NIH Research Gap: Most Funded Studies Fail to Analyze Results by Sex

A groundbreaking analysis has revealed a critical shortfall in biomedical research: despite years of policy mandates, the majority of NIH-funded studies still do not stratify their results by sex. This oversight means that potentially vital differences in how diseases affect men and women, or how treatments perform across sexes, are often overlooked, with profound implications for clinical practice and public health. The study, led by researchers at Northwestern University Feinberg School of Medicine, examined hundreds of publications tied to NIH's prestigious R01 grants and found that while inclusion of both sexes has improved, rigorous analysis lags far behind.

This revelation underscores a persistent challenge in scientific rigor. The National Institutes of Health (NIH), the world's largest public funder of biomedical research, introduced the Sex as a Biological Variable (SABV) policy in 2016 to address historical biases favoring male subjects in preclinical studies. Yet, as recent data shows, translation into meaningful sex-stratified outcomes remains inconsistent, particularly in university-led research where much of this work originates.

Background on the NIH SABV Policy

The SABV policy requires NIH grantees to factor sex into research designs, analyses, and reporting for studies involving vertebrate animals or humans. Full name: Sex as a Biological Variable. This directive aims to enhance reproducibility and relevance by recognizing that biological sex influences physiology, disease susceptibility, and drug responses. For instance, hormonal cycles, genetics, and immune function vary by sex, yet pre-policy, most animal models used males exclusively to avoid 'variability' from estrous cycles—a justification now widely critiqued as scientifically flawed.

Implementation began in 2016 for new grants, with expectations for ongoing compliance. Universities, as primary recipients of R01 awards (over 30,000 active annually), play a pivotal role. Compliance training is often integrated into institutional review board (IRB) processes and grant writing workshops at major research institutions like Harvard, Johns Hopkins, and UC systems.

Key Findings from the Northwestern Study

Published April 27, 2026, in Nature Communications Medicine, the study scrutinized 574 publications from NIH R01 grants funded in fiscal years 2017-2018 across 21 institutes. Methodology involved linking grants via NIH RePORTER to PubMed articles, manually coding for sex inclusion, reporting, and analysis.

• 61% of studies included both sexes (350/574), up from 49% in a pre-policy baseline (2014-2016).
• Of both-sex studies, 83% reported sample sizes by sex, but only 44% performed sex-based analyses (e.g., as covariate or subgroup).
• Human-subject studies showed higher compliance: 75% inclusion, 55% analysis vs. non-human (46% inclusion, 30% analysis).

Single-sex studies (26%) were justified in 34% of cases (e.g., sex-specific diseases like prostate cancer). Variability existed by institute: National Institute of Child Health and Human Development (NICHD) at 75% analysis rate, vs. lower in others like NIAID.

Trends in Compliance Since Policy Launch

Progress is evident but uneven. Pre-SABV (2014-2016), ~49% inclusion; post-policy, 61%. Earlier audits, like a 2020 review, showed ~30-40% analysis rates. A 2021 survey of researchers indicated 33% altered designs for compliance, but awareness alone doesn't ensure execution.

Preclinical gaps persist: Non-human studies lag due to costs of larger samples for powered subgroup analyses. University labs, often resource-constrained, prioritize speed over stratification. Fields like neuroscience (male-biased models) show slowest gains.

Influence of Author Gender and Team Dynamics

A striking pattern: Studies with women as first authors conducted sex analyses 50% of the time vs. 39% for men (p=0.036). Woman-first and woman-last author teams had 2.24 odds ratio for analysis (95% CI: 1.33-3.79). Principal investigators (PIs): 64% male; first authors ~50/50 split.

This aligns with diversity research: Women researchers, facing underrepresentation (e.g., 37% NIH PIs), prioritize inclusivity. Universities fostering gender equity in leadership—via mentorship, tenure reforms—yield better SABV adherence. Programs like NIH's EDGE yield higher female-led sex-stratified outputs.

Photo by Austrian National Library on Unsplash

Real-World Examples of Missed Sex Differences

Failing to stratify masks critical variances. Heart disease: Women experience 'silent' symptoms (nausea vs. chest pain), leading to 50% higher misdiagnosis rates pre-treatment. Autoimmune diseases (78% female) like lupus understudied in male models.

Drug examples: Zolpidem (Ambien) halved dose for women after post-approval data showed next-day impairment 2x higher in females—costing millions in recalls. Digoxin (heart failure): Women 20-30% higher mortality risk due to unadjusted dosing. Oncology: Immunotherapies show sex-dimorphic responses, e.g., better male survival in melanoma.

FDA Ambien adjustment case highlights post-market fixes from overlooked preclinical gaps.

Health Impacts and Precision Medicine Ramifications

Women suffer 1.5-1.7x more adverse drug reactions (ADRs), per FDA data—$30-80B annual US cost. Men underrepresented in autoimmune research; both sexes lose from generic assumptions. COVID-19: Males higher mortality, females more long-haul—stratified trials enabled sex-tailored vaccines/boosters.

Precision medicine falters: Without sex data, algorithms bias (e.g., AI misreads female heart scans). University spin-offs delay sex-specific therapies, perpetuating disparities in chronic diseases (e.g., depression 2x female).

Challenges in University Research Environments

Resource limits hinder: Sex analysis requires 2x sample sizes for power, straining lab budgets. Journals rarely enforce (only 4% papers justify sex choices). Training gaps: Few PhD programs mandate SABV stats modules.

Preclinical bias: Males preferred for 'stability'; females underrepresented (e.g., 2014: 72% male rodents). Universities like Stanford, Yale offer SABV workshops, but adoption varies.

Expert Perspectives and University Responses

Nicole Woitowich (Northwestern): "Including women isn't enough—analyze or lose insights." Leah Welty: "Larger studies pay off for better drugs."

Institutions respond: Johns Hopkins SABV Consortium trains 1,000+ yearly. NIH workshops at AAMC emphasize compliance. Calls for journal mandates (SAGER guidelines) and PI incentives.

Pathways to Improvement: Enforcement and Training

• NIH: Stricter peer review checklists, funding tied to analysis plans.
• Universities: Integrate SABV in curricula, grants (e.g., NSF ADVANCE).
• Journals: Require sex-stratified CONSORT flowcharts.
• Tech: AI tools auto-flag missing analyses.

2026 outlook: Potential policy tweaks post-Trump admin reviews, but momentum builds via ORWH (Office of Research on Women's Health).

Photo by Google DeepMind on Unsplash

Future Outlook for Sex-Inclusive Research

Closing this gap promises equitable health advances. Universities investing in diverse teams, robust stats training will lead. As precision medicine evolves, SABV compliance becomes career-defining—positioning faculty for NIH success and impactful discoveries benefiting all sexes.