American Public Health Association v. National Institutes of HealthAmerican Public Health Association v. National Institutes of Health
COMMONWEALTH OF MASSACHUSETTS; STATE OF CALIFORNIA; STATE OF MARYLAND; STATE OF WASHINGTON; STATE OF ARIZONA; STATE OF COLORADO; STATE OF DELAWARE; STATE OF HAWAI‘I; STATE OF MINNESOTA; STATE OF NEVADA; STATE OF NEW JERSEY; STATE OF NEW MEXICO; STATE OF NEW YORK; STATE OF OREGON; STATE OF RHODE ISLAND; and STATE OF WISCONSIN, Plaintiffs, Appellees, v. ROBERT F. KENNEDY, JR., in his official capacity as Secretary of the United States Department of Health and Human Services; UNITED STATES DEPARTMENT OF HEALTH AND HUMAN SERVICES; JAYANTA BHATTACHARYA, in his official capacity as Director of the National Institutes of Health; NATIONAL INSTITUTES OF HEALTH; NATIONAL CANCER INSTITUTE; NATIONAL EYE INSTITUTE; NATIONAL HEART, LUNG, AND BLOOD INSTITUTE; NATIONAL HUMAN GENOME RESEARCH INSTITUTE; NATIONAL INSTITUTE ON AGING; NATIONAL INSTITUTE ON ALCOHOL ABUSE AND ALCOHOLISM; NATIONAL INSTITUTE OF ALLERGY AND INFECTIOUS DISEASES; NATIONAL INSTITUTE OF ARTHRITIS AND MUSCULOSKELETAL AND SKIN DISEASES; NATIONAL INSTITUTE OF BIOMEDICAL IMAGING AND BIOENGINEERING; EUNICE KENNEDY SHRIVER NATIONAL INSTITUTE OF CHILD HEALTH AND HUMAN DEVELOPMENT; NATIONAL INSTITUTE ON DEAFNESS AND OTHER COMMUNICATION DISORDERS; NATIONAL INSTITUTE OF DENTAL AND CRANIOFACIAL RESEARCH; NATIONAL INSTITUTE OF DIABETES AND DIGESTIVE AND KIDNEY DISEASES; NATIONAL INSTITUTE ON DRUG ABUSE; NATIONAL INSTITUTE OF ENVIRONMENTAL HEALTH SCIENCES; NATIONAL INSTITUTE OF GENERAL MEDICAL SCIENCES; NATIONAL INSTITUTE OF MENTAL HEALTH; NATIONAL INSTITUTE ON MINORITY HEALTH AND HEALTH DISPARITIES; NATIONAL INSTITUTE OF NEUROLOGICAL DISORDERS AND STROKE; NATIONAL INSTITUTE OF NURSING RESEARCH; NATIONAL LIBRARY OF MEDICINE; NATIONAL CENTER FOR ADVANCING TRANSLATIONAL SCIENCES; JOHN E. FOGARTY INTERNATIONAL CENTER FOR ADVANCED STUDY IN THE HEALTH SCIENCES; NATIONAL CENTER FOR COMPLEMENTARY AND INTEGRATIVE HEALTH; and CENTER FOR SCIENTIFIC REVIEW, Defendants, Appellants.
Before Montecalvo, Kayatta, and Rikelman, Circuit Judges.
Brett A. Shumate, Assistant Attorney General, Leah B. Foley, United States Attorney, Daniel Tenny and Benjamin C. Wei, Attorneys, Appellate Staff, Civil Division, for appellants.
Andrea Joy Campbell, Attorney General of Massachusetts, David C. Kravitz, State Solicitor, Gerard J. Cedrone, Deputy State Solicitor, Katherine B. Dirks, Chief State Trial Counsel, Rachel M. Brown, Vanessa A. Arslanian, Phoebe M. Lockhart, Assistant Attorneys General, Rob Bonta, Attorney General of California, Emilio Varanini, Supervising Deputy Attorney General, Daniel D. Ambar, Sophia TonNu, Deputy Attorney General, Anthony G. Brown, Attorney General of Maryland, Julia Doyle, Solicitor General, Adam D. Kirschner, Michael Drezner, James C. Luh, Senior Assistant Attorneys General, Nicholas W. Brown, Attorney General of Washington, Andrew Hughes, Assistant Attorney General, Kristin K. Mayes, Attorney General of Arizona, Joshua G. Nomkin, Assistant Attorney General, Philip J. Weiser, Attorney General of Colorado, Shannon Stevenson, Solicitor General, Kathleen Jennings, Attorney General of Delaware, Ian R. Liston, Director of Impact Litigation, Vanessa L. Kassab, Deputy Attorney General, Anne E. Lopez, Attorney General of Hawai‘i, David D. Day, Special Assistant to the Attorney General, Kaliko‘onālani D. Fernandes, Solicitor General, Keith Ellison, Attorney General of Minnesota, Elizabeth Kramer, Solicitor General, Aaron D. Ford, Attorney General of Nevada, Heidi Parry Stern, Solicitor General, Matthew D. Platkin, Attorney General of New Jersey, Angela Cai, Executive Assistant Attorney General, Raúl Torrez, Attorney General of New Mexico, Astrid Carrete, Assistant Attorney General, Letitia James, Attorney General of New York, Judith N. Vale, Deputy Solicitor General, Dan Rayfield, Attorney General of Oregon, Robert Koch, Senior Assistant Attorney General, Peter F. Neronha, Attorney General of Rhode Island, Jordan Broadbent, Special Assistant Attorney General, Joshua L. Kaul, Attorney General of Wisconsin, and Lynn K. Lodahl, Assistant Attorney General, for appellees Commonwealth of Massachusetts, et al.
Megan Barbero, Kyle H. Keraga, and Venable LLP, on brief for Biological and Biomedical Research Societies, amici curiae.
RIKELMAN, Circuit Judge.
In early 2025, the National Institutes of Health (NIH) and the Department of Health and Human Services (HHS) put into place a new policy that prohibits NIH from funding scientific research grants in certain categories. Two groups of plaintiffs sued, alleging that the new policy and the research grant terminations that directly flowed from it violated the Administrative Procedure Act (APA) and the U.S. Constitution. The plaintiffs claimed, for example, that the new policy was arbitrary and capricious because NIH and HHS never defined the prohibited research categories and their explanation for discontinuing such research rested on circular reasoning. The district court held a trial on the merits, ruled in the plaintiffs’ favor, including on their arbitrary and capricious claims, and entered two orders setting aside the
I. BACKGROUND
This appeal involves two separate cases, which the district court informally consolidated. The plaintiffs in the first case are private research and advocacy organizations and individual researchers who receive NIH funding. The plaintiffs in the second case are states whose public universities and colleges depend on NIH funding to support research projects. The plaintiffs brought APA claims under
After the plaintiffs filed their lawsuits, they moved for a preliminary injunction. The district court treated part of the Department‘s briefing opposing the preliminary injunction as a motion to dismiss, including on jurisdictional issues. After dismissing some of the plaintiffs’ claims, the court consolidated the preliminary injunction hearing on the remaining claims with a trial on the merits under
Ultimately, the district court issued two decisions that are critical to this appeal. First, the court determined that it had subject matter jurisdiction, rejecting the Department‘s argument that the case should have been brought in the U.S. Court of Federal Claims. See Memorandum and Order on Subject Matter Jurisdiction, Massachusetts v. Kennedy, No. 25-cv-10814 (D. Mass. May 12, 2025), Dkt. No. 105. In so ruling, the court distinguished the U.S. Supreme Court‘s recent per curiam order in Department of Education v. California, 145 S. Ct. 966 (2025). The district court reasoned that these cases are “not . . . action[s] for monetary damages” but instead are “action[s] to stop the [Department] from violating the statutory grant-making architecture created by Congress . . . and exercising authority arbitrarily and capriciously, in violation of federal law and the Constitution.” Dkt. No. 105, at 22. Thus, it concluded that the cases belonged in federal district court. Second, the court issued a detailed decision recounting its findings of fact and conclusions of law on a subset of the plaintiffs’ APA claims, and issued a partial final judgment under
In its decision resolving those APA claims, the district court began by laying out the relevant legal background. NIH is authorized by statute to “make grants-in-aid to universities, hospitals, laboratories, and other public or private institutions, and to individuals” to “promote . . . research, investigations, experiments, demonstrations, and studies relating to the causes, diagnosis, treatment, control, and prevention of physical and mental diseases and impairments of man.”
In January 2025, President Donald Trump issued three Executive Orders (EOs) limiting the ability of federal agencies to use federal funds to support research grants in certain categories.1 Contrary to the stated goals of the EOs, the district court concluded that the Department had engaged in “pervasive racial discrimination in selecting grants for termination,” as well as an “unmistakable pattern of discrimination against women‘s health issues.” Am. Pub. Health Ass‘n, 2025 U.S. Dist. LEXIS 125988, at *12 n.4.
The district court found that in the weeks after the EOs, officials at HHS and NIH issued several directives (the “Challenged Directives“) prohibiting funding activities related to the broad categories targeted by the EOs. Initially, then-Acting Secretary of HHS, Dr. Dorothy Fink, released a directive explaining that the agency would no longer be funding activities “that support DEI and similar discriminatory programs,” because such activities were “inconsistent with the Department‘s policy of improving the health and well-being of all Americans.” Id. at *22.
Next, in mid-February, the Acting Director of NIH, Dr. Matthew Memoli, distributed a directive stating that NIH was no longer “supporting low-value and off-mission research programs, including but not limited to studies based on diversity, equity, and inclusion (DEI) and gender identity.” Id. at *26-27. According to Dr. Memoli, all such grants were discriminatory and unscientific.2 NIH then removed various notices of funding opportunities (NOFOs) that purportedly violated these directives. See id. at *27.
After a close review, the district court concluded that the Department‘s “unreasonable and unreasoned agenda of blacklisting certain topics, . . . on this Administrative Record, has absolutely nothing to do with the promotion of science or research.” Id. at *18. The court determined that there was no evidence in the record to support Dr. Memoli‘s assertion that grants in the prohibited categories were “antithetical to the scientific inquiry, do nothing to expand our knowledge of living systems, provide low returns on investment, and ultimately do not enhance health, lengthen life, or reduce illness.” Id. at *55-56. To the contrary, the court found that “Dr. Memoli was taking advice” on what types of research aligned “with the new objectives” not from scientists, but from “official[s] in the so-called Department of Government Efficiency (‘DOGE‘).” Id. at *27.
The district court went on to make comprehensive findings about the grant terminations. It explained that grant recipients were informed of their funding termination via “template letter[s]” (the “Termination Letters“). Id. at *30. The court further determined that NIH was not involved in drafting the Termination Letters. See id. at *28-34. For example, the court pointed to testimony by Michelle Bulls, the Chief Grants Management Officer (CGMO) at NIH, who signed each Termination Letter. See id. at *30-33. As the court highlighted, CGMO Bulls testified that “she did not create any of the language” in the letters and was “unaware whether NIH undertook any assessment at all as to whether a particular grant met the criteria being espoused in the letters.” Id. at *30. Instead, she explained, the template Termination Letter was created by a DOGE staffer, Rachel Riley. See id. at *34. Thus, the court concluded that HHS and NIH were “being force-fed unworkable ‘policy’ supported with sparse pseudo-reasoning, and wholly unsupported statements.” Id. at *51.
As the district court detailed, the template Termination Letter included a space to “INSERT EXPLANATION -- EXAMPLES BELOW” as to why the grant was being terminated. Id. at *35. That text was followed by a “reason-for-termination menu,” that listed: “China,” “DEI,” and “Transgender issues.” Id. “Vaccine Hesitancy,” “COVID,” “Gender-Affirming Care,” “Climate Change,” and “Influencing Public Opinion” were later added to the list of “examples for research activities that NIH no longer supports.” Id. at *40. The court found that “usage of this list was mandatory.” Id. The court also determined that the “boilerplate language” in the Termination Letter about DEI and gender identity tracked “almost verbatim” the language in Dr. Memoli‘s February directive. Id. at *54. The record includes examples of Termination Letters sent to individual grant recipients; the court found that those executed Termination Letters did not provide any additional, grant-specific reasoning but merely adhered to the template language. See id. at *31, *50.
The district court concluded that no NIH scientists were involved in selecting the grants to be terminated. See id. at *54-57. Rather, the evidence showed that DOGE staffers (who had no affiliation with either NIH or
Finally, the district court determined that there was no evidence in the administrative record that the Department considered the “reliance interests that naturally inure to [the] NIH grant process” in terminating the grants, contrary to the requirements of the APA. Id. at *59. Those reliance interests included, as the plaintiffs described in their submissions to the district court, “the risk to human life as research and clinical trials are suspended,” and damage to “the overall scientific endeavor” as a result of abruptly terminating hundreds of studies that had been underway for years, representing millions of hours of work.
Based on these findings, the district court concluded that the Challenged Directives and resulting grant terminations were arbitrary and capricious and contrary to law, all in violation of the APA. See id. at *60, *64 (citing
(1) “the [] Directives . . . are declared . . . arbitrary and capricious, and unlawful, in violation of
5 U.S.C. § 706(2)(A) “; (2) “the Directives . . . are [] of no effect, void, illegal, set aside and vacated“; (3) “[t]he Resulting Grant Terminations pursuant to the Directives are declared to be unlawful“; and (4) “the Resulting Grant Terminations are . . . of no effect, void, illegal, set aside and vacated.”
The court specifically declined to enter an injunction and instead limited its judgments to declaratory relief.3 The Department moved for a stay of those judgments pending appeal.
II. STANDARD OF REVIEW
As the party seeking a stay pending appeal, the Department bears the burden of justifying the extraordinary relief it requests. See Nken v. Holder, 556 U.S. 418, 433-34 (2009). To meet its burden, the Department must make: (1) “a strong showing that [it] is likely to succeed on the merits“; (2) a showing that it “will be irreparably injured absent a stay“; (3) a showing that the “issuance of the stay will [not] substantially injure the other parties interested in the proceeding“; and (4) a showing that “the public interest lies” with the Department, not the plaintiffs. Does 1-3 v. Mills, 39 F.4th 20, 24 (1st Cir. 2022) (quoting Bos. Parent Coal. for Acad. Excellence Corp. v. Sch. Comm. of Bos., 996 F.3d 37, 44 (1st Cir. 2021)). Because the district court held a trial on the merits and issued a partial final judgment, we review its findings of fact for clear error and its legal conclusions de novo. See Aponte v. Calderón, 284 F.3d 184, 191 (1st Cir. 2002).
In its stay motion, the Department focuses primarily on the first stay factor -- likelihood of success on the merits. In particular, it leans heavily on its argument that the district court lacked subject matter jurisdiction to hear this case, claiming that it is a contract dispute about damages and thus belongs in the Court of Federal Claims. The plaintiffs, for their part, respond that they have not brought a breach of contract claim, and they do not seek damages against the United States. Instead, their claims rest on federal statutory and constitutional provisions that are independent of the terms of their research grants. According to the plaintiffs, their lawsuit challenges an overarching agency policy that precludes NIH from funding research grants in certain categories, and the district court‘s orders here provide only quintessential declaratory relief under the APA: They set aside agency action that is arbitrary and capricious.
We conclude that under Supreme Court precedent, the Department has not met its burden of establishing the grounds for a stay in this case.
A. Likelihood of Success on the Merits
1. Jurisdiction
The Department begins by arguing, as it did in the district court, that the Tucker Act bars the district court from exercising jurisdiction in this case. See
In Bowen, the Supreme Court considered whether a federal district court had jurisdiction to review under the APA a final order by the Secretary of HHS refusing to reimburse Massachusetts for a category of expenditures under its Medicaid program. See 487 U.S. at 882. The Secretary of HHS claimed that the case should have been brought in the Court of Federal Claims. See id. at 891. The Supreme Court disagreed. After lengthy analysis, it held that Massachusetts could challenge the “disallowance” of a category of Medicaid expenditures under the APA in federal district court. See id. at 907, 912. The Court explained that Massachusetts had requested declaratory and equitable relief, and thus it was bringing an action “seeking relief other than money damages” under
In language that has been cited for decades, the Court in Bowen held that although the district court‘s orders ultimately would lead to the disbursement of funds by the federal government, that “outcome is a mere by-product of that court‘s primary function of reviewing the Secretary‘s interpretation of federal law” and did not negate the district court‘s jurisdiction. Id. at 910. Separately, the Court concluded that the state‘s claim was not barred by
We now turn to Great-West, which did not involve the APA. Instead, Great-West concerned a lawsuit against an individual to recover “money past due under a contract,” 534 U.S. at 210-11, based on a provision in that individual‘s employee benefit plan; the question before the Court was whether the case had been properly brought in federal court as an action seeking declaratory and injunctive relief under ERISA, see id. at 208. The Court explained that a reimbursement provision in the employee benefit plan was “the basis for the present lawsuit.” Id. That provision specified that “the Plan shall have the right to recover from the [beneficiary] any payment for benefits paid by the Plan that the beneficiary [was] entitled to recover from a third party.” Id. at 207 (quotation marks omitted). Thus, the Court reasoned, the case was “quintessentially an action at law,” not an action for equitable relief, and could not be brought in federal court under ERISA. Id. at 210. In distinguishing Bowen, the Court noted, in part, that Bowen was not a suit “merely for past due sums, but for an injunction to correct the method of calculating payments going forward,” and therefore involved prospective relief. Great-West, 534 U.S. at 212.
Finally, the Supreme Court recently issued a decision in California, granting the government‘s application for a stay pending appeal, which our court had denied. The critical language in the Court‘s short decision states:
The APA‘s waiver of sovereign immunity does not apply “if any other statute that grants consent to suit expressly or impliedly forbids the relief which is sought.”
5 U.S.C. § 702 . Nor does the waiver apply to claims seeking “money damages.” Id. True, a district court‘s jurisdiction “is not barred by the possibility” that an order setting aside an agency‘s action may result in the disbursement of funds. Bowen v. Massachusetts, 487 U.S. 897, 910 (1988). But, as we have recognized, the APA‘s limited waiver of immunity does not extend to orders “to enforce a contractual obligation to pay money” along the lines of what the District Court ordered here. Great-West Life & Annuity Ins. Co. v. Knudson, 534 U.S. 204, 212 (2002). Instead, the Tucker Act grants the Court of Federal Claims jurisdiction over suits based on “any express or implied contract with the United States.”28 U.S.C. § 1491(a)(1) .
California, 145 S. Ct. at 968 (emphasis added). Because of the emergency posture of that case, we focus on the arguments the parties presented in their stay papers to the Court to understand the rationale behind its decision. See New Jersey v. Trump, 131 F.4th 27, 35 (1st Cir. 2025) (“[W]e ‘rely on the parties to frame the issues for decision,’ given our reluctance to definitively opine on issues for which we have been deprived of ‘the benefit of vigorous adversarial testing.‘” (citations omitted)); cf. Labrador v. Poe, 144 S. Ct. 921, 930 (2024) (Kavanaugh, J., concurring in the grant of stay) (the “tight timeline” for resolving stay motions is “not always optimal for orderly judicial decisionmaking“).
In California, the Court framed Bowen (a case that belonged in federal district court) and Great-West (a case that did not) as representing two ends of the jurisdictional spectrum, so we follow that approach in our analysis. Further, we note that Bowen remains binding upon us because only the Supreme Court is granted “the prerogative of overruling its own decisions,” Rodriguez de Quijas v. Shearson/Am. Exp., Inc., 490 U.S. 477, 484 (1989), so we endeavor to harmonize these three cases.
With that framework in mind, we turn to the district court‘s judgments issued here in response to the plaintiffs’ request that the court declare and set aside as illegal both the Challenged Directives and the grant terminations. First, the court declared that the Challenged Directives violate the APA and are thus void. Second, the court declared that the grant terminations made pursuant to the Challenged Directives violate the APA and are thus void. We treat these declaratory judgments separately, given that “a judicial order vacating an agency rule does not automatically void every decision the agency made pursuant to [that] rule.” D.A.M. v. Barr, 486 F. Supp. 3d 404, 414 (D.D.C. 2020).
As to the declaratory judgment vacating the Challenged Directives, the Department does not develop an argument that the district court lacked jurisdiction to order such relief. Thus, we conclude that the Department has waived that particular argument. See United States v. Zannino, 895 F.2d 1, 17 (1st Cir. 1990). Regardless, the district court clearly had jurisdiction to grant “prospective relief” that will govern “the rather complex ongoing relationships” between the Department and grant recipients. Bowen, 487 U.S. at 905.
The declaratory judgment vacating the grant terminations presents a closer question. Nevertheless, we conclude the Department has not established a strong likelihood of success on its jurisdictional argument as to the grant terminations for two key reasons: (1) the district court‘s orders here did not award “past due sums,” but rather provided declaratory relief that is unavailable in the Court of Federal Claims; and (2) neither the plaintiffs’ claims nor the court‘s orders depend on the terms or conditions of any contract. These facts put this case much closer to Bowen than Great-West and distinguish it from California. We flesh out each of these points below.
First, the district court‘s orders -- both as to the Challenged Directives and the grant terminations -- provide declaratory relief that is well within the scope of the APA. See U.S. Bancorp Mortg. Co. v. Bonner Mall P‘ship, 513 U.S. 18, 25 (1994) (describing the “equitable remedy of vacatur“). Indeed, Bowen made clear that this kind of “specific relief,” the effect of which is to “undo the [Department‘s] refusal to reimburse the [plaintiffs],” is not equivalent to “money damages.” 487 U.S. at 910. Instead, it is a type of declaratory relief that will guide an agency as it decides upon its future course of conduct, and such relief is available only in the
And, focusing on the grant terminations in particular, the district court‘s orders afford the same type of relief that the Supreme Court approved in Bowen. The judgment in Bowen “did not order [any] amount to be paid, and it did not purport to be based on a finding that the Federal Government owed [the plaintiff] that amount, or indeed, any amount of money.” Id. at 909-10.
Instead, “the judgment t[old] the United States that it may not disallow the reimbursement on the grounds given.” Id. at 910. Thus, it simply effectuated the court‘s “primary function of reviewing the Secretary‘s interpretation of federal law.” Id. That is an apt description of the district court‘s orders here.
Great-West, by contrast, is clearly distinguishable: It was a breach of contract case, where a party invoked a specific provision in a health insurance plan in seeking to recover payment for medical expenses made by a third party to a beneficiary under that plan. See 534 U.S. at 207. Thus, Great-West explicitly concerned the “enforce[ment] [of] a contractual obligation to pay money past due.” Id. at 212.
We likewise have no difficulty distinguishing California. There, the Supreme Court construed the district court as having ordered “the Government to pay out past-due grant obligations.” 145 S. Ct. at 968; see also id. (government likely to show that “the District Court lacked jurisdiction to order the payment of money under the APA“); Brief for Respondent at 26 n.3, California, 145 S. Ct. 966 (2025) (No. 24A910). Based on that understanding, the Court held that “the APA‘s limited waiver of sovereign immunity does not extend to orders ‘to enforce a contractual obligation to pay money’ along the lines of what the District Court ordered here.” California, 145 S. Ct. at 968 (quoting Great-West, 534 U.S. at 212). In this case, however, the district court did not “enforce a contractual obligation to pay money.” Rather, the court simply declared that the Department unlawfully terminated certain grants. Such relief does not constitute “money damages,” nor would such declaratory relief be available in the Court of Federal Claims.
Second, neither the district court‘s orders nor the plaintiffs’ claims in this case are premised upon the individual terms of the grant agreements. As an initial matter, the plaintiffs distinguish the “grants-in-aid” at issue here from traditional “contracts,” given that relevant statutory and regulatory provisions treat the two differently. See, e.g.,
In any event, the district court neither examined any of the plaintiffs’ grant terms nor interpreted them in reaching its ruling that the grant terminations must be set aside. Instead, as we have explained, the plaintiffs argued that the Challenged Directives are unlawful agency-wide policies because they violate various federal statutes and the Constitution -- classic examples of claims that belong in federal district court -- and that the terminations flowed directly from those unlawful policies. See
Again, Great-West and California are distinguishable. In Great-West, the Supreme Court specifically relied on the employee benefit plan‘s “reimbursement provision [as] the basis for the present lawsuit.” 534 U.S. at 207. And at least one of the respondents’ claims for relief in California depended on the terms and conditions of the grant awards, a fact that the government highlighted for the Court. See Application to Vacate at 16, California, 145 S. Ct. 966 (2025) (No. 24A910) [hereinafter “Appl.“]; Reply Br. at 9.
In sum, we conclude that the Department is unlikely to succeed in showing that the district court lacked “jurisdiction to review [the challenged] agency action . . . and to grant the complete relief authorized by § 706” of the APA. Bowen, 487 U.S. at 912. Instead, the court likely had jurisdiction to enter the orders here -- which provided declaratory relief under the APA independent of any contractual language -- to “set[] aside an agency‘s action[s]” as arbitrary and capricious; the fact that the orders “may result in the disbursement of funds” did not divest the court of its jurisdiction. California, 145 S. Ct. at 968 (citing Bowen, 487 U.S. at 910).5
2. Discretion
Next, the Department argues that the grant termination decisions were committed to agency discretion and are therefore unreviewable under the APA. It relies on Lincoln v. Vigil to assert that agency decisions to reallocate funds acquired via a lump sum appropriation (like NIH grant funds) are nonreviewable. See 508 U.S. 182, 184 (1993) (decisions “committed to agency discretion by law” are “not subject to judicial review under the [APA]” (quoting
The plaintiffs respond that the Department forfeited this argument because it did not raise it in its stay motion to the district court as required under
In any event, the Department, quoting Lincoln, concedes that “an agency is not free simply to disregard its statutory responsibilities: Congress may always circumscribe agency discretion to allocate resources by putting restrictions in the operative statutes.” 508 U.S. at 193. As the district court explained, that is exactly what Congress did here. There are numerous statutory provisions that direct NIH to prioritize or to consider certain research objectives -- including many that would seem to fall within the categories proscribed by the Challenged Directives. See Am. Pub. Health Ass‘n, 2025 U.S. Dist. LEXIS 125988, at *65 (citing, among other provisions,
3. Arbitrary and Capricious
Finally, the Department asserts that the grant terminations will ultimately be upheld under the arbitrary and capricious standard. Based on the briefing we have received to date, we think the Department has failed to carry its burden of showing that result is likely.
“The APA‘s arbitrary-and-capricious standard requires that agency action be reasonable and reasonably explained.” FCC v. Prometheus Radio Project, 592 U.S. 414, 423 (2021). To assess reasonableness, we look to whether the agency “examine[d] the relevant data and articulate[d] a satisfactory explanation for its action including a ‘rational connection between the facts found and the choice made.‘” Motor Vehicle Mfrs. Ass‘n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (citation omitted). And, when the agency enacts a decision that “rests upon factual findings that contradict those which underlay its prior policy; or when its prior policy has engendered serious reliance interests,” it must offer a “more detailed justification” than usual. FCC v. Fox Television Stations, Inc., 556 U.S. 502, 515 (2009); see DHS v. Regents of the Univ. of Cal., 591 U.S. 1, 33 (2020) (when agency is “not writing on a blank slate,” it is “required to assess whether there were reliance interests, determine whether they were significant, and weigh
Although our decision is not a holding on the merits, we see no obvious error in the district court‘s conclusion that the Department‘s actions bear all the hallmarks of arbitrary and capricious decision-making. To recap, the district court concluded that the Department‘s decisions rested on circular reasoning, included no explanation for the about-face in agency-wide policy, and entirely ignored significant reliance interests. For example, the court concluded that the prohibited categories of research grants were never defined, thus allowing the Department to terminate any grant that it wanted to, for any reason. See Am. Pub. Health Ass‘n, 2025 U.S. Dist. LEXIS 125988, at *52, *57. It also concluded that there was no indication in the record that anyone at NIH performed any analysis to support the conclusion that the forbidden categories of grants -- let alone the grants selected for termination -- were unscientific and/or wasteful. See id. at *56-57. To the contrary, after reviewing the “sparse” administrative record and hearing live testimony, the district court found, as a matter of fact, that the decisions about the prohibited categories, as well as which grants fell into those categories, were being “force-fed” to NIH by DOGE. Id. at *51.
In its stay motion, the Department asserts that “NIH and its [Institutes and Centers] reviewed their grant portfolios to identify and cancel specific grants that no longer serve agency priorities.” But the Department provides no record citation for this claim, and the district court found the exact opposite. To the extent the Department is leveling a challenge to the district court‘s factual determination, it does not cite to any contrary evidence in the record.
The Department also contends that the district court was wrong to conclude that grants were “indiscriminately terminated by topic,” pointing out that a few dozen “grants researching minority health” were permitted to continue. But the Department does not dispute the court‘s critical findings that the terminations of hundreds of other grants were unreasonable. For example, the court determined that the categorical language in the Termination Letters was not drafted by anyone at NIH; dozens of grants were terminated very shortly after being flagged by non-NIH staff members; and -- again -- no evidence of any individualized review of any grant material appears in the record. See Am. Pub. Health Ass‘n, 2025 U.S. Dist. LEXIS 125988, at *30-34, *39, *50-51.
Finally, the Department contends that the grantees’ reliance interests were adequately accounted for, because the “terminations provided for additional funds, where necessary.” The record citation the Department provides for this proposition is a single sentence in one of the Termination Letters that a university “may request funds to support patient safety and animal welfare to support an orderly phaseout of the project.” That sentence, which in any event is far from a guarantee of additional funds, does not account for the broad scope of financial and non-financial interests staked on the grant awards, including years of research and millions of hours of work. Nor does it have any bearing on whether the Department considered those myriad interests before issuing and implementing its Directives, which it was required to do under the APA. See DHS, 591 U.S. at 33.
Accordingly, we conclude that the Department has failed to meet its burden under the first Nken factor.
The second Nken factor requires the Department to demonstrate it will suffer irreparable harm absent a stay. See New York v. Trump, 133 F.4th 51, 71 (1st Cir. 2025). On this point, the Department begins by asserting that the district court‘s orders will impair “the President‘s ability to execute core Executive Branch policies.” But we have rejected this as a basis for irreparable harm in the past, insofar as it relies on the premise that the challenged agency action was lawful, in cases where we have concluded that the government has failed to show a likelihood of success on the merits. See id. at 71 (rejecting argument that court order irreparably harmed defendants by “intolerabl[y] intru[ding] on the prerogatives of the Executive Branch” because it rested on the premise that the challenged agency action was lawful).
Next, the Department contends that the district court‘s orders “will result in the immediate outflow of significant amounts of money with limited prospects for recovery.” The Department again relies on California to argue that this constitutes an irreparable harm. See 145 S. Ct. at 969 (crediting the government‘s irreparable harm argument that “it is unlikely to recover the grant funds once they are disbursed“).
The Supreme Court was relying on different facts to find irreparable harm in California. The plaintiffs here, unlike those in California, have cited specific federal regulations that provide for the Department‘s ability to recoup improperly expended funds, and the Department has not argued to us that those regulations are inapplicable. Further, the government‘s primary contention to the Court in California was that the short-term nature of a TRO would incentivize plaintiffs to draw down nearly $65 million in a matter of weeks. See Appl. at 25-26, 29. The district court‘s orders here, which are not time-limited, impose no such concentrated financial pressure.
Nevertheless, the Department is correct that in both California and in this case, the plaintiffs have not “promised to return all funds they receive[] as a result of the district court‘s order if it is ultimately reversed on appeal.” See 145 S. Ct. at 969. So, to the extent that the Department may be unable to recover some funds disbursed during the pendency of this litigation, we conclude that the Department has demonstrated an irreparable harm as California defines it.
Even so, the “stay inquiry calls for assessing the harm to the opposing party.” Nken, 556 U.S. at 435. The plaintiffs provided declarations explaining that the abrupt cutoff in funding will, among other things: cause their studies, some of which have been conducted over the course of many years, to “lose validity“; require animal subjects to be euthanized; force researchers with “project-specific knowledge and experience” to leave; delay treatment for patients enrolled in “clinical trials for life-saving medications or procedures“; and force the closure of community health clinics that provide preventative treatment for infectious diseases. Declarations submitted to the district court described that “[i]n many cases, there is no way to recover the lost time, research continuity, or training value once disrupted,” because studies and researchers cannot be held in stasis. Some declarants explained that the emergency short-term funding provided by their universities was not a sustainable solution and has required layoffs and research cuts; one declarant emphasized that “[u]sing alternative university funds to continue work . . . is neither possible . . . nor practical.” By contrast, the plaintiffs
In response, the Department fails to address any of the non-monetary harms that the plaintiffs detailed, which cannot be remedied by belated payment. Thus, the Department has failed to show that the plaintiffs would not suffer substantial harm if the district court‘s orders were stayed during the pendency of the litigation.
The final Nken factor asks us to consider “where the public interest lies.” 556 U.S. at 434 (citation omitted). The Department‘s one-sentence argument on this point is that “the district court exceeded its authority” by issuing its orders. The Department cites Coggeshall Development Corp. v. Diamond, where we explained that “[f]ederal courts do not have the power to order specific performance by the United States of its alleged contractual obligations.” 884 F.2d 1, 3 (1st Cir. 1989). But, as we have already explained, we do not agree that the Department is likely to succeed on its arguments that this is a breach of contract case that belongs in the Court of Federal Claims or that it did not violate the APA. And there is a substantial public interest “in having governmental agencies abide by the federal laws that govern their existence and operations.” League of Women Voters of the U.S. v. Newby, 838 F.3d 1, 12 (D.C. Cir. 2016) (cleaned up).
Further, the Department does not refute the plaintiffs’ contentions that a stay would result in the setback of “life-saving research by years if not decades” and would eliminate funding for “urgent public health issues.” These are serious concerns that suggest the public‘s interest is aligned with the plaintiffs, at least at this stage of the proceedings, not with the Department.
Accordingly, we cannot say that the balance of equities favors the grant of a stay. Although the Department may suffer some financial loss in the interim, it has neither quantified that potential loss nor provided any evidence that it will occur imminently. By contrast, the plaintiffs have provided concrete examples of economic and non-economic harms to themselves, to the public at large, and to the scientific and medical advancements of the United States if the stay is granted. The Department has failed to rebut plaintiffs’ arguments that these harms are weightier at this stage of the case, especially given our conclusion that the Department has failed to show a likelihood of success on the merits. See Nken, 556 U.S. at 427 (“A stay is an ‘intrusion into the ordinary processes of administration and judicial review,’ and accordingly ‘is not a matter of right, even if irreparable injury might otherwise result to the appellant.‘” (cleaned up)).7
IV. CONCLUSION
For all these reasons, the Department‘s motion for a stay is denied.