Alopecia areata is a chronic, T-cell-mediated disease that targets hair follicles and causes patches of nonscarring hair loss. Despite being relatively common (prevalence of 1-2%), current studies on alopecia areata epidemiology are conflicting, and no large-scale epidemiologic studies have been performed in children in the United States to date. It is still unclear if prevalence differs by age, other demographics such as race/ethnicity and sex, or if there are regional or temporal differences in prevalence and incidence within the US. It is essential to understand which populations are most heavily affected by alopecia areata to improve our understanding of the disease as a whole. 

In order to produce an accurate epidemiologic description of alopecia areata in children, a sizeable cohort of correctly diagnosed patients is necessary. Better algorithms are needed to more reliably identify children with alopecia areata so that large databases can be queried to quickly and accurately produce cohorts of affected patients. Our goal is to develop an electronic health record-based computable algorithm to identify children with alopecia areata and to utilize this algorithm to generate accurate prevalence and incidence data. We will subcategorize prevalence among specific subpopulations based on age, sex, race/ethnicity, insurance type, and economic background. In addition, we aim to describe rates of receipt of novel, off-label therapeutic agents in specific subgroups with the goal of identifying disparities in care among children with alopecia areata.