(A33) A Retrospective Cohort Study Using Administrative Claims to Compare the Incidence of Eosinophilic Granulomatosis with Polyangiitis in Patients with and without Asthma, Stratified by Asthma Severity

Publication Number: 316

Background
Eosinophilic granulomatosis with polyangiitis (EGPA) is a rare condition with systemic effects such as necrotizing vasculitis, extra-vascular granulomas, and eosinophilic inflammation. Notably, asthma can be a symptom of EGPA, and pre-existing asthma is associated with EGPA. Prior studies of EGPA incidence in claims data were limited by non-specific ICD-9 codes. No studies have used the EGPA- ICD-10 code to identify EGPA, nor compared the incidence of EGPA in patients with and without asthma.
Objectives
To compare the incidence of EGPA in patients with and without asthma and to evaluate the influence of asthma severity on EGPA incidence.
Methods
We conducted a retrospective cohort study using Optum claims data from Oct 2015 to Mar 2020. Patients with ≥2 ICD-10 codes for asthma (J45.xx) ≥30 days apart were included in the asthma cohort. Index was the date of the second diagnosis. Asthma patients were matched 1:1 to patients without asthma (controls) on age, sex, and calendar month and year of index. Patients with <6 months of continuous enrollment and evidence of EGPA prior to the index date were excluded. We adapted a published algorithm to identify EGPA; ≥2 claims with an ICD-10 code for EGPA (M30.1) ≥90 days apart plus ≥1 claim from a relevant specialist. Patients were followed from the index date until the first of EGPA, disenrollment, or 31 Mar 2020. Incidence rates (IRs) and incidence rate ratios (IRRs) with 95% confidence intervals (95% CI) were estimated using Poisson regression. Asthma medication use at cohort entry, categorized according to GINA guidelines, was a proxy for asthma severity.
Results
There was a total of 1,555,274 PYs of follow up for the asthma group and 1,728,829 PYs for the non-asthma group. IRs of 2.06 (95% CI 1.35, 2.83) and 0.35 (95% CI 0.12, 0.64) per 100,000 PY were calculated for patients with and without asthma, respectively. After adjusting for the use of asthma medications, which may mask EGPA, the IRR remained similar, but the estimate was less precise (IRR=2.14, 95% CI 0.84, 5.46).
EGPA incidence varied by disease severity; IRs for EGPA in patients with severe, high-moderate, low-moderate, and mild asthma were 23.62 (95% CI 0.00, 70.86), 3.69 (95% CI 1.70, 5.96), 2.39 (95% CI 0.00, 7.17), and 1.47 (95% CI 0.78, 2.25) per 100,000 PY, respectively. IRRs were 12.79 (95% CI 1.60, 102.01) for severe asthma, 2.19 (95% CI 1.06, 4.54) for high-moderate, and 1.52 (95% CI 0.20, 11.51) for low-moderate asthma versus mild asthma.
Conclusions
Using the ICD 10 code, we found a higher incidence of EGPA in patients with asthma as compared to the non-asthma matched population, and the estimate remained similar after adjusting for the use of asthma medications. EGPA incidence was highest in patients with severe asthma.