JMIR Publications

ABSTRACT

Background:

Advancements in science and technology can exacerbate health disparities, particularly when there is a lack of diversity in clinical research, which limits the benefits of innovations for underrepresented communities. With the U.S. population becoming increasingly diverse, it is crucial that clinical research studies reflect this diversity to improve health outcomes. A 2022 report by the National Academies of Medicine and Science emphasizes the need for a comprehensive approach to enhance inclusiveness in clinical research, particularly through better engagement between research teams and communities. However, limited data and small sample sizes in qualitative studies on the inclusion of underrepresented groups hinder progress in this area.

Objective:

Despite the rapid pace of discovery, science has been considerably less effective in translation into solutions that address persistent health disparities. This is largely due to lack of diversity in clinical trials. Because safety and effectiveness may vary in different populations, lack of diversity in clinical trial enrollment compromises healthcare delivery to excluded groups. Whereas others have posited barriers to clinical trial enrollment, little is known about what happens at the point of recruitment.

Methods:

This is the first analysis of conversations between Research Assistants (RAs) and prospective participants in the All of Us Research Program (AOURP). We used structural topic modeling to identify and compare latent topics of conversation in recruitment calls by estimating expected topic proportions in the corpus as a function of enrollment and participation in AOURP.

Results:

In total, our model estimated 45 topics of which 12 coherent topics were identified. Notable topics, that were more likely to occur in conversations between RAs and participants that enrolled and participated, include closing/following up to schedule an appointment, COVID protocols for in-person visits, explaining precision medicine and the need for representation, and working through objections. Topics among participants who did not enroll include technical challenges and describing physical measurement visits.

Conclusions:

Using a mixed methods approach that leverages machine learning to identify topical structure and themes with limited human subjectivity is a promising strategy to identify gaps in, and opportunities to improve, recruitment of underserved communities into clinical trials.