ABSTRACT
BACKGROUND: Atopic dermatitis (AD) is a major public health concern worldwide. Given its rising prevalence, there is a pressing need to study its epidemiology. To date, no validated, cost-effective, and patient self-administered screening instrument exists that can reliably distinguish AD from other skin conditions. Our aim is to develop an instrument that can validate a self-reported AD diagnosis quickly and with reasonable accuracy in a population at a distance. METHODS: A one-page, 13-question atopic dermatitis screening and evaluation questionnaire (ADSEQ) was generated after literature review and consultation with dermatologists. The questionnaire was administered to 121 patients with and without AD in a dermatology clinic. Sensitivity, specificity, positive predictive value, and negative predictive value were calculated for all questions. The questions with the highest sensitivities were selected for logistic regression analysis, and a receiver operating characteristics curve was plotted. RESULTS: Four scoring algorithms using different combinations of responses were then developed. A combination of Q7 (itching) and Qpic (the image questions combined) had the highest sensitivity (96.3%) and specificity (84%) for a diagnosis of AD. Cronbach's alpha coefficient was 0.71, indicating acceptable internal consistency. Other algorithms combining Qpic and questions about past diagnosis (Q1) or family history of atopy (Q5) also had sensitivities and specificities in the 80s or 90s. CONCLUSIONS: In this pilot study, we found that our questionnaire was a valid and reliable self-administered tool for distinguishing individuals with AD from individuals without AD.
