Spatial Modeling of Sociodemographic Risk for COVID-19 Mortality.

In early 2020, the Coronavirus Disease 19 (COVID-19) rapidly spread across the United States (US), exhibiting significant geographic variability. While several studies have examined the predictive relationships of differing factors on COVID-19 deaths, few have looked at spatiotemporal variation at refined geographic scales. The objective of this analysis is to examine this spatiotemporal variation in COVID-19 deaths with respect to association with socioeconomic, health, demographic, and political factors. We use multivariate regression applied to Health and Human Services (HHS) regions as well as nationwide county-level geographically weighted random forest (GWRF) models. Analyses were performed on data from three separate time frames which correspond to the spread of distinct viral variants in the US: pandemic onset until May 2021, May 2021 through November 2021, and December 2021 until April 2022. Multivariate regression results for all regions across three time windows suggest that existing measures of social vulnerability for disaster preparedness (SVI) are predictive of a higher degree of mortality from COVID-19. In comparison, GWRF models provide a more robust evaluation of feature importance and prediction, exposing the value of local features for prediction, such as obesity, which is obscured by coarse-grained analysis. Overall, GWRF results indicate that this more nuanced modeling strategy is useful for determining the spatial variation in the importance of sociodemographic risk factors for predicting COVID-19 mortality.