IPM Take
The antidepressant evidence base has a representation problem, and it is not new.
Three decades after the NIH Revitalization Act, many US antidepressant trials still fail to report race and ethnicity clearly. Several groups remain underrepresented. That means guidelines, labels and prescribing habits are still being shaped by evidence that does not fully reflect the people who will be treated.
This is not administrative tidiness. It is scientific accountability.
Executive Summary
A study published in The Journal of Clinical Psychiatry reviewed completed US antidepressant trials conducted between 1 January 1993 and 1 September 2025, using ClinicalTrials.gov records.
The analysis included 166 trials. Sex was reported by 90.36% of trials, but race by only 39.16% and ethnicity by 33.13%. Among trials reporting race or ethnicity, Asian participants, American Indian/Alaska Native participants, multiracial participants and Hispanic/Latino participants were underrepresented compared with state census benchmarks.
Female participants accounted for 58.27% of participants across trials, higher than the state-level benchmark but lower than the benchmark proportion among antidepressant users.
The authors concluded that reporting gaps and underrepresentation persist, limiting generalizability and supporting the need for clearer reporting standards and targeted recruitment strategies.
Why it matters
- Patients / advocates: Patients deserve evidence that reflects the populations who experience depression and receive antidepressants in real care.
- Clinicians: Underrepresentation weakens confidence that trial findings generalize across demographic groups and health-system contexts.
- Regulators: Reporting standards cannot remain optional if demographic gaps affect interpretation of safety and efficacy.
- Policymakers: Trial diversity is implementation infrastructure. Without it, equity failures are built into the evidence before care begins.
Depression is common. Antidepressant evidence is not equally built around everyone who lives with it.
That is the blunt implication of the new review. Most trials reported sex. Far fewer reported race or ethnicity. When race and ethnicity were reported, several groups were underrepresented compared with census benchmarks.
This should embarrass the field.
Antidepressants are used across populations shaped by different social stressors, comorbidities, access barriers, trauma histories, prescribing patterns and trust in medicine. A trial population does not need to mirror society perfectly to be useful. But when reporting is incomplete and underrepresentation persists for decades, the problem is not statistical noise. It is structural.
The consequence is practical.
A clinician reads a trial, a regulator reads a dossier, a guideline panel weighs evidence, and the underlying population is treated as if it is more complete than it is. Then implementation happens in real communities, where the evidence may be thinner for the people facing the greatest burden.
This is how inequity enters quietly.
Nobody has to write discrimination into a protocol for the result to be exclusionary. It can happen through site selection, recruitment methods, insurance requirements, language access, visit burden, trust, transport and eligibility rules that look neutral on paper.
The fix is not another paragraph saying diversity matters. It is enforceable reporting, serious recruitment plans and trial designs that treat representativeness as part of evidence quality.
If antidepressant trials shape care for millions, they must stop leaving so many patients out of the data.
