369 - Variation in Asthma Adverse Events among Medicaid-enrolled Children in Arkansas

Poster Number: 369
Publication Number: 369.302

Akilah A. Jefferson, University of Arkansas for Medical Sciences College of Medicine, Little Rock, AR, United States; John M. Tilford, University of Arkansas for Medical Sciences, Little Rock, AR, United States; Anthony Goudie, University of Arkansas for Medical Sciences, Little Rock, AR, United States; Mandana Rezaeiahari, UAMS, Little Rock, AR, United States; Clare Brown, UAMS COPH, Little Rock, AR, United States; Arina Eyimina, University of Arkansas for Medical Sciences College of Medicine, Little Rock, AR, United States

Background: Asthma adverse events (AAEs) are likely to vary across providers due to differences in patient mix, access to care (e.g., rural/urban), patient management, and medication adherence. Understanding variation at the provider level is essential for intervention development and resource allocation.

Objective: To determine the extent of variation in AAEs in a population of Medicaid-enrolled children in Arkansas and to identify patient-level factors associated with adverse events at the provider level.

Design/Methods: Using the Arkansas All-Payer Claims Database we identified children with ≥1 primary or secondary asthma diagnosis, and limited to children with ≥ 11 months of coverage in both 2018 and 2019. AAEs were defined as having ≥ 1 hospitalization or emergency department visit with a diagnosis of asthma. Children were assigned to a provider based on 1) the highest number of visits to a given provider, or 2) the last provider seen if the number of visits were tied. Finally, we compared an a priori listing of demographic and clinical variables across quintiles of providers (≥30 patients), defined based on rates AAEs.

Results: 8000 children were attributed to 146 providers with 29-31 providers per quintile. Adverse event rates varied over 20-fold, from 0.51% in the first quintile to 10.28% in the fifth quintile (p< 0.001).  The percentage of Black children was lowest in the quintile with the largest percentage of AAEs at 25.2% compared to an overall percentage of 37.6% Black children, while the percentage of White children was highest in the quintile with the highest AAEs. Rates of atopy and allergic rhinitis were lowest in the quintile with the highest rate of AAEs with mean quintile rates of 64.2% and 49.5% compared to overall rates of 73.7% and 61.6%. Comorbid depression and anxiety rates were 2.0 and 1.7 times as high among the lowest versus highest quintile. Children in the highest quintile had the lowest rate of inhaled corticosteroid monotherapy at 44.6% compared to a rate of 49.43% overall; the rate of oral corticosteroids among the highest quintile was 36.6% compared to 33.09% overall. No explanatory variables were consistent with the large variation in AAEs across quintiles with a ratio of 20.15 suggesting that a significant amount of variation in AAEs is unexplained.

Conclusion(s): AAEs varied considerably across providers in a population of rural, underserved children covered by Medicaid. The unexplained variation in AAEs suggests a need for further analysis and exploration to identify factors associated with better or worse outcomes that can lead to targeted interventions for at-risk children.