Most time-series studies of particulate air pollution and acute health outcomes assess exposure of the study population using fixed-site outdoor measurements. To address the issue of exposure misclassification, we evaluate the relationship between ambient particle concentrations and personal exposures of a population expected to be at risk of particle health effects.
IMPLICATIONSAs epidemiological studies aim to attribute observed health effects to outdoor air pollution, the contribution of outdoor air pollutants to population exposure must be assessed. To support time-series epidemiological studies that evaluate the health impacts of short-term changes in air pollution, an exposure metric is required for which the day-to-day variability in personal exposures reflects the variability of ambient concentrations being measured. We found that personal exposures to particulate sulfate in a group of subjects at risk for adverse health effects of particulate air pollution were highly correlated, over time, with ambient particle concentrations. Personal exposures to sulfate were highly correlated with ambient levels across all individuals and all levels of exposure, whereas correlations between personal exposures to PM 2.5 and ambient particle concentrations were lower and more variable. Overall, we have shown that a personal measure of exposure to outdoor source PM is highly related to variation in ambient levels of PM, lending support to time-series epidemiological studies.Sampling was conducted within the Vancouver metropolitan area during April-September 1998. Sixteen subjects (non-smoking, ages 54-86) with physician-diagnosed chronic obstructive pulmonary disease (COPD) wore personal PM 2.5 monitors for seven 24-hr periods, randomly spaced approximately 1.5 weeks apart. Time-activity logs and dwelling characteristics data were also obtained for each subject. Daily 24-hr ambient PM 10 and PM 2.5 concentrations were measured at five fixed sites spaced throughout the study region. SO 4 2-, which is found almost exclusively in the fine particle fraction and which does not have major indoor sources, was measured in all PM 2.5 samples as an indicator of accumulation mode particulate matter of ambient origin.The mean personal and ambient PM 2.5 concentrations were 18 µg/m 3 and 11 µg/m 3 , respectively. In analyses relating personal and ambient measurements, ambient concentrations were expressed either as an average of the values obtained from five ambient monitoring sites for each day of personal sampling, or as the concentration obtained at the ambient site closest to each subject's home. metric, the median r between personal and average ambient concentrations was 0.96 (range = 0.66 to 1.0). Use of the closest ambient site did not improve the median correlation of the group for either PM 2.5 or SO 4 2-. All pooled analyses resulted in lower correlation coefficients than the median correlation coefficient of individual regressions. Personal SO 4 2-was more highly correlated with all ambient measures than PM 2.5 ....