Short term effects of particulate matter on cause specific mortality: effects of lags and modification by city characteristics

Background: Consistent evidence has shown increased all-cause mortality, and mortality from broad categories of causes associated with airborne particles. Less is known about associations with specific causes of death, and modifiers of those associations. Aims: To examine these questions in 20 US cities, between 1989 and 2000. Methods: Mortality files were obtained from the National Center for Health Statistics. Air pollution data were obtained from the Environmental Protection Agency website. The associations between daily concentrations of particulate matter of aero-diameter <= 10 mu m (PM10) and daily mortality from all-cause and selected causes of death, were examined using a case-crossover design. Temporal effects of PM10 were examined using lag models, in first stage regressions. City specific modifiers of these associations were examined in second stage regressions. Results: All-cause mortality increased with PM10 exposures occurring both one and two days prior the event. Deaths from heart disease were primarily associated with PM10 on the two days before, while respiratory deaths were associated with PM10 exposure on all three days. Analyses using only one lag underestimated the effects for all-cause, heart, and respiratory deaths. Several city characteristics modified the effects of PM10 on daily mortality. Important findings were seen for population density, percentage of primary PM10 from traffic, variance of summer temperature, and mean of winter temperature. Conclusions: There was overall evidence of increased daily mortality from increased concentrations of PM10 that persisted across several days, and matching for temperature did not affect these associations. Heterogeneity in the city specific PM10 effects could be explained by differences in certain city characteristics.