Association of physical activity and PM2.5-attributable cardiovascular disease mortality in the United States

OBJECTIVE: The study aimed to explore the association between physical activity (PA) and PM2. 5-attributable cardiovascular disease (CVD) mortality trends across the United States (US) at the state level. METHODS: We conducted a cross-sectional study using data from the Global Burden of Disease 2019 study for PM2.5-attributable CVD mortality and the Behavioral Risk Factor Surveillance System for PA prevalence. The study covered all 50 US states and the District of Columbia from 2001 to 2019. We utilized Joinpoint Regression to calculate AAPC from 2011 to 2019 and Pearson correlation coefficients to assess state-level associations between PA and PM2.5-attributable CVD mortality AAPC. RESULTS: During 2011–2019, a total of 244,318 PM2.5-attributable CVD deaths were recorded. The age-adjusted mortality rates (AAMR) of PM2.5-attributable CVD declined substantially from 2011 to 2019 across all US states, with the most pronounced reductions observed in industrialized states such as West Virginia (51% decline), Kentucky (32%), and Ohio (22%). AAMR ratios for the US states varied substantially, ranging from 0.1 in Hawaii to 1.7 in Arkansas. The AAPC ranged from −9.4% (West Virginia) to −1.7% (New Mexico) in the majority of states, while a few states such as Alaska, Wyoming, and Washington saw slight positive AAPCs from 0.9 to 2.9%. A significant correlation was found between PA and PM2.5-attributable CVD mortality trends (r = 0.454, p = 0.001), with similar results in subgroup analyses. CONCLUSION: Our findings suggest a correlation between increased physical activity (PA) and increased PM2.5-attributable CVD mortality, highlighting the potential need to consider PM2.5 exposure when engaging in PA to mitigate adverse cardiovascular health impacts. However, further research is warranted to establish causality and underlying mechanisms in the relationship between PA and PM2.5-attributable CVD mortality. Potential limitations include reliance on self-reported PA data.