Su1843 The Prevalence of Eosinophilic Esophagitis in the United States: Estimates From a Large Administrative Database

Abstract

Background: Eosinophilic esophagitis (EoE) is a major cause of upper GI morbidity in both children and adults. However, most assessments of the prevalence of EoE are from single centers, and there are few data on the nationwide prevalence of EoE. Aim: To estimate the prevalence of EoE in the United States. Methods: We performed a retrospective analysis of the LifeLinkTM Health Plan Claims Database. This database is representative of a U.S. national commercially insured population and contains medical and pharmaceutical claims for .90 million individuals. We analyzed data from 2008-2011 in subjects age 0-64 who had ≥6 months of continuous enrollment. EoE was defined as a single instance of the ICD9 code 530.13. This code was approved in 2008 and we have previously validated its extremely high specificity (.99%) for diagnosis of EoE using administrative records among commercially insured patients. For this analysis, we compiled the number of patients with this code over the study timeframe, calculated the prevalence of the code in the database, and then standardized the estimate to the U.S. population by age and sex using 2010 census data. Sensitivity analysis was performed based on the operating parameters of the 530.13 diagnostic code (sensitivity of 37%) that we previously demonstrated. Results: Of the 25,368,850 individuals in this database from 2008-2011, 16,405 (0.06%) were found to have at least one diagnostic code for EoE. The mean age for these EoE patients was 33.5 ± 17 yrs, and 65% were male. Of those with a diagnostic code for EoE, 52.8% also had a diagnostic code for at least one allergic condition. Specifically, 45.9% had a code for allergic rhinitis/sinusitis, 23.4% for asthma, 4.7% for food allergy, 8.6% for atopic dermatitis, and 1.8% for urticaria. Frequency of diagnosis tended to vary by U.S. census region, with 19% in the east (54/100,000), 34% in the midwest (76/100,000), 32% in the south (70/100,000), and 15% in the west ((52/100,000; p=0.08). The cumulative prevalence of EoE increased over time from 7/100,000 in 2008, to 31/100,000 in 2009, 50/100,000 in 2010, and 65/ 100,000 in 2011, likely reflecting increasing use of the code. When standardized to the U.S. population, the estimated prevalence of EoE was 67/100,000, or approximately 178,789 cases. Applying the known sensitivity of the code to the U.S. population, and assuming only 37% of patients with EoE are identified by the code, the prevalence in the U.S. could be as high as 181/100,000 or about 480,000 cases. Conclusions: EoE is becoming more common in the U.S., with an estimated prevalence between 67 and 181/100,000. These values may underestimate the true prevalence given that both knowledge of the new ICD9 code and recognition of EoE are increasing. The estimated prevalence of EoE is beginning to approach that of Crohn's disease.