Aggregate population-based cross-sectional studies using averages across various geopolitical units (cities, metropolitan statistical areas (MSAs), and so on) have examined the relation between mortality and long-term PM exposure. Those community-based studies sought to define the characteristics of a community that are associated with its overall average health status, in this case annual mortality. For example, Ozkaynak and Thurston (1987) analyzed 1980 total mortality in 98 MSAs, using data on PM15 and PM2.5 from the EPA inhalable-particle monitoring network for 38 of these locations. They concluded that the results suggested an effect of particles on mortality that decreased with increasing particle size.
Prospective cohort studies have considered the effect of PM exposure on the relative survival rates of individuals, as modified by age, sex, race, smoking, and other individual risk factors, finding that PM exposure can lead to substantial shortening of life in the general population. That type of analysis has a substantial advantage over aggregate population-based studies, in that the individual analysis allows stratification according to such important risk factors as smoking. Abbey et al. (1991) described a prospective cohort study of morbidity and mortality in a population of about 6,000 white, non-Hispanic, nonsmoking, long-term California residents who were followed for 6-10 years beginning in 1976. TSP and ozone were the only pollutants considered. In a followup analysis, Abbey et al. (1995) considered exposures to SO42-, PM10, and PM2.5, as well as visibility (extinction coefficient). In these analyses, no significant associations with nonspecific mortality (i.e., from all natural causes) were reported, and only high concentrations of TSP or PM10 were associated with respiratory symptoms of asthma, chronic bronchitis, or emphysema. However, a more recent analysis using an additional 5 years of follow-up on this cohort and improved PM10 exposure estimates did predict significant PM-mortality associations among men in this cohort, who reportedly spent significantly more time outdoors than women (Abbey et al. 1999). Dockery et al. (1993) analyzed the mortality experience in 8,111 adults who were first recruited in the middle 1970s in 6 cities in the eastern portion of the United States. The subjects were white and 25-74 years old at enrollment. Dockery et al. (1993) reported that “mortality was more strongly associated with the levels of fine, inhalable, and sulfate particles” than with the other pollutants. Pope et al. (1995) analyzed 7-year survival data (1982-1989) for about 550,000 adult volunteers obtained by the American Cancer Society (ACS). They took great care to control for potential confounding factors on which data were available. For example, several different measures of active smoking were considered, as was the time exposed to passive smoke. The adjusted total-mortality risk ratios for the ACS study, computed for the cities' range of the pollution exposures, were as follows: 1.15 (95% confidence interval, 1.09-1.22) for a 19.9 µg/m3 increase in sulfates and 1.17 (95% confidence interval, 1.09-1.26) for a 24.5 µg/m3 increase in PM2.5. Analysis of life-tables indicate that these effects