Bridging gaps in health care: The impact of investigator diversity on clinical trials

Clinical trials can lead to new, innovative solutions that improve health care, but often they lack the patient diversity to be generalizable.  A randomized survey experiment from J-PAL affiliated researcher Marcella Alsan and coauthors Romaine Campbell, Ayotomiwa Ojo, and Lukas Leister seeks to shed light into how increasing the racial diversity of clinical trial investigators can affect the willingness of Black respondents to participate in a trial. In an interview with the researchers, we delve into their motivations, findings, and the broader implications of their work to increase racial representation in research. 

What motivated you to explore the subject of racial representation in clinical trials?

Our project was motivated by large and persistent racial health disparities in the United States and the lack of representation in current clinical trials for new technologies. Our previous survey experiment showed that lower racial representation in the data for clinical trials can lead to lower take-up rates by Black patients and physicians of new technologies. That project also found that it is less costly to recruit white individuals, which is another factor that can lead to overall lower diversity in study populations. 

This project is concerned with the step before data analysis—study enrollment—and considers whether an investigator’s racial background could influence trial participation and who is included in a study’s results. We recruited Black respondents and randomized them to be shown photographs of National Institutes of Health (NIH) investigators who varied across gender and racial background. We then assessed respondent willingness to participate in a trial led by that investigator to understand the impact of the different investigator characteristics on clinical trial participation. 

Another major motivation was the lived experience of the people on our team. Ayotomiwa Ojo is an OB-GYN resident and has seen firsthand how Black women in labor are treated differently and receive less access to novel technologies:  

 “I  frequently encounter patients who worry whether treatments and protocols will work for them because they know that Black people are less likely to participate in studies, or they do not trust that those developing the treatments have their best interests in mind. As a provider, I want to find solutions so that I can reassure them and provide them with the standard of care.” 

What were the key findings of the study?

We found that Black respondents were more likely to participate in a clinical trial if the investigator was race-concordant—when the race of a patient matches that of their health provider or investigator. Additionally, gender concordance had no effect on participation willingness. We also asked study respondents to rate the NIH investigators on perceived attractiveness, quality, and trustworthiness and found that the perceived trustworthiness and attractiveness of an investigator were the most important factors mediating the effect of concordance.

How does this study build on the broader literature on racial concordance, including your previous work which examined how race concordance affects the take-up of preventive health services by Black men?

We believe this study broadly fits into the theory of the second best. In other words—in an optimal situation, we would have a different health care system that cared for all equally regardless of income or background. We do not have such a system today. Instead, we have a setting where people have legitimate reasons to distrust the health care system and the biomedical industry and face financial barriers to seeking care. Reducing frictions by providing the opportunity to see a physician or work with an investigator with whom they feel more comfortable might help under such circumstances.

What were some of the challenges you faced while running this study?

The underlying question in our study is whether greater representation of Black clinical trial investigators could improve racial representation among trial participants. However, for various logistical reasons, it was not feasible to match participants to trials. Doing so would have allowed us to track actual behaviors rather than stated preferences for participating in clinical trials. Nonetheless, our study documents an important finding and provides insight into what factors may affect Black individuals’ willingness to participate in medical trials. We hope that this study will spur much-needed additional work in this area, including studies that can examine actual participation decisions.

What key takeaways did you gain from the study, and how will they inform your future work?

Two key takeaways come to mind from the study. First is that institutional and historical contexts matter when trying to address any social problem. Prior work investigating the ‘why’ behind racial under-representation in clinical trials provided crucial context for our study. 

Second, our study underscores the importance of efforts that increase opportunities for those historically excluded in STEM fields, especially as these programs are being questioned and defunded. Programming that serves to create equity and improve diversity by providing tuition support, mentorship, and coursework opportunities is critical to expose, recruit, and retain underrepresented students in STEM fields. This study, among others, demonstrates that such efforts might have spillover effects to research, since greater representation among investigators increased Black individuals’ participation in research. We hope our findings help quantify the importance of such efforts across disciplines and research institutes.