Quantifying the Extent of Early COVID-19 Illness

Abstract

To date, there remains much confusion about when the first cases of COVID-19 occurred in the U.S. This knowledge informs models of COVID-19 transmission and impacts its estimated fatality rate. We employed a differencing analysis on the Current Population Survey (CPS) to examine how rates of absence from work due to illness varied this year relative to last year in key industries. We found that absence from work due to illness was higher in almost all industries during December-March of 2020 relative to the same months of 2019. Our findings imply that COVID-19 was spreading in the U.S. as early as December. They also highlight the fact that labor market surveys, like the CPS, can be a valuable tool for the early detection of illness at a national level. Information from labor market surveys should be used to track and combat COVID-19 and future disease outbreaks.