Interactions between cigarette smoking and ambient PM2.5 for cardiovascular mortality

Associations between long-term exposure to ambient fine particulate matter (PM2.5) and all-cause and cardiovascular mortality are well documented however less is known regarding possible interactions with cigarette smoking. We previously reported a supra-additive synergistic relationship between PM2.5 and cigarette smoking for lung cancer mortality. Here we examine interactions for all-cause and cardiovascular mortality among 429,406 current or never smoking participants in the prospective American Cancer Society Cancer Prevention Study-II with modeled PM2.5 concentrations. Cox proportional and additive hazards models were used to estimate mortality associations and interactions on the multiplicative and additive scales. A total of 146,495 all-cause and 64,339 cardiovascular (plus diabetes) deaths were observed. The hazard ratio (HR) (95% confidence interval (CI)) for cardiovascular mortality for high vs. low PM2.5 exposure (>14.44 µg/m3 vs ≤10.59 µg/m3, 75th vs 25th percentile) was 1.09 (95% CI 1.05, 1.12) in never smokers. The HR for cigarette smoking was 1.89 (95% CI 1.82, 1.96) in those with low PM2.5. The HR for both high PM2.5 and cigarette smoking was 2.08 (95% CI 2.00, 2.17). A small significant excess relative risk due to interaction (0.10; 95% CI 0.02, 0.19) was observed. Quantification of the public health burden attributed to the interaction between PM2.5 and cigarette smoking indicated a total of 32 (95% CI −6, 71) additional cardiovascular deaths per 100,000 person-years due to this interaction. In conclusion, PM2.5 was associated with all-cause and cardiovascular mortality in both smokers and never smokers, with some evidence for a small additive interaction with cigarette smoking. Reductions in cigarette smoking will result in the greatest impact on reducing all-cause and cardiovascular death at the levels of PM2.5 observed in this study. However, reductions in PM2.5 will also contribute to preventing a proportion of mortality attributed to cigarette smoking.