Skip to main content
NCBI home page
JAMA Network logoLink to JAMA Network
. 2026 Jun 4;9(6):e2616661. doi: 10.1001/jamanetworkopen.2026.16661

Federal Policy Changes and Online Diversity, Equity, and Inclusion Content Among US Medical Schools

Sreeveda Sreepada 1, Michelle Ko 1,✉
PMCID: PMC13237613  PMID: 42241000

Abstract

Importance

Recent policy changes have prompted US higher education institutions to reevaluate and, in some cases, dismantle their diversity, equity, and inclusion (DEI) efforts.

Objective

To describe how US medical schools displayed DEI language and DEI-related content on their public-facing websites and how this changed across 3 time points in 2025.

Design, Setting, and Participants

This cohort study reviewed public-facing website content of allopathic and osteopathic medical schools at 3 time points (May, July, and September 2025) to evaluate for the presence, removal, or replacement of DEI language and changes over time.

Main Outcome and Measures

Primary outcomes included the percentage of publicly accessible DEI website content and instances of its replacement, removal, or pending change. Institutional characteristics (degree type, ownership type, region, Minority Serving Institution status, and location in a state with a ban on DEI offices, programs, or initiatives) and content changes were compared using Pearson χ2 tests.

Results

A total of 195 school websites were reviewed. In May 2025, 109 schools (55.9%) had publicly accessible DEI content on their websites, 11 (5.6%) had replaced it with other names or structures, 69 (35.4%) had removed it, and 6 (3.1%) indicated content as pending. By July, 78 schools (40.0%) had accessible DEI content, 30 schools (15.4%) had replaced it, 78 (40.0%) had removed it, and 9 (4.6%) were pending. In September, 70 schools maintained accessible DEI content (35.8%), 41 schools (21.0%) had replaced it, 71 (36.4%) had removed it, and 13 (6.7%) were pending. A higher proportion of MD-granting schools maintained DEI content (58 of 151 [38.4%] vs 12 of 44 [27.3%]; P = .006). A higher proportion of schools in states with bans on DEI offices, programs, or initiatives removed content vs those located in states without bans (37 of 65 [56.9%] vs 34 of 130 [26.1%]; P < .001).

Conclusion and Relevance

In this cohort study of US allopathic and osteopathic schools, publicly accessible website DEI content shifted toward replacement of DEI language and removal of content altogether. Our findings suggest that institutional DEI communication was sensitive to legal, political, and regulatory pressures. Ongoing monitoring is needed to understand how US medical schools can best remain accountable to the public and maintain institutional autonomy to build a physician workforce that optimizes health care outcomes for all.

Key Points

Question

How did US medical schools present diversity, equity, and inclusion (DEI) in their public websites during a period of legal and policy changes?

Findings

In this cohort study of 195 allopathic and osteopathic schools, public DEI language was frequently removed or replaced over time, with substantial shifts observed between May, July, and September of 2025. Higher proportions of MD-granting schools maintained DEI content, while higher proportions of those in states with DEI bans removed it.

Meaning

These results suggest that medical schools changed messaging during a period of extensive external pressures, raising concerns about improving health care for all.

This cohort study of US medical school websites examines the removal or alteration of language about diversity, equity, and inclusion at 3 time points in 2025 following legal, political, and regulatory pressures.

Introduction

Diversity, equity, and inclusion (DEI) among physicians is associated with improved health care access, quality, and health outcomes.1,2,3 However, policymakers have increasingly pushed higher education institutions, including medical schools, to terminate programs and policies deemed related to DEI.

Diversity among physicians has been contested throughout the history of modern medicine. The Flexner Report of 1910 supported disproportionate closures of historically Black medical colleges, vastly shrinking the pool of Black physicians for generations.4 The Civil Rights Movement drove school integration, adoption of affirmative action, and immigration reforms, policies that set the path for diversifying the profession. Then, in the Bakke v. Regents of the University of California decision in 1978, the Supreme Court of the United States (SCOTUS) curtailed affirmative action to serve the purpose of “diversity,” rather than racial injustice.5 Over time, higher education institutions pivoted to adopt programs and policies emphasizing DEI.6 Progress stalled amid weak institutional support and rising opposition, with 9 states instituting bans on affirmative action in the 1990s.7 In 2023, SCOTUS ruled race-conscious admissions unconstitutional in Students for Fair Admissions v. Harvard.

Political and legal opponents of affirmative action have since expanded their scope from admissions to dismantling DEI in higher education. In 2025, multiple states banned DEI programming and policies,8 federal Executive Orders directed universities and accrediting bodies to terminate DEI practices,9 and federal funding agencies eliminated training programs that support underrepresented students.10 Although many directives addressed unlawful discrimination by race and sex, officials have targeted DEI programs more broadly, irrespective of activities or legality. By May 2025, the Commission on Osteopathic College Accreditation (COCA), Liaison Committee for Medical Education (LCME), and the Accreditation Council for Graduate Medical Education (ACGME) had all removed or suspended DEI requirements from their standards.11,12,13

Against this backdrop, we examined how US medical schools publicly presented DEI language in their publicly accessible websites. While previous research has examined DEI in training program websites14,15 and school mission statements,16,17 to our knowledge, none has reviewed language during more recent policy changes. Although online language does not equate with institutional commitment, changes in public-facing messaging can signal how schools are adapting to this new climate. We also assess associations with medical school characteristics to provide insight on what types of institutions maintained or changed their language and content.

Methods

Design

We conducted an observational cohort study from March to September 2025. We report our study in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines for cohort studies. The UC Davis institutional review board reviewed the protocol and determined the study as not human participants research.

Study Population

The population consisted of all accredited allopathic (MD) and osteopathic (DO) medical schools in the US, and those with pending accreditation, as listed by the Association of American Medical Colleges (AAMC) and the American Association of Colleges of Osteopathic Medicine (AACOM) as of March 31, 2025 (204 schools total). For schools with multiple campuses, we analyzed the website of the main campus, as not all campuses host independent websites on school programs and services. The final study sample thus consisted of 195 US medical schools.

Data Sources

We searched schools’ websites, supplemented by checks with 2 external sources: Do No Harm,18 an advocacy organization whose stated intent “seeks to highlight and counteract divisive trends in medicine, such as ‘Diversity, Equity and Inclusion’”(70 schools),18 and the Chronicle of Higher Education DEI Tracker,19 which tracks university- and state-level DEI policies. The majority of the data was collected from primary sources, as the external sources tracked only a subset of schools.

Data Collection

We collected data at 3 points (1) between May 1 and May 31, 2025; (2) July 3, 2025; (3) September 5, 2025, spanning multiple policy and institutional changes (timeline of federal, state, and accreditor policy and guidance changes available in eFigure in Supplement 1). Prior to May 2025, several events had likely already prompted changes in school practices, including the federal Executive Orders 14151 and 14173 to eliminate DEI in federal programs, contractors, and grantees; Executive Order 14279, which named both LCME and ACGME as enabling unlawful discrimination in medical education; and bans on DEI in 12 states issued via legislation, executive order, or budgetary action.8 The initial data collection in May required extensive search and documentation process described above. We re-examined sites in July, assuming that schools had sufficient time to adjust to accreditor changes and new state laws. As of September, some universities had entered negotiations to eliminate DEI programming (among other actions, such as affirming biological sex rather than gender) to receive federal funding, and a handful of judges had issued injunctions against federal actions against DEI.8,19 The ACGME removed its DEI accreditation standards and closed its Department of DEI on September 5.11 We conducted the third data collection to check for additional changes, given the potential for replacing formerly removed language or content, or in response to judicial decisions. Of note, COCA proposed permanent elimination of DEI elements on October 9, 2025.20

Search Process

We reviewed school websites for direct content and links to DEI offices, programs, positions, and initiatives. We then explored school sites in the content areas of admissions, students, residents, faculty, organizational mission, values, strategic plans, and annual reports (terms and search process available in eTables 1 and 2 in Supplement 1). Lastly, we used schools’ local search function, eg, if a school provided a window or box which allowed for internal website search, to identify other content potentially missed through the primary protocol and to return archived content. At follow-up assessments, we reviewed content using previously saved links, external DEI trackers, and by rechecking school sites.

The first author (S.S.) selected an initial sample of 20 sites for review and selected 3 sites for representation across potential outcomes. Both authors then met and reviewed the 3 sites in detail, refined search process and definitions, and then repeated the process for another 5 sites. After finalizing search process, terms, and outcomes, the first author then collected data from the remaining sites.

Measures

We defined our outcome of DEI language as using 1 or more of the terms diversity, equity, and/or inclusion, either in title, body text, or navigation label, as corresponding to the text in Executive Order 14151.21 For the purposes of searching for DEI content, we also included terms identified by the Chronicle of Higher Education DEI Tracker, and a database of position descriptions of DEI leaders from a prior study (eTable 1 in Supplement 1).19,22 We included only website content referring to school-organized efforts, and excluded those led by students, residents, staff, or faculty, such as affinity groups and social events.

We defined content status as publicly accessible, replaced, removed, or pending. Publicly accessible material could be located and viewed on at least 1 or more pages as a member of the general public. Sites that remained accessible in follow-up periods and with no identifiable changes were categorized as maintained. Replaced content was accessible but had significant alterations such that 1 or more DEI terms had been eliminated and new language, terms, and/or structures were in their place, determined by a comparison of the navigation links or archived site information. We also documented new terms used. Removed content was when search results, either directly from the school website or identified as live or archived pages from the internal search tool, indicated prior content, such as a former site with URL (uniform resource locator) that subsequently displayed an error message, returned a broken link, redirected to another page, or was listed as no longer available. Pending content had an active link on the website, but with content that described an office, program, or website that was still in progress, under construction, pending, or otherwise forthcoming, with no other sites found.

To determine whether DEI content had been replaced, we first searched and tested school site links labeled with the terms described above. Returned notices such as “404 error” or “under construction” prompted additional checks as to whether another office, program, or positions site had appeared in its place. When possible, we compared URLs directly to confirm that former pages had been redirected, deleted, or edited. If initial search or internal search tools identified archived documents, we also examined those to verify whether terms had changed. If no replacement site could be identified, then we assigned status accordingly (removed or pending).

We also collected the following information from school sites, AAMC, and AACOM: degree (MD or DO), ownership (public, private nonprofit, private for-profit, or federal), and state (categorized into US Census regions: Northeast, South, Midwest, West, and US territories). We referenced school sites and cross-checked with the US Department of Education21 for a Minority-Serving Institution (MSI) designation (such as Hispanic Serving, Asian American Native American and Pacific Islander Serving Institutions, and Alaska Native and Native Hawaiian-Serving Institutions), as well as Historically Black Colleges and Universities (HBCUs) and Tribally Controlled Colleges and Universities.

To describe state policies and actions that banned or terminated offices, initiatives, and programs labeled as DEI in public institutions of higher education, we referenced the Chronicle of Higher Education DEI Legislation tracker, supplemented by additional search and review of relevant state actions. Because of our focus on programmatic language, we did not include state actions limited to one higher education component, such as a ban on diversity statements in faculty hiring, or a ban on mandatory diversity training, when other activities remained permissible. We then created indicators for whether a school was located in a state which enacted a DEI ban as of May 2025, between June 1 and July 1, 2025, or never enacted a ban. There were no additional state actions on DEI from July 1 to September 5, 2025. We also created an indicator for location in a state with no DEI ban during the study period vs any DEI ban.

Statistical Analysis

We tabulated outcomes at each time point and described trends across time periods. We used Pearson χ2 tests to conduct exploratory, descriptive analyses of associations between varying school characteristics and net outcomes from May to September 2025. We used the Bonferroni correction to adjust for multiple comparisons, such that we interpret statistical significance at P < .01. We used Stata SE version 17 (StataCorp). We also conducted an exploratory qualitative thematic examination of replacement terms.

Results

Of the 195 US medical schools included in the study, 151 (77.4%) offered MD degrees and 44 (22.6%) offered DO degrees (Table 1). Half were publicly owned (98 [50.3%]), over one-third located in the South (72 [36.9%]), and one-fifth designated as Minority-Serving Institutions (MSIs) (41 [21.0%]). At baseline, 48 (24.6%) were located in states with an existing ban on DEI activities; by the end of the study period, this increased to 65 schools (33.3%).

Table 1. US Medical School Characteristics, 2025.

Characteristic Schools, No. (%) (N = 195)
Degree type
MD 151 (77.4)
DO 44 (22.6)
Accreditation
LCME 149 (76.4)
COCA 36 (18.5)
Preaccreditation 7 (3.6)
Unclear or none 3 (1.5)
Ownership
Public 98 (50.3)
Private (nonprofit) 91 (46.7)
Private (for-profit) 5 (2.6)
Federal 1 (0.5)
Years in operation
0-5 14 (7.2)
6-22 35 (18.0)
23-45 8 (4.1)
≥46 135 (69.2)
Not yet 3 (1.5)
Region by main campusa
Northeast 41 (21.0)
South 72 (36.9)
Midwest 44 (22.6)
West 35 (18.0)
US territory 3 (1.5)
Minority-serving institutionb
Yes 41 (21.0)
No 150 (76.9)
Unclear 4 (2.1)
State ban on DEI in higher educationc
No 130 (66.7)
Yes (as of May 2025) 48 (24.6)
Yes (as of September 2025) 65 (33.3)
Open in a new tab

Abbreviations: COCA, Commission on Osteopathic College Accreditation; LCME, Liaison Committee on Medical Education.

a

Note that some colleges have campuses in multiple regions: we count them as one and use only the main campus.

b

Includes Historically Black Colleges and Universities, Tribally Controlled Colleges and Universities, and those identified by the Department of Education Hispanic Serving Institutions, Asian American Native American and Pacific Islander Serving Institutions, and Alaska Native and Native Hawaiian-Serving Institutions. Four schools had conflicting documentation on minority-serving institution status.

c

Includes legislation, budget elimination, and/or executive orders. Counts reflect successive tallies of schools, such that all schools at baseline are included in final total at study end period.

At baseline, a slight majority of schools (109 [55.9%]) had publicly accessible DEI website language (Table 2). Over one-third had evidence of removed DEI content (69 [35.4%]), and 11 (5.6%) had replaced their language. Publicly accessible DEI sites were more common among MD vs DO schools (90 of 151 [59.6%] vs 19 of 44 [43.2%]; P = .002). Higher proportions of schools in the Northeast (33 of 40 [82.5%]) and Midwest (30 of 44 [68.2%]) had publicly accessible DEI language vs those in the South (24 of 72 [33.3%]) (P < .001). Of schools located in states with bans on DEI enacted at baseline, 33 (68.7%) had removed DEI content and 3 (6.3%) had replaced content, vs 8 (5.5%) and 35 (24.5%), respectively, of schools in states without a ban (P < .001).

Table 2. US Medical School Characteristics and Publicly Accessible Online Content on Diversity, Equity, and Inclusion (DEI), May 2025.

Characteristic Total, No. Schools, No. (%) P valuea
Pending Removed Replaced Accessible
All schools 195 6 (3.1) 69 (35.4) 11 (5.6) 109 (55.9) NA
Degree type
MD 151 1 (0.7) 52 (34.4) 8 (5.3) 90 (59.6) .002
DO 44 5 (11.4) 17 (38.6) 3 (6.8) 19 (43.2)
Ownership
Public 98 2 (2.0) 43 (43.9) 5 (5.1) 48 (49.0) .23
Private-nonprofit 91 4 (4.4) 23 (25.3) 5 (5.5) 59 (64.8)
Private-for profit 5 0 2 (40.0) 1 (20.0) 2 (40.0)
Federal 1 0 1 (100) 0 0
Region by main campusb
Northeast 40 0 6 (15.4) 1 (2.6) 33 (82.5) <.001
South 72 3 (4.2) 42 (59.2) 3 (4.2) 24 (33.3)
Midwest 44 1 (2.3) 12 (27.9) 1 (2.3) 30 (68.2)
West 34 2 (5.9) 6 (17.6) 6 (17.6) 20 (58.8)
US territory 3 0 2 (66.7) 0 1 (33.3)
Minority serving institution
Yes 41 1 (2.4) 19 (46.3) 1 (2.4) 20 (48.8) .36
No 150 3 (2.0) 49 (32.7) 10 (6.7) 88 (58.7)
Unclear 4 2 (50.0) 1 (25.0) 0 1 (25.0)
State ban on DEIc
No 147 6 (4.0) 8 (5.5) 36 (24.5) 97 (66.0) <.001
In effect 48 0 33 (68.7) 3 (6.3) 12 (25.0)
Open in a new tab

Abbreviations: COCA, Commission on Osteopathic College Accreditation; LCME, Liaison Committee on Medical Education.

a

The threshold for significance in multiple testing was P < .001. Comparisons by region and state ban were the only significant results.

b

Main campus location for schools with multiple regional locations

c

Includes legislation, executive order, and budget actions to ban or terminate offices, programs, and initiatives related to DEI.

The majority of shifts occurred from May to July 2025, in which 31 schools with formerly publicly accessible content replaced language (15), removed their content (13), or changed to pending status (3) (Figure; eTable 3 in Supplement 1). While 13 schools that had initially removed their content reintroduced sites with replacement language, overall shifts led to minimal net change.

Figure. Status of Publicly Accessible Diversity, Equity, and Inclusion (DEI) Content on US Medical School Websites in 2025.

Figure.

Open in a new tab

We observed the following net changes from May to September: 70 schools (35.8%) had maintained publicly accessible DEI language, while 41 (21.0%) had replaced language, 71 (36.4%) removed their content, and 13 schools’ (6.7%) sites remained in a pending status (Table 3). A greater proportion of MD vs DO schools maintained their content (58 of 151 [38.4%] vs 12 of 44 [27.3%]; P = .006). A majority of schools in the Northeast maintained content (54.8%), whereas fewer than 1 in 5 in the South did (23 of 42 [54.8%] vs 14 of 72 [19.4%]; P < .001 for comparisons across all outcomes by region). Of schools located in states with DEI bans, 37 (56.9%) had removed DEI language; in states without bans, just under half (60 of 130 [46.2%]) maintained their DEI content (P < .001). Differences by ownership and MSI status were not significant.

Table 3. US Medical School Characteristics and Net Changes in Website Content on Diversity, Equity, and Inclusion (DEI), From May to September 2025.

Characteristic Total, No. Schools, No. (%) P valuesb
Maintaineda Replaceda Removeda Pendinga
All schools 195 70 (35.8) 41 (21.0) 71 (36.4) 13 (6.7)
Degree type
MD 151 58 (38.4) 32 (21.2) 56 (37.1) 5 (3.3) .006
DO 44 12 (27.3) 9 (20.4) 15 (34.1) 8 (18.2)
Ownership
Public 98 28 (28.5) 24 (24.5) 43 (43.9) 3 (3.1) .02
Private (nonprofit) 91 41 (45.0) 14 (15.4) 26 (28.6) 10 (11.0)
Private (for profit) 5 1 (20.0) 3 (60.0) 1 (20.0) 0
Federal 1 0 0 1 (100) 0
Region by main campusc
Northeast 42 23 (54.8) 9 (21.4) 8 (19.0) 2 (4.8) <.001
South 72 14 (19.4) 13 (18.1) 39 (54.2) 6 (8.3)
Midwest 44 20 (45.4) 7 (15.9) 15 (34.1) 2 (4.5)
West 34 13 (38.2) 12 (35.3) 6 (17.7) 3 (8.8)
US territory 3 0 0 3 (100) 0
Minority serving institution
No 154 59 (38.3) 36 (23.4) 50 (32.5) 9 (5.8) .07
Yes 41 11 (26.8) 5 (12.2) 21 (51.2) 4 (9.8)
State ban on DEI
No 130 60 (46.2) 27 (20.8) 34 (26.1) 9 (6.9) <.001
Yes 65 10 (15.4) 14 (21.5) 37 (56.9) 4 (6.2)
Open in a new tab
a

Includes legislation, executive order, and budget actions to ban or terminate offices, programs, and initiatives related to diversity, equity, and inclusion.

b

The threshold for significance in multiple testing was P < .001. Comparisons by region and state ban were the only significant results.

c

Main campus location for schools with multiple regional locations.

In the course of reviewing replaced language, we found several common themes. This included student support (access, opportunity), organizational culture (inclusive excellence, belonging), academics, engagement, and community (Table 4).

Table 4. Main Concepts and Examples of US Medical School Public Website Language Replacing Programming, Offices, or Initiatives Related to Diversity, Equity, and Inclusion (DEI) in 2025.

Main concept Examples
Student support

  • Opportunity

  • Access

  • Well-being

  • Empowerment
Organizational culture

  • Inclusive excellence

  • Belonging

  • Academic culture

  • Humanism
Academic

  • Academic excellence

  • Student success

  • Enrichment
Engagement

  • Engagement

  • Community engagement

  • Institutional engagement

  • Outreach
Community

  • Community

  • Community impact

  • Unity

  • Collaboration
Professionalism

  • Professional development

  • Professional proficiency
Health and health care

  • Health equity

  • Primary care

  • Culturally inclusive care
Other

  • Infinite worth

  • Hope
Open in a new tab

Discussion

We examined how US medical schools presented DEI content on their public websites during a time of growing legal and political uncertainty. At baseline, over one-third of schools had already removed prior content. Over the ensuing 4 months, the proportion of schools with public websites that used language associated with DEI declined, although one-fifth had retained sites but replaced the content.

The rapid increase in schools with language changes during (rather than prior to) the study period occurred in concert with additional state legislation banning DEI offices, programming, and initiatives in July. By the end of the study period, schools located in states with DEI bans accounted for half of all schools that had removed their DEI content. While 40 out of the 65 schools in these states are publicly owned (and only 1 maintained DEI language), only 9 out of the 25 privately owned schools in these states also maintained DEI language, suggesting additional contextual influences on communication choices. State bans and accompanying unobserved political, cultural, or social influences may also drive regional differences, as 49 schools subjected to state bans were located in the US South.

Shifts also occurred among schools not subject to state bans. The study period coincided with alterations to accreditation standards, which did not require schools to terminate DEI but signaled the loss of regulatory support provided by accrediting bodies. During the study period, the administration also froze or threatened to freeze federal funds to a small number of universities and requested schools eliminate DEI programs. The National Institutes of Health (NIH) issued, retracted, and then reinstated new NIH grant conditions requiring institutions and investigators to certify they did not operate DEI or participate in programs. National associations also signaled their retreat from support in several ways, such as eliminating the Group on Diversity and Inclusion (AAMC) and the Center for Health Equity (American Medical Association).

Altering language over the study period is consistent with prior work showing that DEI presence on institutional websites is often limited and weak, even without the political and legal pressures in place during the study period. Studies of training program websites found that while many programs mentioned DEI, few provided detailed or transparent information, while others have shown listing diversity in school mission statements rose after 2007 but plateaued in 2021.14,16 Prior scholars in higher education have found that some schools employ diversity language as a safe way to signal positive institutional values, while avoiding engagement in deeper reforms that would challenge organizational hierarchies.23,24 A 2025 study of US schools of law websites found changes in DEI content following the SCOTUS 2023 ruling, even though the language and policies were not subject to the ruling.25 In short, public-facing online DEI content may be relatively malleable because the connections to institutional actions are inconsistent, particularly if schools are looking to comply with legal mandates or funding threats.

A small group consistently hosted sites linked to DEI labels, but the content itself remained “under construction” or was otherwise signaled to have content pending. This group predominantly consisted of those with pending accreditation. These findings may reflect institutional uncertainty on how to respond to federal policies, accreditation changes, and/or efforts to maintain DEI commitments without triggering legal scrutiny. Of note, newly accredited medical schools are not more likely to have greater student diversity, even under prior accreditation standards.26

We found some differences by medical school characteristics. A higher proportion of MD- vs DO-granting schools maintained DEI content early in the study period, which may reflect differences in institutional resources, accreditation expectations, or alignment with national policy norms.26,27 In contrast, MSI-designated schools retained or reframed their DEI content rather than remove it entirely. These schools are committed to serving historically underrepresented students and thus prioritize diversity-related messaging.28 However, MSI classifications are tenuous, as the US Department of Education has announced that it will not defend the program to legal challenges, nor will it continue grants (although HBCU and tribal college funding remain intact).21

We note that, nevertheless, a slight majority of schools overall retained public indication regarding inclusive learning environments even without using exact DEI language. Changes in terminology are not new: universities have pivoted from racial justice to affirmative action to diversity following legal actions and societal norms.6,24 Schools later added equity and inclusion in acknowledgment that representational diversity does not necessarily lead to improved outcomes for minoritized groups. Then and now, leaders in higher education have justified language changes as necessary for legal compliance, strategic moves, or to better reflect institutional values.29 Evidence from states and the more recent SCOTUS decision indicates that the switch from affirmative action to DEI led to substantial declines in Black and Latinx medical school enrollment,30,31 but because DEI activities are far more heterogeneous in form and impact,22 the downstream effects of language changes and/or shifting programming are not clear. A study of undergraduate institutions found that, although DEI offices were lost at schools subjected to state DEI bans, these schools were more likely to gain new student affinity groups.31 Boatright et al32 have developed a tool for measuring student perceptions of equity and inclusion within medical schools that elaborates on multiple school-related dimensions that may affect internal climate.

Symbolic changes in medical schools’ public messaging still carry notable implications. For example, removal of all content could suggest to prospective students that DEI is not an institutional priority, even though a 2025 study found medical students (other than those identifying as White and male) named school diversity as a priority.33 Students have previously recognized gaps between public statements and internal climate and actions; when commitments disappear from public view, applicants from historically excluded communities may feel even less welcome or supported.29,34 Even if schools have internally retained institutional values, renamed programs, or moved work to other units, public-facing institutional language may influence who applies and feels they belong.

In addition, changes in public-facing language raise broader concerns about academic freedom and institutional autonomy. Schools may modify or eliminate their language and content not because their missions have changed, but because they are reacting to legal pressures and political scrutiny, especially if they believe it could jeopardize funding or accreditation.35 Programs with a general aim to advance diversity, equity, and inclusion—without preferential treatment by gender, race, or ethnicity—do not violate antidiscrimination law.36 For the majority of schools not subject to state bans, by complying in the present context, schools signal the potential for increasing restrictions on their autonomy. These include control and oversight of curricular content (such as whether schools may be permitted to discuss gender or race in the context of health care disparities), school admissions (federal oversight that ignores the multitude of factors evaluated for medical school admission), and clinical education (as already seen in requirements to discontinue abortion care and gender-affirming care).19,37

Finally, the current scrutiny of language may encourage schools to explore alternative strategies that are not labeled affirmative action or DEI to support the growth of a physician workforce equipped to care for the diversity of communities across the US, based on where care is needed. For example, HBCUs and the University of California Programs in Medical Schools (PRIME) have demonstrated success by recruiting and selecting students using a holistic process that includes criteria based on a mission to care for underserved communities.38,39 Residency programs are similarly developing mission-driven application review processes.40 As of March 26, 2026, the Department of Justice Office of Civil Rights has initiated investigations into 3 medical schools (in addition to an investigation into the David Geffen School of Medicine at UCLA), on the argument that mission-driven admissions are a proxy for racial and ethnic preferences.41 Schools and policymakers can also consider investment in pathway programs, revisions of faculty advancement and promotion criteria to recognize teaching and mentorship in relation to institutional mission, and expansion of scholarship and loan repayment programs in exchange for service.42

Limitations

This study had several limitations. We only assessed publicly available website content and could not observe internal policies or practices. Some schools may have kept DEI efforts in place but removed or relocated them without publicly presenting them online. Further investigation is needed to provide rich detail on institutional impacts. Additionally, the language used on websites may not fully reflect the intentions or priorities of institutional leadership. For example, prior studies have found limited or no associations between school social mission and student or faculty diversity.43,44 Thus, conversely, we cannot conclude that schools were employing effective DEI programming prior to this period. In the wake of the renewed racial justice movements of 2020, school investment in antiracism and, secondarily, DEI remained spotty.22 The study also included only 3 time points in 2025, and we could not track changes before or after that period. We found that several schools already had evidence of changed language but could not identify the timing. Since data collection ended, federal, state, and accreditor actions continued to evolve. In March of 2026, the LCME published updated standards for 2027-2029, in which curriculum content standards would no longer include health disparities or inequities.37 Future research should continue to track DEI content over time and expand to examine institutional climate and changes in recruitment, retention, and career outcomes.

Conclusions

In this cohort study of US medical schools, we found schools quickly shifted their public messaging on DEI. These changes occurred in concert with broader political influences and legal imperatives and may have lasting implications for medical education, institutional trust, and transparency, and ultimately, whether schools of medicine have the autonomy to design and implement programs to serve the needs of all communities.

Supplement 1.

eTable 1. Search terms used for examination of US medical school website content on diversity, equity, and inclusion

eTable 2. Content areas for initial search and identification of diversity, equity, and inclusion and related terms on US medical schools’ websites

eTable 3. US medical schools’ publicly accessible online content on diversity, equity, and inclusion (DEI) and state restrictions on DEI at 3 time points in 2025

eFigure 1. Timeline of federal, state, and accreditor actions regarding diversity, equity, and inclusion, in 2025

jamanetwopen-e2616661-s001.pdf (779.5KB, pdf)
Supplement 2.

Data Sharing Statement

jamanetwopen-e2616661-s002.pdf (16.5KB, pdf)

References

  • 1.Peek ME. Increasing representation of Black primary care physicians—a critical strategy to advance racial health equity. JAMA Netw Open. 2023;6(4):e236678. doi: 10.1001/jamanetworkopen.2023.6678 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Snyder JE, Upton RD, Hassett TC, Lee H, Nouri Z, Dill M. Black representation in the primary care physician workforce and its association with population life expectancy and mortality rates in the US. JAMA Netw Open. 2023;6(4):e236687. doi: 10.1001/jamanetworkopen.2023.6687 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Ma A, Sanchez A, Ma M. The impact of patient-provider race/ethnicity concordance on provider visits: updated evidence from the medical expenditure panel survey. J Racial Ethn Health Disparities. 2019;6(5):1011-1020. doi: 10.1007/s40615-019-00602-y [DOI] [PubMed] [Google Scholar]
  • 4.Campbell KM, Corral I, Infante Linares JL, Tumin D. Projected estimates of African American medical graduates of closed historically Black medical schools. JAMA Netw Open. 2020;3(8):e2015220. doi: 10.1001/jamanetworkopen.2020.15220 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Strelnick AH, Taylor VS, Williams B, et al. Diversity in academic medicine no. 3 struggle for survival among leading diversity programs. Mt Sinai J Med. 2008;75(6):504-516. doi: 10.1002/msj.20081 [DOI] [PubMed] [Google Scholar]
  • 6.Berrey E. The Enigma of Diversity: The Language of Race and the Limits of Racial Justice. University of Chicago Press; 2015. doi: 10.7208/chicago/9780226246376.001.0001 [DOI] [Google Scholar]
  • 7.Ballejos MPUS. US Supreme Court decision implications for affirmative action in medical school admissions. Acad Med. 2024;99(4):349-350. doi: 10.1097/ACM.0000000000005617 [DOI] [PubMed] [Google Scholar]
  • 8.DEI legislation tracker. Chronicle of Higher Education. Accessed September 8, 2025. https://www.chronicle.com/article/here-are-the-states-where-lawmakers-are-seeking-to-ban-colleges-dei-efforts
  • 9.Reforming Accreditation to Strengthen Higher Education. Executive Order 14279. April 23, 2025. Accessed September 8, 2025. https://www.whitehouse.gov/presidential-actions/2025/04/reforming-accreditation-to-strengthen-higher-education/
  • 10.Blinder A. As Trump attacks universities, some are agreeing to negotiate. New York Times. April 23, 2025. Accessed September 8, 2025. https://www.nytimes.com/article/trump-university-college.html
  • 11.Accreditation Council for Graduate Medical Education . Accreditation Council for Graduate Medical Education. September 8, 2025. Accessed December 8, 2025. https://www.acgme.org/newsroom/e-communication/2025/september-8-2025/
  • 12.Commission on Osteopathic College Accreditation . New & Developing Accreditation Standards. Letter to members. April 25, 2025. Accessed December 17, 2025. https://osteopathic.org/index.php?aam-media=/wp-content/uploads/Suspension-of-COCA-Standards-Related-to-DEI.pdf
  • 13.Liaison Committee on Medical Education . LCME voted to eliminate Element 3.3 (diversity programs and partnerships). May 19, 2025. Accessed July 5, 2025. https://lcme.org/announcement-may-19-2025/
  • 14.Sheets ZC, Nguyen M, Hardin B, et al. Diversity statements and pediatric residency program and department characteristics. JAMA Netw Open. 2025;8(5):e2513013. doi: 10.1001/jamanetworkopen.2025.13013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Mallicoat BC, Herstine BAS, Kelly EM, et al. A review of OB/GYN, internal medicine, family medicine, and pediatrics residency program websites for diversity, equity, and inclusion elements. J Grad Med Educ. 2023;15(3):316-321. doi: 10.4300/JGME-D-22-00329.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Stowers J, Mahajan A, Atkinson K, Malmut L. Charting the use of inclusive language in medical school diversity mission statements: a national study. J Natl Med Assoc. 2026;118(1):77-82. doi: 10.1016/j.jnma.2025.10.006 [DOI] [PubMed] [Google Scholar]
  • 17.West K, Oyoun Alsoud L, Andolsek K, Sorrell S, Al Hageh C, Ibrahim H. Diversity in mission statements and among students at US medical schools accredited since 2000. JAMA Netw Open. 2023;6(12):e2346916. doi: 10.1001/jamanetworkopen.2023.46916 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Do No Harm . Medical schools with DEI offices. Accessed September 3, 2025. https://donoharmmedicine.org/dei-tracker/
  • 19.Cutler S, Baiocchi A, Gretzinger E, Hicks M, Dutton C, Smith J. Tracking higher ed’s dismantling of DEI. Chronicle of Higher Education. Updated April 6, 2026. Accessed September 5, 2025. https://www.chronicle.com/article/tracking-higher-eds-dismantling-of-dei
  • 20.American Osteopathic Association . COCA announces revisions to accreditation standards. October 9, 2025. Accessed April 8, 2026. https://osteopathic.org/accreditation/announcements/
  • 21.US Department of Education . Eligibility designations for higher education programs. Updated April 1, 2026. Accessed October 24, 2025. https://www.ed.gov/grants-and-programs/grants-higher-education/eligibility-designations-higher-education-programs
  • 22.Esparza CJ, Simon M, London MR, Bath E, Ko M. Experiences of leaders in diversity, equity, and inclusion in US academic health centers. JAMA Netw Open. 2024;7(6):e2415401. doi: 10.1001/jamanetworkopen.2024.15401 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Ahmed S. On Being Included: Racism and Diversity in Institutional Life. Duke University Press; 2012. [Google Scholar]
  • 24.Thomas JM. Diversity Regimes. Rutgers University Press; 2020. [Google Scholar]
  • 25.Leong N. Diversity messaging after affirmative action. Minn Law Rev. 2025;109(3). [Google Scholar]
  • 26.Oyoun Alsoud L, West K, Sorrell S, Andolsek KM, Al Hageh C, Ibrahim H. A cross-sectional study of newly established medical schools in the United States: student body diversity remains an unmet challenge. Med Educ Online. 2025;30(1):2487660. doi: 10.1080/10872981.2025.2487660 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Ko M, Henderson MC, Fancher TL, London MR, Simon M, Hardeman RR. US medical school admissions leaders’ experiences with barriers to and advancements in diversity, equity, and inclusion. JAMA Netw Open. 2023;6(2):e2254928. doi: 10.1001/jamanetworkopen.2022.54928 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Dens Higano J, Tilburt JC, Hafferty FW. Words matter: Tracing the implicit meaning of diversity language (and its absence) in medical school mission statements. J Natl Med Assoc. 2023;115(1):18-25. doi: 10.1016/j.jnma.2022.12.006 [DOI] [PubMed] [Google Scholar]
  • 29.Nguemeni Tiako MJ, Ray V, South EC. Medical schools as racialized organizations: how race-neutral structures sustain racial inequality in medical education-a narrative review. J Gen Intern Med. 2022;37(9):2259-2266. doi: 10.1007/s11606-022-07500-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Florescu N, Lin A, Temucin S, Rae L. Affirmative action repeal and racial and ethnic diversity in US medical school admissions. JAMA Netw Open. 2025;8(10):e2535020. doi: 10.1001/jamanetworkopen.2025.35020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Schachle-Gordon JL, Coley JS, Tetteh D. Fighting back? The contradictory effects of anti-DEI laws on DEI centers and student-of-color groups at US colleges and universities. Socius. Published online March 25, 2025;11. doi: 10.1177/2378023125132564240857005 [DOI] [Google Scholar]
  • 32.Boatright D, Nguyen M, Hill K, et al. Development of a tool to measure student perceptions of equity and inclusion in medical schools. JAMA Netw Open. 2024;7(2):e240001. doi: 10.1001/jamanetworkopen.2024.0001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Nguyen M, Chaudhry SI, Hajduk AM, et al. Student and faculty diversity in medical school selection. JAMA Netw Open. 2025;8(10):e2533727. doi: 10.1001/jamanetworkopen.2025.33727 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Ghanem N, Goldberg DG, Granger E, Warren JR, Gimm G. A critical qualitative study to understand current black women medical student perspectives on anti-racist reform in US medical education. Med Educ Online. 2024;29(1):2393436. doi: 10.1080/10872981.2024.2393436 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Feingold J. Academic freedom requires DEI; the AAUP gets it. Inside Higher Education. January 9, 2025. Accessed September 8, 2025. https://www.insidehighered.com/opinion/views/2025/01/09/academic-freedom-requires-dei-aaup-gets-it-opinion
  • 36.Brennan T, Cole D. Diversity, equity, and inclusion programs in academic medical centers—their importance and legality. JAMA. 2025;334(1):21-23. doi: 10.1001/jama.2025.6643 [DOI] [PubMed] [Google Scholar]
  • 37.Oza A. As political pressure mounts, medical school accreditor drops requirement to teach about health equity. STAT News. March 27, 2026. Accessed September 8, 2025. https://www.statnews.com/2026/03/27/medical-schools-dei-lcme-drops-structural-competency/
  • 38.Rodríguez JE, López IA, Campbell KM, Dutton M. The role of historically black college and university medical schools in academic medicine. J Health Care Poor Underserved. 2017;28(1):266-278. doi: 10.1353/hpu.2017.0022 [DOI] [PubMed] [Google Scholar]
  • 39.Henderson MC, Fancher TL, Murin S. Holistic admissions at UC Davis—journey toward equity. JAMA. 2023;330(11):1037-1038. doi: 10.1001/jama.2023.15872 [DOI] [PubMed] [Google Scholar]
  • 40.Mieczkowski AE, Dakroub A, Szymusiak J, Gonzaga AM. Building the building blocks: development of a mission-driven residency application scoring rubric for holistic review. Acad Med. 2025;100(8):914-918. doi: 10.1097/ACM.0000000000006070 [DOI] [PubMed] [Google Scholar]
  • 41.Bender MC, Blinder A. Trump Administration begins inquiries into 3 medical schools in show of power. New York Times. March 26, 2026. Accessed September 8, 2025. https://www.nytimes.com/2026/03/26/us/politics/trump-medical-schools-civil-rights.html
  • 42.Guilamo-Ramos V, Thimm-Kaiser M, Benzekri A, Shim RS, Amankwah FK, Rosenbaum S. Overcoming the impact of Students for Fair Admission v Harvard to build a more representative health care workforce: perspectives from ending unequal treatment. Milbank Q. 2024;102(4):853-867. doi: 10.1111/1468-0009.12718 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Campbell KM, Tumin D, Linares JI, Morley CP. The relationship between the social mission content of medical school mission statements and minority faculty representation among faculty and senior leadership. J Immigr Minor Health. 2024;26(2):334-340. doi: 10.1007/s10903-023-01555-5 [DOI] [PubMed] [Google Scholar]
  • 44.Campbell KM, Tumin D, Linares JLI, Porterfield L, Kisel T. Changing missions of medical schools and trends in medical student diversity. Fam Med. 2023;55(7):481-484. doi: 10.22454/FamMed.2023.928475 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1.

eTable 1. Search terms used for examination of US medical school website content on diversity, equity, and inclusion

eTable 2. Content areas for initial search and identification of diversity, equity, and inclusion and related terms on US medical schools’ websites

eTable 3. US medical schools’ publicly accessible online content on diversity, equity, and inclusion (DEI) and state restrictions on DEI at 3 time points in 2025

eFigure 1. Timeline of federal, state, and accreditor actions regarding diversity, equity, and inclusion, in 2025

jamanetwopen-e2616661-s001.pdf (779.5KB, pdf)
Supplement 2.

Data Sharing Statement

jamanetwopen-e2616661-s002.pdf (16.5KB, pdf)

Articles from JAMA Network Open are provided here courtesy of American Medical Association

RESOURCES