Jump to Main Content
Associations of daily mortality with short-term exposure to PM2.5 and its constituents in Shanghai, China
- Wang, Yiyi, Shi, Zhihao, Shen, Fuzhen, Sun, Jinjin, Huang, Lin, Zhang, Hongliang, Chen, Chen, Li, Tiantian, Hu, Jianlin
- Chemosphere 2019 v.233 pp. 879-887
- acute exposure, ammonium, epidemiological studies, human health, linear models, mortality, nitrates, organic carbon, particulates, pollutants, respiratory tract diseases, sulfates, China
- Epidemiological studies have shown that fine particulate matter (PM2.5) has adverse impacts on human health. However, limited studies have investigated the effects of short-term exposure to PM2.5 and its constituents on mortality in China. This study used the generalized linear model (GLM) to investigate the effects of PM2.5 and its constituents, including organic carbon (OC), element carbon (EC), ammonium (NH4+), nitrate (NO3−), sulfate (SO42−), on different causes of mortality in Shanghai from January 1, 2013 to December 31, 2015. The single-day lagged model and the moving average lagged model were used to examine the lagging effects of pollutants on mortality. At lag0 day, PM2.5 had a significant effect on all-cause mortality, and a 10 μg/m3 increase leads to 0.68% increase in all-cause mortality (RR 1.0068, 95%CI 1.0013–1.0123). Among the five constituents, EC had the greatest impact on all-cause mortality in Shanghai, with 10.48% increase of mortality (RR 1.1048, 95%CI 1.0266–1.1891) per 10 μg/m3 increase of concentrations, followed by OC (RR 1.0577, 95%CI 1.0277–1.0886), NH4+ (RR 1.0272, 95%CI 1.0028–1.0522) and SO42− (RR 1.0104, 95%CI 1.0003–1.0206). For respiratory diseases mortality, EC, OC, NO3− and NH4+ had significant impacts and caused an increase of mortality by 44.99% (RR 1.4499, 95%CI 1.1813–1.7794), 10.40% (RR 1.1040, 95%CI 1.0260–1.1880), 5.338% (RR 1.0533, 95%CI 1.0097–1.0989) and 7.34% (RR 1.0734, 95%CI 1.0015–1.1505) per 10 μg/m3 increase of concentrations, respectively. The cumulative effect of PM2.5 on mortality was significant in Shanghai. Except for SO42−, the RR value of the single-day lagged model was smaller than the moving average lagged model.