Pollen has been generally linked to an increased risk for asthma exacerbation. However, the delayed effect (lag), the length of effect duration, and the association heterogeneity by pollen types have not been well characterized. Short-term associations between ambient concentration of various pollen types (tree, grass, and weed) and emergency department (ED) visits for asthma were assessed using data in Wake County, North Carolina, during 2006-2012. Distributed lag nonlinear models (DLNM) were used to characterize the associations, while adjusting for air pollutants, meteorological, and temporal factors. A strong association between same-day tree pollen and asthma ED visits was detected. This association lasted four days, with a 4-day cumulative risk ratio (RR) up to 2.10 (3500 grains/m(3) vs. 0 grains/m(3), 95% confidence interval [CI]=1.21-3.65). The associations of asthma ED visits with weed pollen and grass pollen were weak, suggestively starting from lag 2 and lasting 3 days, with the strongest association a 3-day cumulative RR of 1.08 (32 grains/m(3) vs. 0 grains/m(3), 95% CI=1.01-1.15) and 1.05 (11 grains/m(3) vs. 0 grains/m(3), 95% CI=1.00-1.11). Our results indicate that the association of ambient pollen and asthma exacerbation vary by pollen type, both quantitatively and temporally. These findings have important implications for optimizing targeted allergic disease prevention and management, and helping understand the etiology of ambient exposure-induced allergic diseases.
Keywords: Asthma; Distributed lag nonlinear model; Pollen concentration.
Copyright © 2015 Elsevier B.V. All rights reserved.