Time-Series Study On The Air Pollution And Daily Mortality In Beijing
Xiao-chuan Pan,Guiqiu Chang
DOI: https://doi.org/10.1097/00001648-200407000-00142
2004-01-01
Epidemiology
Abstract:ISEE-119 Abstract: In this study, daily time-series study was used to evaluate quantitatively associations between ambient air pollution and daily mortality in Beijing. Daily time-series data in the period of 1998–2000, including key air pollutants (sulfate dioxide (SO2), carbon monoxide (CO), nitrogen oxidants (NOx), total suspended particles (TSP) and particulate matter aerodynamic diameter <10 um (PM10)), meteorological factors and health outcomes (deaths of respiratory disease, cardiovascular and cerebrovascular disease, chronic obstructive pulmonary disease and coronary heart disease were collected from eight districts of Beijing. We performed both single and multi-factor Poisson regression for cause-specific deaths, controlling the potential confounding factors such as seasons, meteorological factors (air temperature, air humidity), as well as the influence of flu epidemics in winter of 1998. All the missing value of the Data was interpolated by the SAS expand procedure for regression analysis. The results showed that levels of the air pollutants in Beijing are higher than the national air quality standard of China, especially for NOx and TSP. In single-factor models, a significant positive correlation was found between the four pollutants and daily mortality except for TSP and coronary heart disease deaths. In multi-factor Poisson regression analyses, we found that with each increase of 100 ug/m3 of SO2, respiratory deaths, cardiovascular cerebrovascular disease, coronary heart disease deaths, chronic obstructive pulmonary disease deaths were estimated to increase 4.21% (95% CI: 1.85%, 6.86%), 3.97% (95% CI: 2.44%, 5.53%), 10.68% (95% CI: 8.11%, 13.32%), 19.22% (95% CI: 14.52%, 24.1%), respectively. Meanwhile, each increase of 100 ug/m3 of TSP, respiratory disease deaths, cardiovascular and cerebrovascular disease deaths were estimated to increase 3.19% (95% CI: 1.45%, 4.96%) and 0.62% (95% CI: 0.01%, 1.17%), respectively. The results suggested that an increase of the air pollutants level in Beijing would positively associated with daily population mortality and this association could be estimated quantitatively.