TrialDiversityAtlas — Quantifying Demographic Representation Gaps in Global Clinical Trials

Abstract

Can clinical-trial enrolment be compared with real-world disease burden across demographic groups using only public registry and Global Burden of Disease data? We tagged completed ClinicalTrials.gov trials by MeSH area in the April 2026 AACT snapshot, targeting the cardiovascular and oncology cohorts. TrialDiversityAtlas defines a Representation Disparity Index comparing participant-weighted enrolled proportions, read from posted baseline tables, with expected burden-based prevalence. Enrolled female participation was 49.0% in cardiovascular trials and 56.2% in oncology (pooled 52.1%; 43.6 million participants) — at or above parity — but the expected denominator was uncomputable because the public Global Burden of Disease files are sex- and age-aggregated. The female result spans tens of millions of participants, but the enrolled age breakdown covered few trials, and the over-65 gap had no burden comparator. For these cohorts the data contradict simple female-underrepresentation and show the disparity index is not estimable from standard public exports. Fully stratified burden data and race/ethnicity reporting are needed.