Summary:
I was a collaborator on this research project led by Dr. Mathew Kiang and Dr. Andrew Stokes, published in Science Advances in March 2026 that uses machine learning methods to identify unrecognized COVID-19 deaths that occurred in 2020 and 2021. We find that these these deaths occurred disproportionately among decedents with less than a high school education; decedents identified as Hispanic, American Indian, Alaska Native, Asian, and/or Black; counties with lower household incomes and worse preexisting health; and counties in the South. In the paper, we argue that our results showing more unrecognized COVID-19 deaths among counties with worse preexisting health are suggestive that unrecognized deaths occurred disproportionately among chronically ill and disabled people. This finding is indicative of a pattern of structural ableism (along with structural racism and classism) in the death investigation system’s performance during the COVID-19 pandemic that warrants further research and policy attention.
Citation:
Kiang MV, Li ZR, Wrigley-Field E, et al. Applying machine learning to identify unrecognized COVID-19 deaths recorded as other causes of death in the United States. Sci Adv. 2026;12(12):eaef5697. doi:10.1126/sciadv.aef5697
Media Coverage:
COVID probably killed 150,000 more people in its first two years than official U.S. tolls show in Scientific American
