LONG-TERM AMBIENT CONCENTRATIONS OF PARTICULATES AND OXIDANTS AND DEVELOPMENT OF CHRONIC DISEASE IN A COHORT OF NONSMOKING CALIFORNIA RESIDENTS

A cohort of 6340 nonsmoking California Seventh-Day Adventists (SDAs) w ho had resided within 5 miles of their present residence for the past 10 yr has been followed since 1977 for incidence of cancer and myocard ial infarction (MI) through 1982; development of definite symptoms of, and increasing severity of, airway obstructive disease (AOD), chronic bronchitis, and asthma through 1987; and all natural cause mortality through 1987. Cumulative ambient concentrations of specific pollutants have been estimated for study participants from 1967 to 1987 by inter polating monthly statistics from statewide air monitoring stations to ZIP codes of residence and work location. Statistics include excess co ncentrations and exceedance frequencies above a number of cutoffs as w ell as mean ambient concentration and mean ambient concentration adjus ted for time spent indoors. Indoor sources or nitrogen (NO2), and of p articulate pollution such as environmental tobacco smoke, both at home and at work, as well as occupational dusts and fumes, have been adjus ted for in multivariate statistical models. Particulates included tota l suspended particulates (TSP), monitored from 1973 to 1987; inhalable particulates less than 10 mu m in diameter (PM-10), estimated from si te/seasonal-specific regressions on TSP for 1973-1987; fine particulat es less than 2.5 mu m in diameter estimated from airport visibility da ta for 1967-1987; and suspended sulfates (SO4), monitored from 1977 to 1987. A direct measure of visibility, and gaseous pollutants-ozone, s ulfur dioxide (SO2), and (NO2)-monitored from 1973 to 1987 were also i ncluded in analyses. No statistically significant associations between any of the disease outcomes studied and NO2 or SO2 were found in this cohort. None of die pollutants studied showed statistically significa nt associations with all natural cause mortality or incidence of all m alignant neoplasms in males. Statistically significant associations we re observed between elevated ambient concentrations of one or more par ticulate pollutants and each of the other disease outcomes. In additio n, ozone was significantly associated with increasing severity of asth ma, and with the development of asthma in males. Multipollutant analys es indicated that none of the associations between particulate polluta nts and disease outcomes were due to correlations with gaseous polluta nts studied except possibly for PM2.5 and increasing severity of asthm a, which could be due to a correlation with ozone. Observed associatio ns between disease outcomes and PM2.5 or PM-10 could be biased toward the null because of increased measurement error due to their indirect methods of estimation.