A decade after the National Institutes of Health established its landmark Sex as a Biological Variable (SABV) policy, a Northwestern University analysis published in Nature Communications Medicine reveals a troubling pattern: the vast majority of NIH-funded researchers are failing to report outcomes by sex, perpetuating a systemic blind spot that may be costing lives.

The numbers are stark. Of 574 NIH-funded papers published between 2017-2024, only 44% reported results disaggregated by sex—despite 61% including both sexes as subjects. More alarming: 13% of publications don't even disclose the sex of their subjects at all. In non-human studies, where foundational biological discoveries occur, only 28% analyzed findings by sex.

This isn't merely an administrative oversight. It represents a structural failure in how we generate medical knowledge, with profound implications for treatment efficacy and patient safety across the gender spectrum.

THE HIDDEN COST OF MISSING DATA

The consequences of sex-agnostic research are well-documented. Women experience adverse drug reactions at nearly twice the rate of men, according to a 2001 GAO report that remains relevant today. Eight of ten prescription drugs withdrawn from the U.S. market between 1997 and 2001 posed greater health risks to women than men. The mechanism is clear: differences in body composition, hormonal environments, and drug metabolism create distinct pharmacokinetic profiles between sexes.

Yet the problem extends far beyond pharmacology. Cardiovascular disease research has historically focused on male subjects, leading to decades of delayed diagnosis in women whose symptoms present differently. Women experiencing heart attacks are 50% more likely to receive an incorrect initial diagnosis, according to research published in Circulation. The typical "crushing chest pain" taught in medical schools describes the male presentation; women more frequently experience fatigue, nausea, and back pain—symptoms often dismissed as anxiety.

The Northwestern study reveals this pattern hasn't improved despite explicit NIH guidance. Lead author Nicole Woitowich notes researchers often use vague language: "we used mice of both sexes" rather than specifying "10 female mice and 10 male mice." This opacity makes replication impossible and obscures whether findings apply universally or to specific populations.

WHY GUIDELINES FAIL: THE INCENTIVE PROBLEM

The SABV policy's fundamental flaw lies in its permissive structure. While it "expects" consideration of sex differences, it doesn't mandate analysis or reporting of sex-disaggregated results. Only 4% of papers in the Northwestern sample even discussed their rationale for how they approached SABV.

Anne Murphy, who served on the committee that drafted the SABV policy, articulated the disconnect: "It was finally recognition that we need to start looking at both sexes, but that was never the ultimate end goal. The end goal was to generate sex-specific knowledge to improve health outcomes for everyone."

The gap between intention and implementation reflects deeper structural problems in research funding and publication. Including both sexes increases study costs—requiring larger sample sizes and more complex statistical analyses. A 2016 analysis in Biology of Sex Differences estimated sex-balanced preclinical studies cost 15-20% more than male-only studies. For researchers facing tight budgets and "publish or perish" pressures, the path of least resistance remains clear.

Journal requirements compound the problem. While some high-impact journals now require sex-disaggregated reporting, enforcement remains inconsistent. Peer reviewers, themselves products of the same research culture, often don't flag missing sex analyses as grounds for rejection.

THE PRECLINICAL PIPELINE PROBLEM

The Northwestern findings are particularly troubling in non-human research, where only 41% of studies included both sexes. This matters because preclinical research forms the foundation for all subsequent human trials. If basic mechanistic studies use exclusively male animal models, downstream translational research inherits those blind spots.

Historically, researchers justified male-predominant animal studies by citing the "confounding" effects of female hormonal cycles. This reasoning has been thoroughly debunked—a 2014 meta-analysis in Neuroscience & Biobehavioral Reviews found that male mice actually showed more variable responses than females across most traits measured. Yet the preference for male models persists, likely due to tradition rather than scientific justification.

The implications cascade through the research pipeline. When Phase I safety trials build on male-predominant preclinical data, they may miss sex-specific toxicities. By the time problems emerge in Phase III or post-market surveillance, millions of patients may be affected.

BEYOND BINARY: WHAT THE ANALYSIS MISSED

The Northwestern study, while valuable, reveals limitations in how we conceptualize sex and gender in research. The authors grouped various analytical approaches together without distinguishing between studies that merely stratified by sex versus those that tested for sex interactions. Previous research has shown many scientists employ inappropriate statistical methods—testing males and females separately rather than including sex as a covariate or interaction term.

More critically, the study doesn't examine whether papers considered gender (social/cultural factors) distinct from biological sex, or how researchers defined sex itself. In an era of growing recognition that sex characteristics exist on a spectrum and gender identity affects health outcomes independently of biology, binary sex categorization may be insufficient.

A 2023 National Academies report on women's health research found that even as NIH's overall budget increased, the proportion allocated to women's health research declined. This suggests the problem isn't merely methodological but reflects deeper priorities about which questions get funded.

ENFORCEMENT WITHOUT TEETH

The NIH has promoted sex-inclusive research since establishing the Office of Research on Women's Health in 1990 and mandating inclusion of women in clinical trials in 1993. The 2016 SABV policy was meant to extend these principles to preclinical research and strengthen analytical requirements.

Yet without meaningful enforcement mechanisms, compliance remains voluntary. The NIH could require sex-disaggregated reporting as a condition of publication or future funding. It could mandate that grant applications include detailed plans for sex-based analysis with adequate statistical power. It could establish a registry where researchers pre-register their analytical plans, creating accountability.

Instead, the current system relies on researcher goodwill and journal policies that vary widely. The result, as the Northwestern data show, is widespread non-compliance masked by superficial inclusion.

THE PATH FORWARD: FROM GUIDELINES TO REQUIREMENTS

Several interventions could address this systematic failure:

Funding-level reforms: The NIH should require detailed sex-analysis plans in grant applications, including power calculations demonstrating adequate sample sizes to detect sex interactions. Studies proposing single-sex designs should be required to justify why sex differences are irrelevant to their research question.

Publication standards: Major journals should adopt and enforce mandatory reporting requirements. The SAGER (Sex and Gender Equity in Research) guidelines provide a framework, but voluntary adoption has proven insufficient. Journal editors should be trained to recognize inadequate sex-based analyses during peer review.

Statistical infrastructure: Researchers need better training in appropriate methods for analyzing sex differences. Many current practices—such as conducting separate analyses in males and females without testing for interactions—can miss genuine sex effects or generate false positives.

Incentive alignment: Academic promotion and tenure decisions should value rigorous sex-based analysis. Currently, the "least publishable unit" mentality incentivizes simpler studies with faster publication timelines.

Expanded definitions: Future guidelines should address gender identity, gender expression, and sex characteristics beyond binary categorization. Transgender and intersex populations face unique health disparities that current frameworks ignore entirely.

Alina Salganicoff, who co-chaired a National Academies committee on NIH women's health research, framed the stakes clearly: "A lot of the basic research that we already have is basically based on male models. We realized we need to do more, which is where the SABV comes in and tries to address that and fix that. But it's not clear that we are fixing it."

CONCLUSION: KNOWLEDGE GAPS AS HEALTH INEQUITY

The systematic failure to report sex differences in NIH-funded research represents more than methodological sloppiness. It reflects how research funding priorities create and perpetuate health inequities. When scientists default to male subjects—whether human or animal—they generate incomplete knowledge that gets translated into clinical guidelines, treatment algorithms, and pharmaceutical approvals.

Women, who represent 51% of the population, receive medical care based on research conducted predominantly in men. The consequences range from adverse drug reactions to delayed diagnoses to less effective treatments. And by focusing exclusively on binary sex categories, current research frameworks exclude transgender and intersex populations entirely.

The solution requires more than better guidelines. It demands structural changes in how research is funded, conducted, reviewed, and rewarded. After a decade of the SABV policy, the Northwestern data make clear that voluntary compliance and researcher education have failed. The next phase must include meaningful enforcement, adequate funding, and cultural transformation in how science values comprehensive, equity-focused knowledge generation.

The $47 billion question—reflecting NIH's approximate annual budget—is whether the agency will continue issuing guidelines without teeth, or finally implement requirements that ensure its funded research serves all populations equally. The answer will shape health outcomes for generations to come.