Remote identification of new onset pediatric atopic dermatitis - a comparison of diagnostic methods from a post-hoc analysis of the CASCADE prevention trial

BACKGROUND: Few validated diagnostic criteria for atopic dermatitis (AD) can diagnose AD without a clinician's exam, and even fewer are validated in very young children. Recently, the Community-based Assessment of Skin Care, Allergies, and Eczema (CASCADE) prevention trial remotely captured parent/caregiver responses to the Childhood Eczema Questionnaire and a Proxy Report of a Clinician AD Diagnosis in children ≤2 years of age. OBJECTIVES: To evaluate the performance of Childhood Eczema Questionnaire and Proxy Report of a Clinician AD Diagnosis for the purpose of remote AD diagnosis. METHODS: Data from the CASCADE trial was used to calculate the sensitivity, specificity, positive predictive value, and negative predictive value of Childhood Eczema Questionnaire or Proxy Report of a Clinician AD Diagnosis at 12 and 24 months of age in 933 participants with data available for comparison against the reference standard of a clinician's chart diagnosis of AD, present in 26.2% (244/933 at 24 months) of the cohort. Clinicians who acted as the reference standard were primary care clinicians or pediatricians who received additional training in diagnosing AD. RESULTS: Proxy Report of a Clinician AD Diagnosis displayed sensitivity/specificity at 12 months (65.8%/91.3%) and 71.6%/93.2% at 24 months. It also displayed positive and negative predictive values of 72.5%/88.5% at 12 months and 88.3%/82.2% 24 months. The Childhood Eczema Questionnaire showed sensitivity, specificity, positive and negative predictive values of 46.3%/91.5%/64.5%/83.5% at 12 months and 50.9%/93.2%/84.1%/72.8% at 24 months. A lack of proxy-observed or reported itch or rash were the most common reasons for Childhood Eczema Questionnaire false negative cases. CONCLUSIONS: Proxy Report of a Clinician AD Diagnosis performed well for sensitivity, specificity, positive and negative predictive values, followed by Childhood Eczema Questionnaire. This U.S. study provides new evidence that supports Proxy Report of a Clinician AD Diagnosis as an adequate approach to identify eczema cases remotely for research in children <2 with access to healthcare.