Associations between short-term exposure to ambient sulfur dioxide and increased cause-specific mortality in 272 Chinese cities

Background: Ambient sulfur dioxide (SO2) remains a major air pollutant in developing countries, but epidemiological evidence about its health effects was not abundant and inconsistent. Objectives: To evaluate the associations between short-term exposure to SO2 and cause-specific mortality in China. Methods: We conducted a nationwide time-series analysis in 272 major Chinese cities (2013-2015). We used the over-dispersed generalized linear model together with the Bayesian hierarchical model to analyze the data. Two-pollutant models were fitted to test the robustness of the associations. We conducted stratification analyses to examine potential effect modifications by age, sex and educational level. Results: On average, the annual-mean SO2 concentrations was 29.8 mu g/m(3) in 272 cities. We observed positive and associations of SO2 with total and cardiorespiratory mortality. A 10 mu g/m(3) increase in two-day average concentrations of SO2 was associated with increments of 0.59% in mortality from total non-accidental causes, 0.70% from total cardiovascular diseases, 0.55% from total respiratory diseases, 0.64% from hypertension disease, 0.65% from coronary heart disease, 0.58% from stroke, and 0.69% from chronic obstructive pulmonary disease. In two-pollutant models, there were no significant differences between single-pollutant model and two-pollutant model estimates with fine particulate matter, carbon monoxide and ozone, but the estimates decreased substantially after adjusting for nitrogen dioxide, especially in South China. The associations were stronger in warmer cities, in older people and in less-educated subgroups. Conclusions: This nationwide study demonstrated associations of daily SO2 concentrations with increased total and cardiorespiratory mortality, but the associations might not be independent from NO2.