Targeted termination of scientific grants and minoritised researcher status in a national survey: a cross sectional analysis Authors: Rebecca Fielding-Miller, Natalie Vawter, Laramie R. Smith, Nicholas Metheny, Abigail M. Hatcher, Sarah Peitzmeier Published: 2026 ABSTRACT SummaryBackgroundBetween January and May 2025, the National Institutes of Health terminated over 2000 grants primarily relating to health equity, particularly for Black, Indigenous, and people of color (BIPOC) and sexual and gender minority (SGM) communities. 600 grants were terminated as part of blanket, institution-wide actions that were purportedly to “combat anti-Semitism.”MethodsN = 1918 investigators with grant terminations documented in the Grant Witness database were sent a unique link for an online demographic survey. We built logistic regression models to assess the odds of receiving different grant termination types, for BIPOC, SGM, female, and Jewish researchers.Findings941 investigators participated (49.1% response rate). Nearly half (48.6%) of investigators with equity-related terminations were BIPOC and 60.0% of investigators with gender-related terminations were SGM, including 16.5% who identified as transgender or non-binary. Additionally, 20.5% of investigators with institutional terminations (for purported antisemitism) were Jewish. Among investigators of terminated grants, BIPOC investigators had higher odds of receiving an equity-related termination (aOR = 2.69, 95% CI: 1.71–4.25 for BIPOC women and trans/nonbinary investigators, aOR = 1.82, 95% CI: 1.10–3.02 for BIPOC men) compared to White men. SGM investigators had 11 times higher odds of receiving a gender-related termination (aOR = 11.14, 95% CI: 6.85–18.12) than heterosexual, cisgender investigators, and White women and trans/non-binary investigators had double the odds (aOR = 2.24, 95% CI: 1.25–4.02) of a gender-related termination than cisgender White men.InterpretationGrant terminations targeting health research for BIPOC and SGM communities have disproportionately targeted the careers of BIPOC, SGM, and female researchers and researchers from disadvantaged backgrounds. Ensuring the health of all Americans will require systematic re-investment in these topics and scientists.FundingThe authors received no external funding for this study. Research in contextEvidence before this studyIn February 2025, the United States' National institutes of Health (NIH) began terminating thousands of federally funded research projects. The majority of these grants were targeted for termination because they focused on newly “deprioritized” research topics, particularly research on the health of Americans who are Black, indigenous, and people of color (BIPOC) and sexual and gender minorities (SGM). A smaller set of research grants were terminated not because of their topic area, but through an “institutional termination,” i.e. when all federal grants at a specific university were frozen or terminated in response to allegations of antisemitic actions. A robust literature has already documented preexisting funding disparities in NIH funding for BIPOC researchers, as well as women. Some of this funding disparity is due to the fact that BIPOC and women investigators are more likely to conduct research on BIPOC and women's health, and these topics tend to receive lower funding prioritization, particularly when there are fewer women or BIPOC reviewers represented on NIH peer review study section panels. However, to our knowledge, no original research has yet been conducted on how the wave of grant terminations may have exacerbated these funding disparities by disproportionately terminating the grants of BIPOC, women, and SGM investigators. A search of Pubmed using the terms “research grant termination”; and “grant termination” yielded no previously published studies, although a hand-search identified two products: a research letter which found that as of April 2025 the National Institutes of Minority Health and Health Disparities (NIMHD) had lost the greatest proportion of grants from January to April 2025, and a brief correspondence from June 2025 describing the career and community impacts of grant terminations on lesbian, gay, bisexual and transgender health.Added value of this studyWe obtained self-reported demographics from N = 941 investigators whose research was terminated between January 20 and May 30, 2025 and found that nearly half (48.6%) of investigators with equity-related terminations identified as BIPOC and 60.0% of investigators with gender-related terminations identified as a sexual or gender minority, including 16.5% as transgender or non-binary. Additionally, 20.5% of investigators affected by institutional terminations (for purported antisemitism) identified as Jewish. Among investigators of terminated grants, BIPOC investigators, SGM investigators, and investigators from a disadvantaged background are all more likely to have received a termination that specifically targeted their work for its topic, rather than an institutional termination for reasons that were topic-agnostic.Implications of all the available evidenceRecent federal grant terminations have disproportionately impacted minority scientists, amplifying existing inequities in NIH funding and threatening the diversity and quality of the U.S. scientific workforce. In the absence of future policy interventions to mitigate these external career shocks, the targeted disruption of BIPOC, SGM, and female scientist's careers may lead to declines in BIPOC, SGM, and women's health for years beyond the current moment. Research in context Evidence before this study In February 2025, the United States' National institutes of Health (NIH) began terminating thousands of federally funded research projects. The majority of these grants were targeted for termination because they focused on newly “deprioritized” research topics, particularly research on the health of Americans who are Black, indigenous, and people of color (BIPOC) and sexual and gender minorities (SGM). A smaller set of research grants were terminated not because of their topic area, but through an “institutional termination,” i.e. when all federal grants at a specific university were frozen or terminated in response to allegations of antisemitic actions. A robust literature has already documented preexisting funding disparities in NIH funding for BIPOC researchers, as well as women. Some of this funding disparity is due to the fact that BIPOC and women investigators are more likely to conduct research on BIPOC and women's health, and these topics tend to receive lower funding prioritization, particularly when there are fewer women or BIPOC reviewers represented on NIH peer review study section panels. However, to our knowledge, no original research has yet been conducted on how the wave of grant terminations may have exacerbated these funding disparities by disproportionately terminating the grants of BIPOC, women, and SGM investigators. A search of Pubmed using the terms “research grant termination”; and “grant termination” yielded no previously published studies, although a hand-search identified two products: a research letter which found that as of April 2025 the National Institutes of Minority Health and Health Disparities (NIMHD) had lost the greatest proportion of grants from January to April 2025, and a brief correspondence from June 2025 describing the career and community impacts of grant terminations on lesbian, gay, bisexual and transgender health. Added value of this study We obtained self-reported demographics from N = 941 investigators whose research was terminated between January 20 and May 30, 2025 and found that nearly half (48.6%) of investigators with equity-related terminations identified as BIPOC and 60.0% of investigators with gender-related terminations identified as a sexual or gender minority, including 16.5% as transgender or non-binary. Additionally, 20.5% of investigators affected by institutional terminations (for purported antisemitism) identified as Jewish. Among investigators of terminated grants, BIPOC investigators, SGM investigators, and investigators from a disadvantaged background are all more likely to have received a termination that specifically targeted their work for its topic, rather than an institutional termination for reasons that were topic-agnostic. Implications of all the available evidence Recent federal grant terminations have disproportionately impacted minority scientists, amplifying existing inequities in NIH funding and threatening the diversity and quality of the U.S. scientific workforce. In the absence of future policy interventions to mitigate these external career shocks, the targeted disruption of BIPOC, SGM, and female scientist's careers may lead to declines in BIPOC, SGM, and women's health for years beyond the current moment. Introduction The National Institutes of Health (NIH) funds $47 billion in scientific research annually—25 times the investment of the world's next largest biomedical and health research funder.1 The scale of NIH investment in biomedical research means that American priorities shape the direction of health science innovations globally. Despite the NIH's commitment to serving the health of all Americans, systematic barriers have impeded the pace and quality of NIH-funded research. Just one-third of the NIH's large research grants (R01s) are led by women,2 and a 2011 analysis of scientists whose work was supported by the NIH found that racial minority investigators were significantly less likely to be funded than White investigators with the same training and publication track records.3 This funding gap is particularly striking given that female and/or racial minority scientists are more likely to conduct innovative work but less likely to have their innovations acknowledged and rewarded when compared to their White male peers.4 Disparities in research funding and career progression are partially driven by what has been called the Matthews Effect, in which relatively minor initial advantages or disadvantages compound to create dramatic differences over the course of a researcher's career.5 Research grant applications related to racial minorities, health disparities, community-engaged research, women's health, and social-structural drivers of health tend to receive lower impact scores from peer reviewers, regardless of NIH-stated priorities.6 This trend is the most pronounced in study sections with the highest proportion of White male reviewers.7 Investigators who are women and/or racial minorities are significantly more likely to submit applications related to these topics, resulting in a small but systematic funding disadvantage for women and/or racial minority scientists.8,9 This effect compounds over time and appears to be a significant driver of racial and gender funding gaps, particularly for women of color.10,11 The consequences of these funding and priority gaps extend beyond the careers of individual researchers: By one estimate, they may have resulted in up to 26% fewer studies of gender-based violence, 25% fewer studies of women's reproductive health, and 20% fewer studies on the health of immigrants and minorities than would be expected if the proportion of Black, Indigenous, and people of color (BIPOC) and women researchers was representative of the US population.10 This funding gap runs counter to the ideals of scientific research funded by US taxpayers of all backgrounds, with a mission to ensure that the most innovative ideas with the highest potential to improve the health of all Americans are identified, funded, and integrated into national policy. In 2013, recognizing their mandate to improve the health of all Americans, the NIH began implementing incremental reforms to support the careers of outstanding underrepresented scientists, mitigate the role of implicit reviewer biases, and decrease the cumulative impact of the Matthews Effect in the funding process.12 In the 2019 “Notice of NIH's Interest in Diversity” (NOT-OD-20-031), the NIH provided a definition of “under-represented populations” based on racial/ethnic identity, socioeconomic status, disability, and gender. Shortly after January 20, 2025, this notice of interest in diversity was rescinded, and new agency priorities began to be implemented. Between February 28 and May 31, 2025, the federal government terminated roughly 2000 NIH grants as part of these revised priorities. An early analysis of the funding mechanism and topics of terminated grants found that 58% of terminated grants were to conduct research (R01, U01, UM1, U19, UM2) and 20% were career grants meant to support early career scholars (T32, K12, K01, K23, F31).13 The National Institute of Minority Health and Health Disparities (NIMHD) lost both the largest absolute number of grants and the largest proportion of their funded grant portfolio, followed closely by the National Institute of Nursing Research (NINR)—another Institute focused on promoting health equity.13 Grant terminations have been justified using two distinct sets of arguments. The first, which we label targeted, is that either the topic of study (for research grants) or the socio-demographic identity of the researcher (for fellowships and diversity supplements), has been deprioritized by new NIH leadership. The second justification, which we label institutional, is agnostic to research topic and investigator socio-demographics. Institutional terminations froze or rescinded every NIH-funded grant at several high-profile institutions as a tool to “combat campus anti-Semitism [sic]” (Executive Order 14188), with the first wave of terminations occurring at Harvard, Columbia, and Northwestern. Throughout this paper we will use the term “antisemitism” rather than “anti-Semitism” (as used in EO 14188), following guidance from the International Holocaust Remembrance Alliance: “The hyphenated spelling allows for the possibility of something called “Semitism,” which … legitimizes a form of pseudo-scientific racial classification that was thoroughly discredited by association with Nazi ideology”. Many investigators who received targeted terminations reported receiving letters that attributed the termination to either: a) being “diversity, equity, and inclusion (‘DEI’) studies,” i.e. studies that included a focus on the health of BIPOC or other minoritized communities, which we call equity-related terminations or b) being “unscientific” due to a focus on “gender identity,” i.e. studies that included a focus on the health of sexual and gender minority (SGM) communities, which we call gender-related terminations. We hypothesized that the NIH's systematic defunding of research related to BIPOC and SGM health through targeted terminations would also lead to a systematic defunding of BIPOC and/or SGM investigators, as well as systematic defunding of investigators along other axes of inequality such as female investigators and investigators from disadvantaged backgrounds. To test this hypothesis, we compared the demographics of investigators who received targeted terminations with those who received topic-agnostic institutional terminations. A secondary goal was to test the hypothesis that institutional terminations, while purportedly issued to support Jewish communities at the impacted universities, may have disproportionately defunded Jewish investigators, given the relatively large Jewish populations at universities that received such terminations. Methods We conducted a rapid demographic survey of investigators who had grants terminated between January 20 and May 30, 2025 for targeted and/or institutional justifications. Sample and data collection In April 2025, 2183 terminated grants were documented in the Grant Witness database.14 Grant Witness tracks National Institutes of Health (NIH) grant terminations via direct submissions from impacted investigators and by scraping government websites such as HHS TAGGS and NIH RePORTER. To our knowledge, it is the most complete dataset documenting federal research grant terminations that is publicly available. Full details are available at https://grant-witness.us. The publicly available dataset includes grant number, activity code, total funding envelope, and institution. Grant Witness lead investigators provided our team with a dataset that included investigator names, which are not available on the public-facing dataset. A team of volunteers then conducted internet searches to find institutional email addresses for all investigators in the dataset. Between June 24 and July 31, 2025, we sent each investigator up to three invitations to participate in a brief, self-administered, online demographic survey using the REDcap platform (see Appendix 1). Together, these crowd-sourced and publicly available data, combined with responses from our demographic survey, provided the data for this cross-sectional study. Measures No item was required, and participants could select “I do not feel comfortable answering this question” for any item. Outcome of interest: termination type Most investigators who had grants terminated were notified via form letters. The process was not systematic: While most investigators did receive a notification letter, some received no letter, some received individual emails from program officers, and some learned of their termination through their institution. The National Institutes of Health did not provide an official list of rationales for targeted grant terminations. However, we reviewed publicly available documentation and used social media and personal contacts to identify eight of the most common rationale statements used in termination letters. Survey participants could select more than one reason and were also provided with the option to select “another reason” and provide a brief description. The 8 options for grant termination were: (1) “Bolstering Chinese universities”, (2) “Diversity, equity, and inclusion,” or a letter that contained language related to “amorphous equity objectives”, (3) “Gender identity” or a letter that included language similar to “Research programs based on gender identity are often unscientific …”, (4) Vaccine hesitancy or a letter that included the language “research activities that focuses gaining scientific knowledge on why individuals are hesitant to be vaccinated …”, (5) “The end of the [COVID-19] pandemic”, (6) “Climate change”, (7) “Attempts to influence the public opinion”, or (8) “anti-Semitic [sic] harassment”. Four binary outcomes related to the type of grant termination were considered:•Participants were considered to have received a targeted termination if they indicated that their grant(s) had been terminated for reasons 1–7;•Participants were considered to have received an equity-related termination if they indicated their grant was terminated due to “amorphous equity objectives” (reason 2);•Participants were considered to have received a gender-related termination if they indicated that their grant termination was due to “gender identity” (reason 3); and•Participants were considered to have received an institutional termination if they indicated option 8 in the survey, and/or if they were based at Harvard, Columbia, or Northwestern University. •Participants were considered to have received a targeted termination if they indicated that their grant(s) had been terminated for reasons 1–7; Participants were considered to have received a targeted termination if they indicated that their grant(s) had been terminated for reasons 1–7; •Participants were considered to have received an equity-related termination if they indicated their grant was terminated due to “amorphous equity objectives” (reason 2); Participants were considered to have received an equity-related termination if they indicated their grant was terminated due to “amorphous equity objectives” (reason 2); •Participants were considered to have received a gender-related termination if they indicated that their grant termination was due to “gender identity” (reason 3); and Participants were considered to have received a gender-related termination if they indicated that their grant termination was due to “gender identity” (reason 3); and •Participants were considered to have received an institutional termination if they indicated option 8 in the survey, and/or if they were based at Harvard, Columbia, or Northwestern University. Participants were considered to have received an institutional termination if they indicated option 8 in the survey, and/or if they were based at Harvard, Columbia, or Northwestern University. Participants who endorsed option 1–7 and option 8 were classified as targeted terminations (i.e. a participant based at Harvard who indicated that their research was terminated because it addressed health equity was classified as receiving an equity-related termination). Correlates of interest Race/ethnicity Participants were asked to report their race (options: American Indian or Alaska Native, Asian, Black or African American, Middle East or North African, Native Hawaiian and other Pacific Islander, White) and ethnicity through a separate item (Hispanic or Latino). Participants were coded as BIPOC (Black, Indigenous, and people of color) if they endorsed any non-White racial identity or Hispanic or Latino ethnicity. Gender identity and sexual orientation Participants were asked about their gender (cisgender man, cisgender woman, genderqueer, non-binary, transgender man, transgender woman, or “I use a different term”) and sexuality (gay or lesbian, straight, bisexual, queer, or “I use a different term”). Respondent gender was re-categorized as cisgender man, cisgender woman, or transgender/non-binary. Write-in responses were recoded as one of the three categories based on study team discussion and consensus. Respondents were categorized as a sexual and gender minority (SGM) if they identified as transgender, nonbinary, gay, lesbian, bisexual, queer, or if they indicated they use a different term that could be reasonably categorized as SGM (e.g. “queer Femme”). Race × gender To understand the impact of the terminations at different intersections of race and gender, we constructed a categorical race/gender variable. Race was considered dichotomously as described above (BIPOC and White). When cell sizes permitted, we used a 6-category race x gender variable based on trichotomized gender (men, women, and trans/non-binary individuals); otherwise we used a 4-category variable based on dichotomized gender (men vs women, trans, and non-binary individuals). Socioeconomic background Participants were given the full NIH definition of “disadvantaged background” (briefly, measures of low socioeconomic status)15 and asked if they met the definition. Jewish identity Participants were asked whether they identified as Jewish (“yes”, “no”, or “it's complicated” with a write-in response option). The decision to use “it's complicated” and an open response box to allow participants to clarify their own perspective on their identity was chosen to reflect the fact that “the experience of being Jewish is complex, highly personal, and fluid over a lifetime,”16 and based on the first author's judgment as a Jewish woman in consultation with several Jewish-identified academics. We recoded the write-in responses of participants who indicated “it's complicated” (n = 14) using broad inclusion criteria that encompassed familial relationships, cultural affinity, and ethnic identity (e.g. “agnostic and culturally Jewish” or “half Jewish” were recoded as Jewish). Career stage Participants were asked to self-report their career stage (student, within 10 years of terminal degree, 10 years or more past terminal degree). Analysis Sample assessment We calculated response rates among investigators who received targeted terminations and investigators at universities that received institutional terminations. We calculated descriptive statistics and cross-tabulations to understand the demographic characteristics of investigators who received specific types of terminations (e.g. institutional, targeted, equity-related, gender-related). We compared racial and gender (men and women) demographics of those receiving targeted terminations to publicly available NIH-Funded Research Workforce data on the demographic characteristics of all NIH-funded principal investigators in 202417 using basic chi square tests. Data on trans identity, sexual orientation, and intersectional identities (e.g. race x gender) of all NIH funded researchers are not publicly available. Among our survey sample, we use multivariate logistic regression to model the odds of receiving each termination type (any targeted, equity-related, or gender-related) vs any other type of termination for each of our demographics of interest after adjusting for all other demographic characteristics. To see if findings were driven by the termination of workforce diversity mechanisms, we conducted a secondary analysis excluding diversity supplements from the dataset and reran the demographic characteristics and regression models to assess whether the results were qualitatively similar. Analyses were conducted in Stata v19. Missing data were handled with model-wise deletion. Ethical considerations This study was reviewed by the UC San Diego Institutional Review Board and determined to be exempt (Protocol 812631). All survey participants provided informed consent. Institutional email addresses, grant numbers, and demographic data were kept in three separate databases that could be linked with a unique ID. To maximize participant privacy, a staff research associate whose personal and professional networks did not overlap with survey participants merged these datasets and then shared a final deidentified dataset combining survey demographics and grant characteristics with the study team. Positionality The current study posits that an investigator's identity shapes the research questions they choose to investigate and the methodology with which they approach the question. As such, the positionality of the study team is salient to the research presented here. All authors have had one or more grant frozen, terminated, or removed from review. RFM is a tenured Associate Professor in a school of public health in California. She is a White cisgender Jewish woman who conducts research on gendered and racial health disparities in the United States and southern Africa with a focus on infectious disease and gender-based violence. NV is a Staff Research Associate in a school of public health in California. She is a White cisgender woman. LRS is an Associate Professor at a school of medicine in California who studies stigma and other syndemic drivers of HIV and substance use-related health inequities among medically marginalized communities. She is a queer multiethnic woman. NM is a tenure-track Assistant Professor in a school of nursing in Georgia. He is a White, cisgender, gay man whose research focuses on the prevention and mitigation of sexual violence among women and SGM. AMH is a research track Associate Professor of health behavior at a school of public health in North Carolina. She is White, cisgender woman whose research focuses on testing interventions for social determinants of health. SP is a tenured Associate Professor at the University of Maryland School of Public Health who researches gender-based violence prevention and LGBTQ+ health promotion. She is a multiracial Asian American woman. Role of the funding source The authors received no external funding for this study. Results Termination type and response rate Nine-hundred and forty-one of the 1918 investigators who received a unique invitation participated in the online survey between June 24 and July 31, 2025 (49.1% response rate). Forty respondents from institutions that received institutional terminations also reported receiving targeted termination letters. The response rate was higher for investigators who received targeted terminations (n = 725/1290, 56.2%), than investigators who received institutional terminations (n = 256/668, 38.3%). The large majority of investigators who received a targeted termination reported that the termination was equity-related (n = 530/725, 73.1%); 15.5% of these (n = 82/530) were diversity supplements. Sixteen percent of targeted terminations (n = 117/725, 16.1%) were gender-related. Twenty individuals attributed their grant termination to both equity and gender. The remaining targeted terminations were attributed to vaccine hesitancy (n = 39/725), the purported end of the COVID-19 pandemic (n = 19/725), and influencing public opinion (n = 5/725). Twenty-one respondents who did not receive institutional terminations indicated that they did not know why their project had been terminated (typical responses included “no longer aligns with NIH priorities” and “change in agency priorities”), suggesting that the 7 targeted categories we identified were sufficiently comprehensive. Demographic characteristics of terminated investigators Survey respondent demographics are shown in Table 1. Nearly half (n = 246, 48.6%) of investigators with equity-related terminations identified as BIPOC and over one third (n = 177, 35.0%) were from a disadvantaged background. Among investigators with gender-related terminations, 60.0% (n = 69) identified as SGM, including 16.5% (n = 19) as transgender or nonbinary. Additionally, 20.5% of investigators (n = 46) affected by institutional terminations identified as Jewish. Compared to the all NIH-funded investigators in 2024, investigators who received targeted terminations were more likely to self-identify as women (n = 406, 56.0% vs 42.4%), Hispanic/Latino (n = 122, 17.7% vs 7.0%), Black/African American (n = 86, 11.9% vs 4.0%), American Indian/Alaska Native (n = 19, 2.6% vs 0.3%), and Native Hawaiian/Pacific Islander (n = 5, 0.7% vs 0.1%) and less likely to identify as white (n = 484, 66.8% vs 70.2%) (Supplementary Table S1).Table 1Survey respondent demographics.Survey respondent demographicsFull sample N = 941Institutional termination N = 256Any targeted termination N = 725aEquity-related termination N = 530bGender-related termination N = 117bBIPOC trans/nonbinary individuals1.0% (9/859)1.4% (3/222)1.2% (8/674)1.4% (7/506)2.6% (3/116)BIPOC women24.0% (206/859)17.6% (39/222)26.1% (176/674)30.6% (155/506)10.3% (12/116)BIPOC men14.4% (124/859)11.7% (26/222)15.4% (104/674)16.6% (84/506)10.3% (12/116)White trans/nonbinary individuals2.9% (25/859)1.8% (4/222)3.4% (23/674)1.2% (6/506)14.7% (17/116)White women35.5% (305/859)38.3% (85/222)33.5% (226/674)31.4% (159/506)42.2% (49/116)White men22.1% (190/859)29.3% (65/222)20.3% (137/674)18.8% (95/506)19.8% (23/116)Sexual and gender minority21.8% (186/852)17.1% (38/222)24.3% (162/668)19.9% (100/503)60.0% (69/115)Disadvantaged background27.6% (235/853)12.5% (28/224)32.6% (218/668)35.0% (177/505)20.1% (23/112)Jewish13.3% (115/863)20.5% (46/224)11.5% (78/678)11.0% (56/511)17.2% (20/116)Career stage Student13.1% (117/894)17.3% (41/237)11.5% (80/697)12.8% (67/525)9.4% (11/115) Early (≤10 years from terminal degree)23.3% (208/894)31.7% (75/237)21.2% (148/697)19.6% (103/525)29.9% (35/115) Mid/Senior (>10 years from terminal degree)63.7% (569/894)51.1% (121/237)67.3% (469/697)67.6% (355/525)60.7% (71/115)a40 respondents reported both institutional and targeted terminations.b20 respondents reported both equity-related and gender-related terminations. Table 1Survey respondent demographics.Survey respondent demographicsFull sample N = 941Institutional termination N = 256Any targeted termination N = 725aEquity-related termination N = 530bGender-related termination N = 117bBIPOC trans/nonbinary individuals1.0% (9/859)1.4% (3/222)1.2% (8/674)1.4% (7/506)2.6% (3/116)BIPOC women24.0% (206/859)17.6% (39/222)26.1% (176/674)30.6% (155/506)10.3% (12/116)BIPOC men14.4% (124/859)11.7% (26/222)15.4% (104/674)16.6% (84/506)10.3% (12/116)White trans/nonbinary individuals2.9% (25/859)1.8% (4/222)3.4% (23/674)1.2% (6/506)14.7% (17/116)White women35.5% (305/859)38.3% (85/222)33.5% (226/674)31.4% (159/506)42.2% (49/116)White men22.1% (190/859)29.3% (65/222)20.3% (137/674)18.8% (95/506)19.8% (23/116)Sexual and gender minority21.8% (186/852)17.1% (38/222)24.3% (162/668)19.9% (100/503)60.0% (69/115)Disadvantaged background27.6% (235/853)12.5% (28/224)32.6% (218/668)35.0% (177/505)20.1% (23/112)Jewish13.3% (115/863)20.5% (46/224)11.5% (78/678)11.0% (56/511)17.2% (20/116)Career stage Student13.1% (117/894)17.3% (41/237)11.5% (80/697)12.8% (67/525)9.4% (11/115) Early (≤10 years from terminal degree)23.3% (208/894)31.7% (75/237)21.2% (148/697)19.6% (103/525)29.9% (35/115) Mid/Senior (>10 years from terminal degree)63.7% (569/894)51.1% (121/237)67.3% (469/697)67.6% (355/525)60.7% (71/115)a40 respondents reported both institutional and targeted terminations.b20 respondents reported both equity-related and gender-related terminations. Survey respondent demographics. 40 respondents reported both institutional and targeted terminations. 20 respondents reported both equity-related and gender-related terminations. Among our sample of investigators with terminated grants, BIPOC women and trans/non-binary people had nearly three times higher odds of having an equity-related termination compared to White cisgender men (aOR 2.69, 95% CI: 1.71–4.25) and BIPOC cisgender men had 1.82 times higher odds (95% CI: 1.10–3.02) than White cisgender men (Table 2). Investigators from disadvantaged backgrounds had 2.54 times higher odds of receiving an equity-related termination than non-disadvantaged investigators (aOR 1.75–3.69). White women, trans, and nonbinary researchers had more than double the odds of a gender-related termination compared to White cisgender men (aOR 2.24, 95% CI: 1.25–4.02). SGM investigators had more than 10-fold higher odds of gender-related termination compared to heterosexual, cisgender researchers (aOR 11.14, 95% CI: 6.85–18.12). Jewish investigators had 71% higher odds of having a grant terminated to address campus antisemitism than non-Jewish investigators (Supplementary Table S2).Table 2Adjusted logistic regression models.Model 1 (n = 818) Targeted vs institutional termination aOR (95% CI)Model 2 (n = 818) Equity-related vs other termination aOR (95% CI)Model 3 (n = 818) Gender-related vs other termination aOR (95% CI)BIPOC women/trans/nonbinary2.33 (1.32–4.13)2.69 (1.71–4.25)0.72 (0.34–1.52)BIPOC cisgender men1.82 (0.96–3.44)1.82 (1.10–3.02)0.88 (0.37–2.07)White women/trans/nonbinary1.18 (0.76–1.83)0.96 (0.66–1.39)2.24 (1.25–4.02)White cisgender men---ref------ref------ref---Sexual and gender minority2.93 (1.73–4.96)0.70 (0.55–1.14)11.14 (6.85–18.12)Heterosexual and cisgender---ref------ref------ref---Disadvantaged background5.58 (3.11–10.00)2.54 (1.75–3.69)0.57 (0.32–1.01)Non-disadvantaged background---ref------ref------ref---Career stage Student---ref------ref------ref--- Early career (0–10 years)2.01 (1.13–3.59)0.87 (0.52–1.44)3.09 (1.31–7.29) Mid/senior career (>10 years)5.30 (3.06–9.18)1.73 (1.09–2.75)3.34 (1.47–7.58) Table 2Adjusted logistic regression models.Model 1 (n = 818) Targeted vs institutional termination aOR (95% CI)Model 2 (n = 818) Equity-related vs other termination aOR (95% CI)Model 3 (n = 818) Gender-related vs other termination aOR (95% CI)BIPOC women/trans/nonbinary2.33 (1.32–4.13)2.69 (1.71–4.25)0.72 (0.34–1.52)BIPOC cisgender men1.82 (0.96–3.44)1.82 (1.10–3.02)0.88 (0.37–2.07)White women/trans/nonbinary1.18 (0.76–1.83)0.96 (0.66–1.39)2.24 (1.25–4.02)White cisgender men---ref------ref------ref---Sexual and gender minority2.93 (1.73–4.96)0.70 (0.55–1.14)11.14 (6.85–18.12)Heterosexual and cisgender---ref------ref------ref---Disadvantaged background5.58 (3.11–10.00)2.54 (1.75–3.69)0.57 (0.32–1.01)Non-disadvantaged background---ref------ref------ref---Career stage Student---ref------ref------ref--- Early career (0–10 years)2.01 (1.13–3.59)0.87 (0.52–1.44)3.09 (1.31–7.29) Mid/senior career (>10 years)5.30 (3.06–9.18)1.73 (1.09–2.75)3.34 (1.47–7.58) Adjusted logistic regression models. The effect size and significance of these relationships remained similar in a subanalysis that excluded diversity supplements (Supplementary Table S3). Discussion Our data document which scientists were most impacted by the wave of NIH grant terminations, and we found that grant terminations that target topics related to the health of specific communities also target investigators from those same communities. The grant terminations represented within our dataset mirror the variety of mechanisms through which the current administration has sought to reshape the research priorities of NIH-funded institutions and investigators. Participants who were SGM, BIPOC, and/or from disadvantaged backgrounds were significantly more likely to receive targeted terminations (i.e. where their work was explicitly targeted for its topic) than their less minoritized peers. These findings persisted when restricting the analysis to exclude investigators whose only terminated grant was a diversity supplement, indicating that this effect is not only produced through directly targeting workforce diversity mechanisms. Investigators in our sample who received targeted terminations were significantly more likely to be women, Black, Hispanic, American Indian/Alaska Native, and Native Hawaiian/Pacific Islander than the publicly available demographics of 2024 NIH-funded investigators. Nearly half of grants terminated for focusing on equity are held by BIPOC investigators. There are no current and publicly available data on the sexual orientation, gender identity (inclusive of transgender identities) of all NIH-funded investigators. However, it seems likely that SGM investigators have been disproportionately impacted given that 60% of investigators with gender-related terminations were SGM. Regardless of intent, the impact of these terminations was to target specific demographic groups as surely as they targeted specific topics. While many institutional terminations have been reinstated through negotiated agreements or legal action by universities, faculty associations, or attorneys general, targeted terminations based on topic have garnered less support from universities and may be more difficult to reverse. Even if reinstated, investigators will continue to have difficulty obtaining funding on these “deprioritized” topics, threatening their livelihoods and career advancement if they do not significantly modify their research agenda to align with current political priorities. Our findings highlight the importance of considering these impacts through an intersectional lens to account for the ways in which multiple axes of inequity operate simultaneously to produce outcome disparities. For example, BIPOC women had the highest odds of receiving any targeted termination, while the odds of a White woman receiving any targeted termination letter was not significantly different from that of a White man (the reference group)–echoing previous work on NIH funding success by race and gender.10,18 The elevated odds of termination for researchers from disadvantaged backgrounds also suggest the importance of considering class and socioeconomic status in future intersectional analyses—and NIH funding policy considerations. While institutional terminations were justified by the Office of Civil Rights as an intervention to combat campus antisemitism, we found that 1 out of every 5 investigators who lost grants via this mechanism were Jewish themselves, and that Jewish investigators were disproportionately more likely to lose their research funding from this termination mechanism than their colleagues who did not identify as Jewish. This is partially an artifact of the large academic Jewish community at Harvard, Columbia, and Northwestern where institutional terminations had occurred during our survey. Despite the purported intent to “protect” Jewish faculty and staff at these institutions, our findings question the utility of terminations justified via EO 14188 as a mechanism to support this community. As others have noted, any study attempting to accurately document the rationale for federal grant terminations is limited by a lack of transparency regarding how they were categorized and selected for termination.13,19 Some survey respondents selected more than one termination reason in the survey, indicating that investigators themselves may not know the primary reason for a grant termination. Additionally, while a 49% response rate is higher than average for an online survey,20 and significantly higher than standard response rates for online surveys of researchers in academic medicine,21 our findings must still be interpreted in the light of potential systematic response biases. Our sample of participants who received targeted terminations is consistent with previous findings on investigator demographics in health disparity research.6,22 However, the lower response rate among investigators who received institutional terminations, and differences between our respondent sample and the overall NIH workforce data, suggest that we may have oversampled women and younger investigators and under sampled older White men–a consistent trend in online survey responses.20 If older White men are systematically underrepresented in the institutional termination sample, it is possible that racial and gender disparities in grant termination type are even more pronounced than we found in our analyses. In June 2025 a federal judge ruled that targeted NIH grant terminations reflected “racial discrimination and discrimination against America's LGBTQ community” and “unmistakable discrimination against women's health,” leading him to reinstate a subset of targeted grants.19 However, even for the 61% of grants that have been reinstated,14 the temporary stop-work orders can have major impacts on human subjects research: Researchers who collaborate with minority communities often spend years overcoming their legitimate mistrust of academia and laying the groundwork for large research projects.23 This slow work of relationship building is seldom recognized or compensated by institutions. Losing a research project that is the culmination of this long, slow, work has the potential to damage or end these community relationships, either ending an entire line of research or requiring the investigator to invest significant (unpaid, unrecognized) effort to maintain or rebuild damaged trust. Because NIH funding is an integral part of career stability and progression for many biomedical researchers, even a single grant termination has the very real potential to disrupt or end investigators’ careers.22 This is particularly true for minoritized researchers. Bibliometric analyses have documented the phenomenon of “choice privilege”—in which BIPOC and women investigators are more likely to concentrate their work in areas related to their racial and gender identities than their White cisgender male peers.10 This means that targeted grant terminations are more likely to cause meaningful career disruption for BIPOC, SGM, and women investigators with deep expertise and a narrow research focus on “deprioritized” topics, when compared to investigators who received a targeted termination but may have a broader portfolio that includes other lines of research that they can fall back on. Systematically removing BIPOC, SGM, and women researchers from the scientific workforce will have ripple effects beyond the immediately terminated projects. Study sections are typically composed of senior investigators with significant subject matter expertise and a strong track record of NIH funding. Without deliberate interventions, large-scale grant terminations that systematically disrupted the careers of minority researchers will go on to result in fewer minority researchers at the study section table. In 2019, Hoppe et al. found that research applications which focused on socioeconomic, racial, and gendered health disparities had significantly lower than average success rates, regardless of study section or institute.6 This funding gap is largely driven by the discussion stage, at which point applications related to minority health disparities, socio-structural determinants of health and/or sexual and reproductive health tend to receive lower impact scores relative to other topics.6 Analyses of intramural research in the Veterans Administration—a process that is analogous to NIH study panels—has found that increasing the proportion of women and racial minorities on a study section panel can meaningfully counteract this: Women's applications had 5 times higher odds of being funded when reviewed by study sections with the highest proportion of female reviewers and racial minority investigators had three times higher odds of application success if 8% or more of reviewers were also racial or ethnic minorities.7 The scale of career disruption across the more than 1000 NIH-funded investigators who received targeted grant terminations risks creating distal impacts on the prioritization of research on health disparities for decades beyond the initial shock. The 2025 grant terminations we documented in this study, along with ongoing changes to NIH research priorities, will likely intensify the already significant racial and gender disparities in the NIH-funded workforce. Unless the voices of minority researchers are deliberately and systematically re-invited to the table, there will be long-term ramifications for the health of many communities. If the NIH is to serve the health of all Americans, it requires urgent course correction to ensure all scientific careers are sustained without targeted discrimination. Contributors The project was conceived and designed by RFM, SP, LRS, NM, and AMH. NV supported data curation and project administration. RFM led data analysis and manuscript drafting, with collaboration from SP. All authors had access to the underlying data and contributed to reviewing and editing the final draft. Data sharing statement We invite principal investigators of terminated grants to visit www.RebeccaFieldingMiller.com/frozenpipelines for information on how to participate in the survey or to collaborate on further analysis of this dataset. De-identified data that support the findings of this study are available from the corresponding author, RFM, upon reasonable request. Declaration of interests This project was unfunded, and the authors have no conflicts of interest to declare.