Abstract
BACKGROUND: Prior studies have identified predictors of asthma exacerbations; however, most lack integration of type 2 (T2) inflammatory markers. RESEARCH QUESTION: In a large electronic health record database, what are predictors of asthma exacerbation rates and is there interaction by T2 inflammation, female sex, and obesity? STUDY DESIGN AND METHODS: This is a retrospective cohort study using electronic health record data of patients with asthma followed for at least 1 year in the UCHealth system. The primary outcome was asthma exacerbation rate, defined by the prescription of an oral corticosteroid burst. Predictors of interest included T2 high inflammation, defined as absolute eosinophil count (AEC) ≥ 300 cells/μL, BMI, and sex. Predictors of the numbers of exacerbation and prespecified interactions were identified with negative binomial models. A natural cubic spline was used to model the dose response between AEC and exacerbation rate. RESULTS: The cohort included 70,939 patients with asthma; 52% had T2 high inflammation and 62% were female, with 70% of patients being overweight or obese. Individuals with T2 high inflammation had higher adjusted rates of exacerbation (adjusted incidence rate ratio, 1.13; 95% CI, 1.10-1.16). AEC predicted exacerbation frequency in a dose-dependent manner. There was significant effect modification by sex, with female participants with T2 high inflammation having increased exacerbation rates compared with male participants with T2 high inflammation. INTERPRETATION: This study finds an increase in exacerbation rate among patients with T2 high inflammation with asthma and shows a dose-dependent response to AEC. To our knowledge, this is the first study to find effect modification by sex and T2 status, identifying a group of patients who could potentially benefit from T2-targeted biologic therapy to decrease their exacerbation rate.