Seasonal Pattern of the Acute Mortality Effects of Air Pollution

Evidence of seasonal variation of acute mortality effects of air pollution is inconsistent. The seasonal patterns of associations between daily mortality and daily mean concentrations of particulate matter 10 microm or less in aerodynamic diameter (PM10), sulfur dioxide (SO2), and nitrogen dioxide (NO2) were examined using 4 yr of data (2001-2004) in Wuhan, China. Four distinct seasons occur in Wuhan, where approximately 4.5 million residents live in the city core area of 201 km2. Air pollution levels are higher and pollution ranges are wider in Wuhan than in most cities. Quasi-likelihood estimation within the context of the generalized additive models (natural spline [NS] models in R) was used to model the natural logarithm of the expected daily death counts as a function of the predictor variables. The estimates of the interaction between seasons and pollution were obtained from the main effects and pollutant season interaction models. It was found that the interactions between three pollutants and cause-specific mortality were statistically significant (P < 0.05). The strongest effects occurred consistently in winter for all-natural, cardiovascular, stroke, and respiratory mortality. Every 10-microg/m3 increase in PM10 daily concentration at lag 0-1 days was associated with an increase in all-natural mortality of 0.69% (95% confidence interval [CI]: 0.44-0.94%) for winter, 0.34% (95% CI: 0.00-0.69%) for spring, 0.45% (95% CI: -0.13 to 1.04%) for summer, and -0.21% (95% CI: -0.54 to 0.12%) for fall. The results show a clear seasonal pattern of acute mortality effects of ambient air pollution and the strongest effects occurred during winter in the study city.