The scientific validity of medical research depends on the assumption that the population studied is representative of the population treated. For most of the history of clinical medicine, that assumption has been false. Research has been conducted disproportionately in white men. The findings have been applied to everyone.
This is not an artifact of early medicine that has been corrected. It is an ongoing, documented, and consequential problem.
How Exclusion Became Standard Practice
The systematic exclusion of women from clinical trials in the United States was effectively codified in 1977 when the FDA issued guidance recommending that women of childbearing potential be excluded from Phase I and early Phase II trials. The concern was teratogenicity — protecting potential fetuses from drug exposure. The consequence was a decades-long evidence base that could not answer basic questions about drug dosing, efficacy, and side effects in women.
The FDA reversed this guidance in 1993. The NIH Revitalization Act, also passed in 1993, required the inclusion of women and minorities in NIH-funded clinical research. Progress has been made. The inclusion mandate exists on paper. But inclusion is not the same as analysis: a trial can enroll women while being underpowered to detect sex differences in outcomes. Many do.
The Downstream Effects
Drug dosing that was established in male populations is administered to women. Side effect profiles documented in male-dominant trials underreport adverse effects in women. Diagnostic criteria developed in male cohorts produce false negatives in female patients. This is not theoretical. Zolpidem was prescribed at doses that caused next-morning impairment in women for decades before the FDA issued corrected dosing guidance in 2013. The cardiovascular symptom profile taught in medical schools — chest pain, arm pain, diaphoresis — is the male presentation. Women’s atypical symptoms are systematically underrecognized.
The Diversity Gap Within Women’s Health Research
Women who are included in clinical trials are not representative of all women. Enrollment skews toward white, educated, and insured populations. Black, Hispanic, Indigenous, and Asian women are underrepresented in proportion to their disease burden. Research that does not include diverse populations cannot produce guidelines that serve them.
SWHR and allied organizations have called for enrollment reporting that is not merely demographic box-checking but is designed to ensure that trial populations can support race- and sex-stratified analyses. The science cannot serve everyone if everyone is not in the science.
