Spatiotemporal analysis for the effect of ambient particulate matter on cause-specific respiratory mortality in Beijing, China

This study explored the association between particulate matter with an aerodynamic diameter of less than 10 μm (PM₁₀) and the cause-specific respiratory mortality. We used the ordinary kriging method to estimate the spatial characteristics of ambient PM₁₀ at 1-km × 1-km resolution across Beijing during 2008–2009 and subsequently fit the exposure-response relationship between the estimated PM₁₀ and the mortality due to total respiratory disease, chronic lower respiratory disease, chronic obstructive pulmonary disease (COPD), and pneumonia at the street or township area levels using the generalized additive mixed model (GAMM). We also examined the effects of age, gender, and season in the stratified analysis. The effects of ambient PM₁₀ on the cause-specific respiratory mortality were the strongest at lag0-5 except for pneumonia, and an inter-quantile range increase in PM₁₀ was associated with an 8.04 % (95 % CI 4.00, 12.63) increase in mortality for total respiratory disease, a 6.63 % (95 % CI 1.65, 11.86) increase for chronic lower respiratory disease, and a 5.68 % (95 % CI 0.54, 11.09) increase for COPD, respectively. Higher risks due to the PM₁₀ exposure were observed for females and elderly individuals. Seasonal stratification analysis showed that the effects of PM₁₀ on mortality due to pneumonia were stronger during spring and autumn. While for COPD, the effect of PM₁₀ in winter was statistically significant (15.54 %, 95 % CI 5.64, 26.35) and the greatest among the seasons. The GAMM model evaluated stronger associations between concentration of PM₁₀. There were significant associations between PM₁₀ and mortality due to respiratory disease at the street or township area levels. The GAMM model using high-resolution PM₁₀ could better capture the association between PM₁₀ and respiratory mortality. Gender, age, and season also acted as effect modifiers for the relationship between PM₁₀ and respiratory mortality.