431 Assessment of racial and ethnic differences of atopic dermatitis severity and treatment patterns in a diverse outpatient cohort in the United States

Abstract

Abstract Previous population-based studies in the USA found racial and ethnic differences of atopic dermatitis (AD) severity and treatment patterns. It is unclear whether these differences are from differences of disease characteristics or disparities. This study aims to examine racial and ethnic differences in severity and treatment patterns in a diverse outpatient patient cohort of patients with AD (n = 833). A retrospective observational study was performed of patients with physician-diagnosed atopic AD from a large metropolitan tertiary care medical center. Medical records were reviewed for race, ethnicity, disease severity and use of therapies from 10/01/2015-10/1/2021. Atopic dermatitis cases were identified using Hanifin–Rajka criteria. Chi-square or Fisher’s exact tests or were used to assess associations of race and ethnicity with AD severity. Generalized linear models were used to assess interactions of race or ethnicity with gender or age as predictors of individual treatments. There were no significant associations of body surface area (BSA; Fisher’s exact test, P = 0.19 and P = 0.44) or physician’s global assessment (PGA; P = 0.63 and P = 0.57) with race or ethnicity; nor interactions of race/ethnicity with gender or age as predictors of BSA or PGA. Asian and multiracial/other patients were more likely than White or Black patients to use topical calcineurin inhibitors (Chi-square, P = 0.01). Dupilumab use differed by race (Multiracial/other = 35.0%; White = 20.1%; Asian = 15.7%; Black = 13.6%; Chi-square, P = 0.03), but not ethnicity (P = 0.88). Use of oral corticosteroids (Chi-square, P = 0.74), immunosuppressants (P = 0.98) or GABAergics (P = 0.16), or NBUVB (P = 0.42) did not differ by race. There were no statistical interactions of race or ethnicity with gender or age as predictors of treatment use. This study demonstrates that AD severity and specific treatments did not differ overall between racial and ethnic groups in this diverse urban academic cohort.