Associations Between Ambient Nitrogen Dioxide and Daily Cause-specific Mortality Evidence from 272 Chinese Cities

Background: There has been a long history of debate regarding whether ambient nitrogen dioxide (NO2) directly affects human health. Methods: We conducted a nationwide time-series analysis in 272 major Chinese cities (2013-2015) to evaluate the associations between short-term exposure to NO2 and cause-specific mortality. We used the overdispersed generalized linear model together with the Bayesian hierarchical model to estimate the associations between NO2 and mortality at the national and regional levels. We examined two-pollutant models with adjustment of fine particles, sulfur dioxide, carbon monoxide, and ozone to evaluate robustness for the effects of NO2. Results: At the national-average level, we observed linear and positive associations between NO2 and mortality from all causes and main cardiorespiratory diseases. A 10 g/m(3) increase in 2-day average concentrations of NO2 would lead to increments of 0.9% (95% posterial interval [PI], 0.7%, 1.1%) in mortality from total nonaccidental causes, 0.9% (95% PI, 0.7%, 1.2%) from total cardiovascular disease, 1.4% (95% PI, 0.8%, 2.0%) from hypertension, 0.9% (95% PI, 0.6%, 1.2%) from coronary heart disease, 0.9% (95% PI, 0.5%, 1.2%) from stroke, 1.2% (95% PI, 0.9%, 1.5%) from total respiratory diseases, and 1.6% (95% PI, 1.1%, 2.0%) from chronic obstructive pulmonary disease. There were no appreciable differences in estimates from single-pollutant and two-pollutant models. The associations were stronger in the south of China, in the elderly, and in females. Conclusions: The present study provided robust epidemiologic evidence of associations between day-to-day NO2 and mortality from total natural causes and main cardiorespiratory diseases that might be independent of other criteria air pollutants.