Identifying Areas of Low-Access to the COVID-19 Vaccine: A New Objective Framework Incorporating Mobility Data

Abstract

Abstract Background Methods have been proposed to identify areas of low access to resources that are embedded with subjective parameters. In this work, we propose and validate a new Mobility Data-Driven (MDD) framework to identify areas that have low access to the COVID-19 vaccine. Methods We collected geospatial mobility data to an objective approach for determining areas of low access. We identify census tracts in Oregon with low access to the COVID-19 vaccine through two approaches: (1) An adapted United States Department of Agriculture (USDA) food desert definition and (2) our proposed MDD framework. Ten spatial and social measures of access are utilized to compare these two approaches. Results Tracts identified by the MDD definition have lower spatial accessibility scores (0.072 – 0.162) than those identified by the USDA adapted definition (0.239 – 0.32). During the Spring season, the MDD identified census tracts have a higher rate of poverty (15.2%), unemployment (8.4%), uninsured individuals (7.1%), and a lower per capita income ($28,261). Moreover, we find that the proportion the American Indian and Alaskan Native population in MDD identified low-access census tracts is higher than that in USDA definition (4.85% versus 0.95%) which indicates the framework’s ability to capture known disparities in access amongst this population. Conclusions During the creation of plans for equitable distribution of a resource like the COVID-19 vaccine, leaders should utilize objective data (like mobility data) to assist them in determining parameters that represent a populations ability to obtain that resource. Our proposed framework provides a starting point for achieving this goal.