Abstract Details

Equitable selection of subjects is required under 45CFR46. Representative participation is crucial to improve validity and generalizability of study results.

The purpose of this study was to investigate study participant representativeness within intramural clinical protocols, with an initial focus on epilepsy, to assess equitable selection of subjects.

A retrospective assessment of all intramural clinical studies from January 2005 to December 2022 was performed. Demographic data included race, ethnicity, zip-code, and home state was compared to CDC, US Census Bureau, and published literature estimates.

20797 participants across 331 protocols from 2005-2022, including 739 participants in four epilepsy protocols were analyzed (2008-2022). Compared to Census (2008-2021), overall White enrollment was lower (73.3% vs 77%) and Black enrollment was only slightly lower (13.1% vs 13.26%). Total Hispanic/Latino recruitment was lower (7.9%), compared to Census (17.4%). However, the epilepsy protocol recruitment compared to 2013-2015 CDC epilepsy data demonstrated higher Black Non-Hispanic cohort enrollment (29% vs 18%), and lower rate of White Non-Hispanic enrollment (56% vs 66%). Epilepsy participants on average for DC-specific enrollment were more likely to be Black Non-Hispanic (47% vs 31%) and less likely to be White Non-Hispanic (42% vs 53%). Rates of Hispanic/Latino were similar (5% epilepsy vs 4% overall; p=0.02 for overall proportions). Published estimates demonstrate significantly higher Black Non-Hispanic epilepsy prevalence in DC than is captured by cohort (74%; p=0.001).

Overall, demographic enrollment demonstrated lower White Non-Hispanic, higher Black Non-Hispanic, and comparable other racial groups compared to Census data. The largest disparity was a 9.5% difference in Hispanic/Latino enrollment, indicating a potential recruitment target. The epilepsy cohort revealed that while rates of Hispanic/Latino were comparable to the local DC population, there were significantly fewer Black Non-Hispanic subjects enrolled than expected, although the overall trend was a higher Black Non-Hispanic enrollment.