An Exploratory Analysis of the Relationship Between Mortality and the Chemical Composition of Airborne Particulate Matter

Abstract

We explored relationships between daily mortality and the major sources of airborne particulate matter (PM) using a newly developed approach, Factor Analysis and Poisson Regression (FA/PR). We hypothesized that by adding information on PM chemical speciation and source apportionment to typical PM epidemiological analysis, we could identify PM sources that cause adverse health effects. The FA/PR method was applied to a merged data set of mortality and extensive PM chemical speciation (including trace metals, sulfate, and extractable organic matter) in New Jersey. Statistically significant associations were found between mortality and several of the FA-derived PM sources, including oil burning, industry, sulfate aerosol, and motor vehicles. The FA/PR method provides new insight into potentially important PM sources related to mortality. For the data set we analyzed, the use of FA/PR to integrate multiple chemical species into source-related PM exposure metrics was found to be a more sensitive tool than the traditional approach using PM mass alone.