Changes in PM2.5 exposure due to residential relocation and mortality among United States Veterans

Air pollution remains a leading cause of mortality. Few studies have leveraged residential relocation to examine impacts of intraindividual changes in PM2.5 exposure. As the largest integrated health system in the United States, the Veterans Health Administration (VA) is an ideal setting to examine relocation-induced changes in air pollution. We constructed a cohort of veterans who relocated from one ZIP code to another between 2011 and 2017 (N = 762 905). We linked spatially averaged annual PM2.5 estimates to each person based on residential address. We examined relocation-induced PM2.5 changes and post-move mortality in discrete-time models, accounting for individual-level and area-level confounding variables. We estimated pooled odds ratios (ORs) over 5 years of follow-up and examined effect-measure modification by age (<65, 65+), sex, and neighborhood-level poverty. The pooled OR associated with a 2 μg/m3 increase in PM2.5 levels between the origin and destination areas was 1.03 (95% CI = 1.01-1.03). Associations appeared more pronounced among older individuals when assessing potential additive effect-measure modification. Examining intraindividual changes in PM2.5 exposure and subsequent health outcomes can elucidate potential impacts of air pollution and pollution-reduction policies. Results suggest survival benefits from policies that further reduce PM2.5 levels.