Ambient Particle Source Apportionment and Daily Hospital Admissions among Children and Elderly in Copenhagen

Abstract

TM3-O-03 Introduction: An association between particulate air pollution and morbidity and mortality is well established. However, little is known about which sources of particulate matter contribute most to the adverse health effects. Identification of responsible sources would be valuable in both, providing additional insight into biologic mechanisms, and targeting of air pollution control. Methods: For a 6-year period (January 1, 1999 to December 31, 2004), we studied the effect of PM10 and gaseous pollutants (CO, NO2) on hospital admissions in Copenhagen, Denmark, and used source apportionment of PM10 data from a 1.5-year campaign (May 3, 2003 to December 22, 2004) to evaluate the hypothesis that certain sources (biomass, secondary, oil, crustal, sea salt, and vehicle) are more closely associated with health than total PM10 mass concentration. We studied 3 health outcomes, hospital admissions due to cardiovascular disease in elderly (age ≥65 years), respiratory disease in elderly (age ≥65 years), and asthma in school-aged children (5–18 years), in an attempt to cover the array of diseases and susceptible populations and provide a comprehensive study of short-term effects of air pollution on morbidity. We used Poisson-generalized additive model adjusted for season, day of the week, public holidays, school holidays, influenza epidemics, grass pollen, and meteorology, with up to 5 days lagged air pollution exposure. Results: Increase of one interquartile range (14 μg/m3) for 4-, 5-, and 6-day PM10 concentration averages, respectively, resulted in 2.7% (95% CI, 1.3%–4.2%) increase in cardiovascular, 3.7% (1.4%–6.0%) increase in respiratory, and 7.7% (0.4%–15.5%) increase in asthma hospital admissions the following day. The PM10 effect remained robust in the presence of CO and NO2. We found that crustal and secondary sources of PM10 showed strongest associations with cardiovascular, biomass with respiratory, and vehicle and biomass with asthma admissions. No single PM10 source could be attributed to all morbidity. Discussion and Conclusions: We found short-term associations between PM10 and morbidity in Copenhagen, with strongest associations for asthma in children, implying higher susceptibility of children than adults. This study brings new evidence that different PM10 sources are relevant for different diseases, implicating biomass and vehicle sources relevance for respiratory disease and asthma, and secondary and crustal sources importance for cardiovascular disease. These novel results on PM10 sources may give insight into causal pathways, pathophysiologic mechanisms, and toxicity. However, without identifying single important source for all morbidity, our results reinforce the importance of control of total mass PM10.