Social Class, Race/Ethnicity, and COVID-19 Mortality Among Working Age Adults in the United States

Importance: Substantial racial/ethnic disparities in COVID-19 mortality have been documented. Social class is a likely explanation of mortality disparities across and within racial/ethnic groups. This is the first U.S. study of social class and COVID-19 mortality in working age adults. Objectives: To determine the joint effects of social class, race/ethnicity, and gender on the burden of COVID-19 mortality. A secondary objective was to determine whether differences in opportunities for remote work were correlated with COVID-19 death rates for sociodemographic groups. Design: Annual mortality study which used a special government tabulation of 2020 COVID-19 related deaths stratified by decedent social class (educational attainment) and race/ethnicity. Setting: United States in 2020. Participants: COVID-19 decedents aged 25 to 64 years old (n=69,001). Exposures: Social class (working class, some college, college graduate), race/ethnicity (Hispanic, Black, Asian, Indigenous, multiracial, and non-Hispanic white), and gender (women, men). Detailed census data on occupations held by adults in 2020 in each of the 36 sociodemographic groups studied were used to quantify the possibility of remote work for each group. Main Outcomes and Measures: Age-adjusted COVID-19 death rates for 36 sociodemographic groups defined by social class, race/ethnicity, and gender. Disparities were quantified by relative risks and 95% confidence intervals. College graduates were the (low risk) referent group for all relative risk calculations. Results: A higher proportion of Hispanics, Blacks, and Indigenous people were working class in 2020. COVID-19 mortality was five times higher in the working class vs. college graduates (72.2 vs. 14.6 deaths per 100,000, RR=4.94, 95% CI 4.82-5.05). The joint detriments of lower socioeconomic position, Hispanic ethnicity, and male gender resulted in a COVID-19 death rate which was over 27 times higher (178.0 vs. 6.5 deaths/100,000, RR=27.4, 95%CI 25.9-28.9) for working class Hispanic men vs. college graduate white women. In regression modeling, percent employed in never remote jobs explained 72% of the variance in COVID-19 death rates. Conclusions and Relevance: SARS-CoV-2 infection control efforts should prioritize the working class (i.e. those with no college education), particularly those employed in never remote jobs with inflexible and unsafe working conditions (i.e. blue collar, service, and retail sales workers).