J Med Virol. 2026 Jul;98(7):e71040. doi: 10.1002/jmv.71040.
ABSTRACT
This study estimates the incidence of symptomatic COVID-19 cases, both documented and undocumented, among U.S. Veterans across demographic groups from the beginning of the pandemic to the end of the public health emergency on May 11, 2023. By analyzing a cohort of Veterans alive as of March 1, 2020, we extended a mortality-based estimation approach to measure COVID-19 incidence. We relaxed the assumptions of a constant infection fatality rate (IFR) over time and across age groups and broadened the model from considering only excess respiratory deaths to including excess all-cause deaths. Descriptive analyses were performed to understand differential ascertainment biases among demographic groups. Resulting estimates suggested a significantly higher number of COVID-19 cases among Veterans than those documented in the electronic health record. We also identified varying biases among different demographic groups. These estimates offer a clearer view of COVID-19’s impact on Veterans, accounting for missed cases among those who sought care outside of the VA. Differences between documented and estimated cases were substantial. Policymakers should recognize that actual numbers are likely much higher than documented and that documented rates may not be directly comparable across populations or time periods.
PMID:42366842 | DOI:10.1002/jmv.71040
