Fine particulate air pollution and mortality in two southern California counties

Several epidemiologic studies have reported associations between respiratory disease and particulate matter less than 2.5 mu m in diameter (PM2.5) or fine particles. However, since daily fine particulate data from ambient monitors are seldom available, many studies have used estimates of PM2.5 based on visual range observed at local airports. This paper examines the impact of visibility-based estimates of PM2.5 on mortality from 1980 through 1986 in San Bernardino and Riverside Counties, located in the Los Angeles metropolitan area, Multiple regression analysis was used to isolate the effects of estimated fine particles on mortality while controlling for covariates, including season, day of week, maximum temperature, and dew point. The results are dependent on season. During the summer quarters, there was a small but statistically significant association of estimated fine particles with both total mortality (RR = 1.03, 95% CI = 1.00-1.05, evaluated at the mean PM2.5 value of 32.5 mu g/m(3)) and respiratory-specific mortality. However, for the year taken as a whole, estimated fine particles were not associated with mortality (RR = 1.00; 95% CI = 0.99-1.02). The use of estimated fine particles introduces additional measurement error into the analysis, During the summer quarters, an effect of ozone on mortality was also detected, but this association could be due to confounding with temperature. Sensitivity analysis indicated that the fine particulate association found during the summer quarters was robust to alternative estimates deriving fine particles from visual range, alternative regression specifications, functional forms, and exposure averaging times. These findings are consistent with another recent study of mortality in Los Angeles County and add to the other findings of associations between particulate matter and various respiratory diseases outcomes. (C) 1995 Academic Press, Inc.