Abstracts: ISEE 21st Annual Conference, Dublin, Ireland, August 25-29, 2009: Oral Presentations
Background and Objective:
A relatively small number of air pollutants are regularly monitored. The “criteria” pollutants are monitored with far greater frequency than others. These measures are supplemented at a limited number of sites by measurements of several particulate matter components. Further supplementation is rare. Epidemiological studies require monitoring data to estimate exposures. Cohort studies require these estimates for locations where study subjects reside; the studies are limited to the air quality components monitored.
A cohort study was undertaken to examine several pollution variables, including those regularly monitored and measures of pollutants less regularly monitored. All pollution variables were related to the mortality experience of the cohort using Poisson regression, adjusting for other factors which influence health status.
Initial analyses related the survival status of this group to the “criteria” pollutants; positive associations were found for both ozone and NO2. With more extensive particulate matter species data included in the analyses, the “criteria” pollutants were no longer important, and EC, nitrate, vanadium and nickel were of greatest concern. Further analyses were undertaken using estimated concentrations of several hazardous air pollutants variables, derived by US EPA. These analyses pointed to several pollutants which were more important than the previous variables in joint analyses; these pollutants included benzene, formaldehyde, and polycyclic organics, none of which are routinely monitored.
Epidemiological studies are limited by available monitoring data. Limited data can affect the ability to find statistically significant relationships as well as the ability to determine the details of these relationships. This may stymie the identification and subsequent control of key air quality components that may cause health impacts.