County-Level Estimates of Excess Mortality associated with COVID-19 in the United States

Abstract

The coronavirus disease 2019 (COVID-19) pandemic in the US has been largely monitored on the basis of death certificates containing reference to COVID-19. However, prior analyses reveal that a significant fraction of excess deaths associated with the pandemic were not directly assigned to COVID-19 on the death certificate. The percent of excess deaths not assigned to COVID-19 is also known to vary across US states. However, few studies to date provide information on patterns of excess mortality and excess deaths not assigned to COVID-19 for US counties, despite the importance of this information for health policy and planning. In the present study, we develop and validate a generalized linear model of expected mortality in 2020 based on historical trends in deaths by county of residence between 2011 and 2019. We use the results of the model to generate county estimates of excess mortality and excess deaths not assigned to COVID-19 for each county in the US along with bootstrapped prediction intervals. Overall, in counties that had statistically significant increases in excess mortality, the proportion of excess deaths assigned to COVID-19 was 82%, meaning that 18% of excess deaths were not assigned to COVID-19. The proportion assigned to COVID-19 was lower in the South (77%) and West (78%) as compared to counties in the Midwest (82%) and Northeast (94%). Across US Census Divisions, the proportion was especially low in the East South Central Division (68%). Rural counties across all divisions (68%) reported lower proportions of excess deaths assigned to COVID-19 than urban areas (84%). For instance, in the Pacific Division, less than half of excess deaths were assigned to COVID-19 in nonmetro areas. In contrast, the New England Census Division stood out as the only division where directly assigned COVID-19 deaths actually exceeded excess deaths, meaning there were 1.22 directly assigned COVID-19 deaths for every 1 excess death. However, this finding did not extend to nonmetro areas within New England where less than three-quarters of excess deaths were assigned to COVID-19. The finding that metro areas in New England reported higher direct COVID-19 mortality than excess mortality suggests that reductions in mortality from other causes of death may have occurred in these areas, at least among some populations. Across individual counties, the percentage of excess deaths not assigned to COVID-19 varied substantially, with some counties’ direct COVID-19 tallies capturing only a small fraction of total excess deaths, whereas in other counties the direct COVID-19 death rate far exceeded the number of estimated excess deaths. Taken together, our results suggest that regional inequalities in the mortality burden associated with COVID-19 are not fully revealed by data at the state level and that consideration of excess deaths across US counties is critical for a full accounting of the disparate regional effects of the pandemic on US mortality.