Neighborhood data allow a hyper-local look at COVID-19 burden across the US, challenging assumptions

In this paper, Noppert et al. presented their first findings from the COVID Neighborhood Project (CONEP), an effort to compile ZIP-code-level and census-tract-level COVID-19 case data, to better understand COVID-19 burden variation across neighborhoods and states, including both urban and rural areas. They examined how neighborhood-level COVID-19 burden varied across 21 US states and what local factors were associated with higher case rates. They did this by linking CONEP’s case data to spatially referenced, nationwide measures of physical and social environments, which are contained in studies housed in the National Neighborhood Data Archive (NaNDA). Specifically, they used measures from two NaNDA studies that focused on neighborhood disadvantage, affluence, rurality, population density, and county-level political partisanship. Noppert et al. found that the distribution of COVID-19 cases at the neighborhood level differed substantially both within and between states. Additionally, the specific neighborhood factors linked to higher COVID-19 rates diverged in size and direction across the different states analyzed. These findings challenge traditional assumptions about disadvantaged areas universally experiencing higher infectious disease burden. And they underscore the importance of examining hyperlocal patterns and neighborhood data to understand the social determinants driving unequal COVID-19 impact. Two of the paper’s authors, Noppert and Clarke, are members of the SBE COVID Consortium of scientists funded by the National Institutes of Health to research COVID-19’s impact on minorities and underserved populations. As members, they work closely with the Social, Behavioral, & Economic COVID Coordinating Center at ICPSR, which promotes communication and collaboration among the behavioral and social science community, multiple NIH award recipients, and the public.