Back

The Dearth of Representation in FDA Approved Drug Trials

Ibilibor, C.; Armbruster, S.; Parker, R.; Yu, J.-R.; Barros, A.

2024-01-17 health policy
10.1101/2024.01.16.24301376 medRxiv
Show abstract

The generalizability of data derived from randomized controlled trials is of paramount importance given their utility in the Food & Drug Administration (FDA) drug approval process. An essential part of this process is the inclusion of reliably reported gender, race and ethnicity data in trials that lead to FDA drug approval. Despite previous mandates by the FDA and Clinicaltrials.gov, gender and race-specific data remains under reported. We reviewed 100 most recently approved FDA medications, and abstracted the clinical trial data from Clinicaltrials.gov that supported their approval. We then compared these FDA approved trials to non-FDA approved trials from the same year and of similar size. We found that 40% of the FDA trials were missing race/ethnicity information, while 24% of these trials did not include gender information. We demonstrate that there remains a significant amount of missing gender and racial/ethnic data in trials that lead to FDA-approved medications.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.