Long-term exposure to particulate matter and COPD mortality: Insights from causal inference methods based on a large population cohort in southern China

Background: Evidence of the association between long-term exposure to particulate matter (PM) and chronic obstruc-tive pulmonary disease (COPD) mortality from large population-based cohort study is limited and often suffers from residual confounding issues with traditional statistical methods. We hereby assessed the casual relationship between long-term PM (PM2.5, PM10 and PM10-2.5) exposure and COPD mortality in a large cohort of Chinese adults using state-of-the-art causal inference approaches.Methods: A total of 580,757 participants in southern China were enrolled in a prospective cohort study from 2009 to 2015 and followed up until December 2020. Exposures to PM at each residential address were obtained from the Long-term Gap-free High-resolution Air Pollutant Concentration dataset. Marginal structural Cox models were used to investigate the association between COPD mortality and annual average exposure levels of PM exposure.Results: During an average follow-up of 8.0 years, 2250 COPD-related deaths occurred. Under a set of causal inference assumptions, the hazard ratio (HR) for COPD mortality was estimated to be 1.046 (95 % confidence interval: 1.034-1057), 1.037 (1.028-1.047), and 1.032 (1.006-1.058) for each 1-mu g/m3 increase in annual average concentra-tions of PM2.5, PM10, and PM10-2.5 respectively. Additionally, the detrimental effects appeared to be more pronounced among the elderly (age >= 65) and inactive participants. The effect estimates of PM2.5, PM10, and PM10-2.5 tend to be greater among participants who were generally exposed to PM10 concentrations below 70 mu g/m3 than that among the general population.Conclusion: Our results support causal links between long-term PM exposure and COPD mortality, highlighting the urgency for more effective strategies to reduce PM exposure, with particular attention on protecting potentially vulner-able groups.