COVID-19-related state-wise racial and ethnic disparities across the USA: an observational study based on publicly available data from The COVID Tracking Project

Abstract

Objective: To evaluate COVID-19 infection and mortality disparities in ethnic and racial subgroups in a state-wise manner across the USA.

Methods: Publicly available data from The COVID Tracking Project at The Atlantic were accessed between 9 September 2020 and 14 September 2020. For each state and the District of Columbia, % infection, % death, and % population proportion for subgroups of race (African American/black (AA/black), Asian, American Indian or Alaska Native (AI/AN), and white) and ethnicity (Hispanic/Latino, non-Hispanic) were recorded. Crude and normalised disparity estimates were generated for COVID-19 infection (CDI and NDI) and mortality (CDM and NDM), computed as absolute and relative difference between % infection or % mortality and % population proportion per state. Choropleth map display was created as thematic representation proportionate to CDI, NDI, CDM and NDM.

Results: The Hispanic population had a median of 158% higher COVID-19 infection relative to their % population proportion (median 158%, IQR 100%-200%). This was followed by AA, with 50% higher COVID-19 infection relative to their % population proportion (median 50%, IQR 25%-100%). The AA population had the most disproportionate mortality, with a median of 46% higher mortality than the % population proportion (median 46%, IQR 18%-66%). Disproportionate impact of COVID-19 was also seen in AI/AN and Asian populations, with 100% excess infections than the % population proportion seen in nine states for AI/AN and seven states for Asian populations. There was no disproportionate impact in the white population in any state.

Conclusions: There are racial/ethnic disparities in COVID-19 infection/mortality, with distinct state-wise patterns across the USA based on racial/ethnic composition. There were missing and inconsistently reported racial/ethnic data in many states. This underscores the need for standardised reporting, attention to specific regional patterns, adequate resource allocation and addressing the underlying social determinants of health adversely affecting chronically marginalised groups.

Keywords: COVID-19; epidemiology; health policy.

Figure 1

Choropleth map showing crude disparity estimates for COVID-19 infections, computed as absolute percentage point difference between % COVID-19 infection and % population proportion by ethnic/racial groups across US states and the District of Columbia. AA, African American; AI, American Indian; AN, Alaska Native.

Figure 2

Choropleth map showing normalised disparity estimates for COVID-19 infections computed as CDI normalised by % population proportion of each ethnic/racial group across US states and the District of Columbia. AA, African American; AI, American Indian; AN, Alaska Native; CDI, crude disparity estimates for COVID-19 infection.

Figure 3

Choropleth map showing crude disparity estimates for COVID-19 mortality, computed as absolute percentage point difference between % COVID-19-related mortality and % population proportion by ethnic/racial groups across US states and the District of Columbia. AA, African American; AI, American Indian; AN, Alaska Native.

Figure 4

Choropleth map showing normalised disparity estimates for COVID-19 mortality, computed as CDM normalised by % population proportion of each ethnic/racial group across US states and the District of Columbia. AA, African American; AI, American Indian; AN, Alaska Native; CDM, crude disparity estimate for COVID-19 mortality.