Inequities in the Geographic Accessibility of COVID-19 Biomedical Therapeutic Trials in the United States

Coronavirus disease 2019 (COVID-19) has disproportionately impacted marginalized communities across the United States (US). However, racial/ethnic minority and elderly populations experiencing the highest COVID-19 incidence, hospitalization, and mortality rates have not been equitably enrolled in clinical tr ials invest igat ing potential COVID-19 therapeutics. We descriptively evaluated the geographic proximity of demographic subpopulations to COVID-19 biomedical therapeutic trial sites. We hypothesized that trial sites would be more accessible to urban populations and subgroups who more often live in urban areas (racial/ethnic minority and younger populations).


INTRODUCTION
Coronavirus disease 2019 (COVID-19) has disproportionately impacted marginalized communities across the United States (US). 1 However, racial/ethnic minority and elderly populations experiencing the highest COVID-19 incidence, hospitalization, and mortality rates have not been equitably enrolled in clinical trials investigating potential COVID-19 therapeutics. 2,3 We descriptively evaluated the geographic proximity of demographic subpopulations to COVID-19 biomedical therapeutic trial sites. We hypothesized that trial sites would be more accessible to urban populations and subgroups who more often live in urban areas (racial/ethnic minority and younger populations).
We geocoded trial site addresses using Google Places API. We calculated drive times from the center of population for each census tract to the ten geographically closest sites and selected the site with the shortest time. We stratified rural and urban tracts using 2010 USDA ERS Rural-Urban Commuting Area codes. We calculated the proportion of each demographic subgroup residing within x minutes of the nearest trial site by weighting each tract by population demographics (age, race, ethnicity) from the 2015-2019 US Census American Community Survey (ACS). We calculated median drive times with 95% confidence intervals by bootstrap.
We performed statistical analyses using RStudio v1.3.1073 (R Foundation for Statistical Computing), and plotted maps using ArcMap v10.7.1. The University of Virginia Institutional Review Board deemed this study exempt.
Trial sites were clustered near metropolitan centers ( Figure 1A), with corresponding shorter drive times near urban areas ( Figure 1B)

DISCUSSION
Similar to the geographic inaccessibility of clinical trials for other diseases, 4 the opportunity to enroll in biomedical therapeutic trials throughout the first 8 months of the COVID-19 pandemic was not equitably available across the US. Nearly one-third of the overall US population, over one-half of AIAN people, and over threefourth of the rural population lived more than an hour from the nearest trial site. Rural-urban differences in trial distribution explain longer overall drive times for White and elderly populations, since these groups disproportionately resided in rural census tracts. However, the AIAN population faced longer drive times even when accounting for rurality, suggesting they are uniquely geographically isolated from novel therapeutics.
Non-Hispanic and White individuals were wellrepresented in COVID-19 trials despite rural trial inaccessibility and lower hospitalization rates. 2 Furthermore, the underrepresentation of Black and Hispanic populations in COVID-19 therapeutic trials is especially striking given their relative geographic proximity to trial sites and disproportionate hospitalization rates, both of which suggest greater opportunity for recruitment. 2 Factors unexplored herein-including racism, mistrust, language barriers, and the persistent segregation of wellresourced hospitals-should be investigated further as potential mediators of decreased trial enrollment. 5 Our study has limitations, including that our use of tract centers of population assumes demographic groups are not clustered within tracts. We also did not account for vehicle access or reliance on public transportation. Thus, our tract-level analyses may misestimate travel times for vehicle-less and demographically segregated urban populations. Beyond the COVID-19 era, innovations like decentralized, Internet-based clinical trials may help mitigate geographic inequities. 6 However, it remains clear that geographic accessibility alone may not improve racial/ethnic representation in the absence of additional structural interventions. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons. org/licenses/by/4.0/. Figure 2 One-way drive times to the nearest COVID-19 biomedical therapeutic trial site for demographic subgroups, stratified by rurality. Bar graphs display the percentage of the population with less than (i.e., left of 0% on the x-axis) or greater than (i.e., right of 0% on the x-axis) a 60min drive time to the nearest COVID-19 biomedical therapeutic trial site. For each sociodemographic subgroup, the median and 95% confidence interval are displayed to the right of the bar.