber of Com. of the U.S. v. FTC, 820 F. Supp. 3d 473, 491 (E.D. Tex. 2026).

On the cost side, Defendants’ analysis “was egregiously flawed and incomplete.” Presidents’ All. on Higher Educ. & Immigr. v. DHS, 2026 WL 2690098, at *15 (D. Mass. Sept. 14, 2026). They estimate that covered entities will spend over $537 million annually if their Rebate Program takes effect. That number is astonishing, but the reality is even worse. Their own half-billion-dollar figure runs counter to overwhelming record evidence (based on first-hand information submitted by 340B hospitals and other covered entities) showing that data-submission costs will require far more work than Defendants predict. Worse still, Defendants arbitrarily excluded entire categories of administrative costs. For example, Defendants disregarded backend reconciliation of paid and denied rebates—the costliest operational burden of a rebate mechanism. They completely wrote off startup, vendor, and training costs, even though they admit these costs will occur. They entirely dismiss “float[] costs” (i.e., having to provide “an interest-free loan from the hospitals to” drug companies, AHA II, 164 F.4th at 32) based on two drug-industry funded studies that concededly do not address the full scope of those costs. And they wholly discount harms to patients and communities, even though this Court previously held that spending hundreds of millions of dollars on administrative costs will have the “downstream effect [of] causing [hospitals] to cut back services.” AHA I, 820 F. Supp. 3d at 46.

On the benefits side, Defendants’ errors are just as fatal. They never quantified the scope of the purported duplicate discount problem they claim to solve. They never evaluated how the Program will incrementally solve that purported duplicate discount problem. And they unquestioningly accepted predictions from drug companies about the purported benefits of a rebate mechanism, even though the drug companies’ figures contradict Defendants’ own audits and arithmetic. See Compl. ¶¶ 18–20, 138–41, 255–57.

Other errors abound. Defendants dismissed “reliance interests” as unjustified as a matter of law only months after conceding they were “significant.” AHA II, 164 F.4th at 33. They rejected less burdensome alternatives on unreasonable, speculative, and circular grounds. Most fundamentally, they did not apply the correct legal standard—what payment mechanism is “the most effective and most efficient from the standpoint of each type of ‘covered entity,’” H.R. Rep. No. 102-384, pt. 2, at 16 (1992). Defendants therefore leave the Rebate Program “unmoored from the purposes and concerns” of the relevant law. Judulang v. Holder, 565 U.S. 42, 64 (2011).

To be clear, Plaintiffs are not “asking for a policy judgment against the Agency.” AHA I, 820 F. Supp. 3d at 45. They ask only for a ruling that Defendants, in making their own policy judgment, failed to consider the “right stuff.” 33 Charles Alan Wright & Arthur R. Miller, Federal Practice and Procedure § 8415 (2d ed. 2018). After all, “to pass muster under the APA’s ‘arbitrary and capricious’ test, courts must ‘ensure that agency decisions are founded on a reasoned evaluation of the relevant factors.’” Penobscot Air Servs., Ltd. v. FAA, 164 F.3d 713, 719 (1st Cir. 1999) (quoting Marsh v. Or. Nat. Res. Council, 490 U.S. 360, 378 (1989)). Because Defendants did not do that here, Plaintiffs are likely to succeed on the merits. And because irreparable harm is even more acute than last year and the balance of equities still weighs in Plaintiffs’ favor, this Court should enjoin the new Rebate Program before it goes into effect on January 1, 2027.

Background

Congress created the 340B Program so safety-net providers could buy drugs at discounts and “stretch scarce Federal resources as far as possible.” H.R. Rep. No. 102-384, pt. 2, at 12; see 42 U.S.C. § 256b. Congress charged the Health Resources and Services Administration (HRSA) with implementing discounts through the payment mechanism “that is the most effective and most efficient from the standpoint of each type of ‘covered entity.’” H.R. Rep. No. 102-384, pt. 2, at 16; see Novartis Pharms. Corp. v. Kennedy, 182 F.4th 1027, 1030–31, 1035 (D.C. Cir. 2026). For the past 33 years, HRSA has concluded that a rebate mechanism is “both inferior to Section 340B’s current upfront-discount model and disruptive to safety-net hospitals.” AHA II, 164 F.4th at 32.¹

In 2025, however, HRSA attempted to compel more than 14,000 rural and safety-net providers to participate in a so-called “pilot” rebate program. This Court enjoined that program. The First Circuit denied a stay, and the government agreed to vacatur. Compl. ¶¶ 58–64.

Just one week after this Court vacated the program, Defendants set about reviving it. Id. ¶¶ 65–69. The new Rebate Program is nearly identical to the one this Court enjoined, only now with 25 potential drugs instead of 10. See Notice Regarding 340B Rebate Model Pilot Program, 91 Fed. Reg. 48,883 (Aug. 3, 2026) (“Notice”). HRSA has already approved plans for 18 drugs. Compl. ¶¶ 195–201. As before, covered entities must float the full purchase price of 340B drugs to manufacturers, submit claims-level data, and await a rebate.

Legal Standard and Reviewability

A district court may grant a temporary restraining order when a movant shows “(1) it is likely to succeed on the merits; (2) it is likely to suffer irreparable harm in the absence of preliminary relief; (3) the balance of equities tips in its favor; and (4) an injunction is in the public interest.” Doe v. Trump, 157 F.4th 36, 46 (1st Cir. 2025). Preliminary relief “preserve[s] the relative positions of the parties until a trial on the merits can be held.” Starbucks Corp. v. McKinney, 602 U.S. 339, 346 (2024). The APA separately allows the Court, “to the extent necessary to prevent irreparable injury,” to “postpone the effective date of an agency action” pending review. 5 U.S.C. § 705.

Agency action is arbitrary and capricious when the agency “relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, [or] offered an explanation … that runs counter to the evidence before the agency.” Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983). An agency’s post-remand explanation “‘must be viewed critically’ to ensure that the [action] is not upheld on the basis of impermissible ‘post hoc rationalization.’” DHS v. Regents of the Univ. of Cal., 591 U.S. 1, 21–23 (2020); Greyhound Corp. v. ICC, 668 F.2d 1354, 1358 (D.C. Cir. 1981).

Notes

1. HRSA had approved a rebate exactly once—for AIDS Drug Assistance Programs, whose purchasing systems “prevented their participation in the section 340B discount program,” and only after finding the change benefited those particular covered entities. 62 Fed. Reg. 45,823, 45,824 (Aug. 29, 1997); 63 Fed. Reg. 35,239, 35,241 (June 29, 1998). [back ^]