Long-term Exposure to Fine Particulate Matter and Mortality A Longitudinal Cohort Study of 400,459 Adults

Background: Cohort studies on the association between long-term exposure to fine particulate matter (PM2.5) and mortality have been well established for America and Europe, but limited and inconsistent in Asia with much higher air pollution. This study aims to investigate the associations between ambient PM2.5 and all-cause and cause-specific mortality over a period of rising and then declining PM2.5. Methods: We enrolled a total of 400,459 adults from an open cohort between 2001 and 2016, and followed them up until 31 May 2019. We obtained mortality data from the National Death Registry maintained by the Ministry of Health and Welfare in Taiwan. We estimated ambient PM2.5 exposures using a satellite-based spatiotemporal model. We performed a Cox regression model with time-dependent covariates to investigate the associations of PM2.5 with deaths from all causes and specific causes. Results: This study identified 14,627 deaths and had a total of 5 million person-years of follow-up. Each 10 mu g/m(3) increase in PM2.5 was associated with an increased hazard risk of 29% (95% confidence interval: 24%-35%) in all-cause mortality. Risk of death increased by 30% for natural causes, 20% for cancer, 42% for cardiovascular disease (CVD) causes, and 53% for influenza and pneumonia causes, for each 10 mu g/m(3) increase in PM2.5. Sensitivity analyses generally yielded similar results. Conclusion: Long-term exposure to ambient PM2.5 was associated with increased risks of all-cause mortality and deaths from cancers, natural causes, CVD, and influenza and pneumonia. Longitudinal study design should be encouraged for air pollution epidemiologic investigation.