Racial Disproportionality in Covid Clinical Trials

There are sufficient data demonstrating that coexisting conditions in patients with Covid-19 influence clinical outcomes and that older age and male sex are associated with a greater risk of death. But despite disproportionately higher rates of Covid-19 infection, hospitalization, and death in racial and ethnic minority groups, the direct effects of genetic or biologic host factors remain unknown.¹

As we strive to overcome the social and structural causes of health care disparities, we must recognize the underrepresentation of minority groups in Covid-19 clinical trials. Although the Food and Drug Administration hailed remdesivir as the standard of care for Covid-19 and is actively distributing supplies throughout the United States, data supporting the drug’s efficacy and safety in minority groups are limited. Data on race and ethnicity were not provided for the 53 patients treated with remdesivir under the “compassionate use” program.² Though we acknowledge that these early results were obtained from a limited data set, Black Americans accounted for only about 20% of the 1063 patients in the placebo-controlled Adaptive Covid-19 Treatment Trial (ACTT-1) funded by the National Institute of Allergy and Infectious Diseases (NIAID)³ and 11% of the 397 patients randomly assigned to 5 or 10 days of remdesivir in the Gilead-funded study (GS-U.S.-540-5773).⁴ The proportions of Latinx and Native American patients were provided only for ACTT-1 and were 23% and 0.7%, respectively.³

Covid-19 Cases and Deaths, According to Black or White Race and Latinx Ethnicity.

Data were obtained from the COVID Tracking Project at The Atlantic (https://covidtracking.com/race) for U.S.-based study sites in the Adaptive Covid-19 Treatment Trial (ACTT-1) and the GS-U.S.-540-5773 trial. Race and ethnicity data are not reported consistently by all states or in their entirety, since some states do not provide race and ethnicity data for Covid-19 cases (NY and LA), some use nonstandard categories, and others fail to provide data altogether (NE, not included here). Data were current as of June 10, 2020.

Both trials included sites throughout the United States where Black, Latinx, and Native Americans are overrepresented among people with Covid-19 and related deaths, but these groups were substantially underrepresented in the study samples (see graph). Indeed, despite widespread underreporting of patients’ race or ethnicity, we know that Black, Latinx, and Native Americans are dying from Covid-19 at rates disproportionate to their representation in the population in multiple U.S. regions. The modest benefit seen in time to clinical improvement with remdesivir may not be generalizable to minority populations, given the differences in disease severity and outcomes.

Lack of diversity in these clinical trials may stem from long-standing medical distrust on the part of minority communities, but the problem may be compounded by cost (in particular hidden costs for such requirements as fuel, parking, meals, and lodging), poor health literacy, lack of information, language barriers, limited accessibility, and implicit biases against minorities.⁵ Another possible explanation is the lack of diversity among principal investigators who can obtain funding, since such diversity might curtail bias in recruitment of participants from underrepresented populations. Despite these barriers, federal law (Public Health Service Act sec. 492B, 42 U.S.C. sec. 289a-2) and National Institutes of Health (NIH) policy (NOT-OD-18-014) mandate inclusion of minorities in NIH-funded research. Furthermore, NIH-defined phase 3 clinical trials are required to report outcomes stratified according to sex or gender and according to race and ethnicity unless data are presented to suggest that differences are unlikely to be observed. These requirements, as well as the principle of justice, demand equitable selection and enrollment of participants and detailed presentation of demographic data and outcomes.

Yet the remdesivir studies failed to provide equal representation of Black, Latinx, and Native Americans, and the ACTT-1 investigators have yet to provide outcome data according to sex or gender, race, and ethnicity. Multivariate analysis of the Gilead study suggested that both Black race and White race were associated with shorter time to clinical improvement,⁴ but these results are unreliable because of the low number of Black participants. It is alarming that long-standing racial health disparities have been extended to Covid-19 clinical trials when racial and ethnic minority groups have so much to gain from this research, including the opportunity to receive lifesaving treatment.

To provide the necessary data for generalizing efficacy and safety outcomes across racial groups, Covid-19 clinical trials must prioritize inclusion of patient populations that reflect the demographics of the ongoing pandemic, especially in the United States, and funding for scientists from underrepresented ethnic groups must be increased. Appropriate random sampling and the expansion of clinical trial sites to vulnerable communities with high-risk patients would improve the representativeness of samples. Furthermore, standardized reporting of race and ethnicity categories would allow for improved assessment of the generalizability of evidence-based interventions.

In addition, we believe that sponsors, regulatory agencies, research funders, medical journals, and peer reviewers have a responsibility to ensure that clinical trials satisfy the requirements of the aforementioned federal law and NIH policy and present all demographic data. These groups can deem it unacceptable for proposals or manuscripts to fail to justify the demographic makeup of a study population. Researchers will have to consider strategies for recruiting participants from underserved groups by proactively addressing the possibility or fear of exploitation. To improve the response to Covid-19 and allocate resources appropriately, states and municipalities will need to report race and ethnicity data on cases and deaths promptly, while Covid-19 clinical trials incorporate marginalized communities by targeting populations at greatest risk — most notably, Black, Latinx, and Native American.