the relationship between pollen air pollution and asthma exacerbations in children in allegheny county pennsylvania a case crossover analysis

Abstract Background: Exposures to outdoor air pollutants have been linked to asthma exacerbations in children. Few studies have examined the association between exposure to outdoor pollens and asthma outcomes in the context of its association with multiple air pollutants. Methods: Time-stratified case-crossover design with conditional logistic regression was used to study the short-term effects of three major pollens (grass, tree, and weed) and four criteria pollutants (PM 2.5, Ozone, SO2 and NO2) on asthma Emergency Department (ED) visits in children age 5-17 reported in Allegheny County, Pennsylvania from April to October 2003-2011. Multivariable regression was conducted to investigate the effects of pollen and pollutant levels on the day of the ED visit, lags of day 1 to 5 and moving averages of day 0-2 and day 0-5. Results: A total of 8,711 asthma ED visits were reported during the study period. In multivariable models, tree and weed pollen were significant positive predictors of asthma ED visits across multiple lags when controlling for temperature and air pollutants. Strongest effects were reported for the 3-day moving average of tree pollen (odds ratio, OR=1.02, 95% CI 1.01-1.02) and the 6-day moving average of weed pollen (OR=1.04, 95% CI 1.03-1.06). PM 2.5 and NO2 were significantly positively associated with ED visits across multiple lags, whereas SO2 was negatively associated with ED visits at several lags. Discussion: Higher tree and weed pollen levels were associated with increased odds of asthma ED visits in children, independent of air pollution levels. Implementing methods to control allergen exposure during particular seasons may prevent adverse asthma outcomes. Keywords Air pollution; Asthma exacerbations; Children; Pennsylvania; Pollen Abbreviations ED: Emergency Department; PM: Particulate Matter; O3: Ozone; SO2: Sulfur Dioxide; NO2: Nitrogen Dioxide; LUR: Land Use Regression; NLDAS: North America Land Data Assimilation System; C-CAT: Case-crossover Analysis Tool; OR: Odds Ratio; CI: Confidence Interval