Use of Insurance Claims Data to Determine Prevalence and Confirm a Cluster of Sarcoidosis Cases in Vermont

Abstract

Objectives. In 2006, the Vermont Department of Health was asked to respond to a potential cluster of sarcoidosis cases related to a Vermont office building. Sarcoidosis prevalence has not been formally described for the United States. A range of <1–40/100,000 is commonly reported; however, we have not identified primary sources supporting this conclusion. Because of the wide prevalence range and lack of a local estimate, confirming existence of a cluster was difficult. Methods. We ascertained the prevalence of sarcoidosis cases in Vermont by using insurance claims data to determine whether or not a cluster of sarcoidosis cases was related to the office building. We calculated county and state annual prevalence proportions for sarcoidosis for 2004 and 2005 and annual building prevalences for 1992–2006. Results. The pooled sarcoidosis case prevalence for Vermont was 66.1/100,000. The pooled building annual prevalence (1,128/100,000) was statistically different from the county in which the building is located (odds ratio = 15.5, 95% confidence interval 3.0, 50.3). Conclusions. We reported the first statewide sarcoidosis prevalence in the United States. This prevalence exceeded previous limited and unsubstantiated U.S. reports. Even with Vermont's elevated sarcoidosis prevalence, the presence of a cluster in this building was apparent.