Assessment of the American Academy of Dermatology diagnostic criteria for pediatric atopic dermatitis and modification into a checkbox form: A cross-sectional study.

Abstract

Diagnostic criteria for atopic dermatitis (AD) are limited in their performance and/or usability. The American Academy of Dermatology (AAD) consensus criteria include hierarchical categories of disease features to improve these metrics but have not been validated. Our objective was to create and validate a checkbox form of the AAD consensus criteria in the pediatric population. We performed a cross-sectional study of 100 pediatric patients with AD (n = 58) and diseases in the differential diagnosis of AD (n = 42). Having three or more "Essential," ≥2 "Important," ≥1 "Associated" features of the AAD criteria was optimal for the diagnosis of AD in children. This combination was 91.4% (95% CI, 84.2%-98.6%) sensitive and 95.2% (88.8%-100%) specific. The UK working party criteria and the Hanifin-Rajka criteria had sensitivities of 96.6% (95% CI 91.9%-100%) and 98.3% (95% CI 94.9%-100%) and specificities of 83.3% (95% CI 72.1%-94.6%) and 71.4% (95% CI 57.8%-85.1%), respectively. The AAD criteria had significantly greater specificity than the Hanifin-Rajka criteria (p = .002). This study represents an important step in validating the AAD consensus criteria and formulating a useable checkbox form for diagnosing AD in the pediatric population.