Abstract
BACKGROUND AND AIMS: A better understanding of distinct subgroups in atopic dermatitis (AD) is warranted. The aim was to identify and determine characteristics of clusters based on anatomical location of AD. METHODS: In this 8-week, observational, decentralized study, patients with AD completed a baseline questionnaire about anatomical location and severity of AD, and a principal component analysis (PCA) was applied to identify clusters. The Patient-Oriented Eczema Measure (POEM) was completed weekly and photographs of affected body areas were captured by the participants' own smartphones. From the weekly photographs, the AD severity was evaluated using the intensity part of the SCORing Atopic Dermatitis. RESULTS: Fifty-five participants were recruited, of which 53 completed the baseline questionnaire with a mean POEM of 14.5 (SD: 5.6). The PCA analysis revealed three clusters, with AD predominantly on the shins, knees, and genitals (Cluster 1), with involvement of the upper body (Cluster 2), and with AD on the hands and feet (Cluster 3). Cluster 1 had a lower mean POEM score (11.12, SD: 5.3) compared with Clusters 2 (12.64, SD: 4.5) and 3 (15.98, SD: 4.7), respectively (p = 0.007). Further, Cluster 1 had the highest age of AD onset (mean 9.5 vs. 2.5 and 4.7 years, p = 0.02) and the lowest proportion of asthma/allergy (47% vs. 82% and 90%, p = 0.01). CONCLUSION: Three clusters of patients with AD based on affected body areas were identified. The cluster with involvement of legs and genitals was characterized by the oldest age of AD onset and the lowest prevalence of asthma/allergy.