Analysis of methods to improve engagement of under-represented and socioeconomically deprived patients in clinical research

ABSTRACT

The current under-representation of racial and ethnic minorities and socio-economically disadvantaged participants in clinical trials represents an important problem, because it reduces generalizability of trial results and should urgently be addressed in all diseases. Decentralized trials may improve engagement of under-represented populations with long-standing health disparities and may be relevant to patients with pancreatic cancer who would benefit from at home trial participation. We completed a fully decentralized randomized double-blind phase II clinical trial in New York State for participants with mild-to-moderate COVID-19. Electronic data were collected for 28 days (5 vital readings and 1 survey with 20 questions per day) from 55 non-hospitalized participants. Home monitoring devices, HIPAA compliant data submission technology, and internet access were provided free-of-charge. We enrolled 40% White, 33% Black or African American and 27% Other/Mixed/Unknown participants. Of these, 25% self-identified as Hispanic or Latino. This exceeded national and New York state averages of minority populations, in contrast with the current clinical trial landscape. We found that the local area within a 30-minute return car journey from our main research hospital disproportionately over-represented socio-economically advantaged white inhabitants. We found that decentralization enabled the inclusion of participants living up to a 2-hour journey from this hospital in socio-economically deprived geographies with higher minority race representation. We excluded selection bias, by demonstrating that our trial population represented the differences of social deprivation observed between races at the national and state level (p = 0.003). In addition to trial enrollment, completeness of trial data has an important impact on the veracity of trial results. Half our participants were assigned a dedicated team member to make telephone call reminders if participants had not submitted data by mid-day despite of a pre-ceding automated notification to the supplied electronic device. Daily telephone follow-up significantly reduced missing electronic data in participants living above the median deprivation index (submitted data per day 4 out of 6 vs 6 out of 6 p = 0.03), thereby aiding equitable data collection from traditionally under-represented participant groups. Our findings require further validation and refinement from multiple centers and expansion to patients with reduced mobility and cachexia due to progressing pancreatic cancer. Also, other factors such as language assistance and recruitment methods, need to be addressed in clinical trials to mitigate against their negative impact on equitable recruitment. Nevertheless, for now we identify decentralization combined with engagement telephone calls as readily actionable methods to improve inclusion of under-represented participants in clinical trials.