Abstracts: ISEE 22nd Annual Conference, Seoul, Korea, 28 August–1 September 2010: Air Pollution - Cardiovascular Health Effects
Substantial evidence supports an association of particulate matter (PM) with cardiorespiratory illnesses, but less is known regarding characteristics of PM and/or sources that contribute to this association. Here, we present results from a time-series investigation of emergency department (ED) visits in St. Louis, MO.
Individual-level data were obtained from the Missouri Hospital Association for all ED visits to acute-care hospitals in the St. Louis Metropolitan area during 1 January 2001 to 30 June 2007. The data set included data for 36 hospitals and >5,600,000 ED visits. Daily ambient air monitoring data for the major urban pollutants were obtained from the USEPA Air Quality System for representative sites within the St. Louis study area. Daily speciated data (including PM2.5 metals and non-polar organics) were obtained from the St. Louis—Midwest Supersite for a 2-year period, and were used in an 11-factor positive matrix factorization analysis to obtain daily source contribution estimates. We used Poisson generalized linear models, controlling for long-term temporal trends and meteorological variables, to examine associations between 0-1-2 day air quality measures and daily respiratory ED visits, and between same day air quality measures and cardiovascular ED visits.
Over the study period, we observed mean ED visit counts of 282 per day for respiratory outcomes and 100 per day for cardiovascular outcomes. Analyses of the criteria pollutants indicated strong positive associations between ozone and ED visits for all respiratory outcomes, URI, and asthma, as well as between NO2 and PM2.5 and asthma. With cardiovascular ED visits, associations were observed with CO. Preliminary analyses of speciated pollutants have focused on selected PM2.5 elemental species (including elemental [EC] and organic carbon [OC], sulfur, zinc, silicon, and potassium). Results to date indicate elevated associations between cardiovascular ED visits and traffic-related indicators (zinc, EC, OC). Further analyses of the PM2.5 elements, as well as key indicators of sources using non-polar organics data (including chrysene, benzo(b+k)fluoranthene, benzo(a)pyrene, indeno(123-cd)pyrene, norhopane, hopane, n-Octacosane, n-Nonacosane) and source apportioned estimates.
Preliminary results indicate impacts of both primary (eg, CO, NO2, EC) and secondary (eg, ozone, PM2.5) pollutants on acute cardiorespiratory morbidity in St. Louis, and are comparable with results of previous similar studies in other cities. Future analyses will consider associations between these outcomes and nonpolar organics, and may help to further describe important emission sources for pollutant-related health effects.