Floods and cause-specific mortality in the United States applying a triply robust approach

Abstract

The health impact of floods has not been well characterized. This study evaluated long-term associations between cause-specific mortality rates and county-level monthly flood days (excluding coastal floods caused by tropical storms) in the post-flood year in the contiguous U.S., using a triply robust approach incorporating propensity score, counterfactual estimation, and confounder adjustment. Death records came from the CDC National Center for Health Statistics (2001-2020) and floods came from the NOAA Storm Events Database (2000-2020). We found that one flood day was associated with 8.3 (95% CI: 2.5 to 14.1) excess all-cause deaths per 10 million individuals, 3.1 due to myocardial infarction, 2.4 due to respiratory diseases, and 5.9 due to external causes. From 2001 to 2020, 22,376 (95% CI: 6,758 to 37,993) all-cause deaths were attributable to floods. Our findings highlight the long-term health risks after floods, and a need for measures to reduce these risks.