Racial, Socioeconomic, and Structural Determinants of Age-adjusted Excess COVID-19 Mortality Across U.S. Counties

To examine racial, socioeconomic, and structural disparities in age-adjusted excess COVID-19 mortality across U.S. counties and identify contextual factors associated with disproportionate mortality.

County-level data from all 3,142 U.S. counties and the District of Columbia were assembled from ten national datasets. Age-adjusted excess COVID-19 mortality was calculated using national age-specific COVID-19 mortality rates. Data-driven variable selection was performed using Least Absolute Shrinkage and Selection Operator (LASSO) regression, followed by population-weighted linear regression with state-clustered standard errors.

Excess mortality varied substantially across counties (mean 29.2 per 100,000; range - 593.4 to 879). Counties with larger Black populations, greater racial segregation, higher socioeconomic disadvantage, lower educational attainment, and higher uninsured prevalence experienced significantly greater excess mortality. Percentage of female-headed family households with no husband present (β = 0.38, 95% CI: 0.34-0.41) and Republican vote share (β = 0.22, 95% CI: 0.19-0.24) were positively associated with mortality, whereas remote work prevalence was inversely associated.

Racial and structural inequities remain strongly associated with unequal mortality across U.S. communities. Addressing persistent inequities in education, healthcare access, neighborhood disadvantage, and economic opportunity may improve resilience to future pandemics and other population health threats.