Mortality burden attributable to long-term ambient PM2.5 exposure in China: using novel exposure-response functions with multiple exposure windows

Background: Exposure to ambient fine particulate matter (PM2.5) increases the mortality burden. Exposure windows and exposure-response functions (ERFs) are two critical components of accurate mortality burden estimation. Objective: We explored the potential heterogeneity of exposure windows and reassessed the PM2.5-attributable mortality burden in China with novel ERFs. Method: Based on 1 km x 1 km satellite-retrieved PM2.5 and population data, provincial-level age structure, and mortality data, we applied the recent Global Exposure Mortality Model (GEMM) with multiple exposure windows (1-year to 6-year during 2010-2015) to estimate age-specific PM2.5-attributable mortality burden in China in 2015. Then, the Global Burden of Disease (GBD) 2017 Integrated Exposure-Response (IER) and Log-Linear (LL) models were exercised for comparative analysis. Result: The PM2.5-attributable mortality was the highest with a 3-year average exposure window (2013-2015). The GEMM-based total premature deaths were 133.2% [95% confidence interval (95% CI): 93.6%-226.4%] and 14.2% (95% CI: 13.9%-16.8%) higher than the values obtained from the GBD2017 IER model and LL model, respectively. The national mortality burden attributable to PM2.5 was 1.94 (95% CI: 1.63-2.23) million, of which IHD and stroke were the leading causes, accounting for 27.3% and 23.0% of the total burden respectively. The mortality burden for the people over 80 years old was 0.62 (95%CI: 0.52-0.71) million, accounting for 31.9% (95%CI: 31.8%-32.0%) of the total burden. Conclusions: This study demonstrates the potential heterogeneity of PM2.5-attributable mortality burden associated with different exposure windows, especially when there are spatial-temporal variations in PM2.5 concentrations. The model comparison results suggest that the health impacts attributed to long-term PM2.5 exposure in China may be much higher than previously estimated. The population over 80 years old has the highest PM2.5-attributable mortality burden. These findings have important policy implications for addressing air pollution at the provincial and national level in China.