OBJECTIVES : Previous ecologic studies showed increased asthma-related healthcare use when pollen levels are high. This study described the association between pollen levels and likelihood of asthma-related inpatient (IP) admissions or emergency department (ED) visits in children with asthma.

METHODS : Children with asthma were identified from the MarketScan Commercial Database and linked to pollen data in the IBM MarketScan Weather Database between March and October 2016. Those with an asthma-related IP admission or ED visit were cases; date of the IP/ED visit was the index date. Controls had no asthma-related IP/ED utilization during 2016 and the index date was the date of first asthma-related outpatient encounter in 2016. Cases and controls were matched on index month at a 1:1 ratio. Grass and ragweed pollen levels (none, low, moderate, high) were assigned using the maximum pollen value in the five days before the index date. Adjusted odds ratios (OR) and 95% confidence intervals (CI) were calculated using conditional logistic regression to describe the likelihood of asthma-related IP/ED visits while controlling for baseline patient demographic and asthma-related clinical characteristics.

RESULTS : 11,502 cases and controls were matched at 1:1 ratio; 63% were male. Likelihood of asthma-related IP/ED use was significantly higher when the maximum pollen level was high for both grass pollen (OR: 1.25; 95% CI: 1.09,1.43) and ragweed pollen (OR: 1.18; 95% CI: 1.02,1.38) vs. none. Moderate grass pollen levels were also associated with higher likelihood of asthma-related IP/ER visits (OR: 1.17; 95% CI: 1.09, 1.43).

CONCLUSIONS : Probability of an asthma-related IP/ER visit in children with asthma was significantly increased when grass or ragweed pollen levels were high after controlling for baseline patient characteristics. Further research is necessary to evaluate the role of seasonal variation in pollen type on the effectiveness of asthma medications and asthma-related healthcare utilization.