Picture-based acne lesion counts: A validation study to assess accuracy and reliability of acne lesion counts via photography.

Abstract

Acne lesion counting (ALC) is widely used to evaluate efficacy of new acne treatments. Although such evaluations are precise and highly discriminative, if assessed live, it can be a time-consuming and intrusive measurement. Photographic assessment is a viable mode for ALCs and for training and/or qualifying evaluators. The purpose of this study was to validate photographic methods for performing ALCs and to provide an objective measurement tool to train and/or qualify lesion count evaluators for deployment in both small- and large-scale studies. Assess accuracy and reliability of acne lesion counting via photographic methods in 8 subjects aged 16 to 40 years, with Fitzpatrick Skin Types I to VI. Frontal and 45°angle images taken at the single in-clinic visit. Each subject underwent 3 counts per 4 evaluators: 1 set of live counts and 2 sets of photo counts. Intra-evaluator and inter-evaluator reliability measures for photo counts were evaluated using calculations of intraclass correlation coefficients (ICCs). Eight subjects (2 males and 6 females) age between 16 and 40 (min 16 years, max 25 years) diagnosed with facial acne vulgaris participated in the study. Fitzpatrick Skin Types ranged from III to V (three Type-III, four Type-IV, and one Type-V). ICC values for intra-evaluator reliability were found to be >0.95 for each evaluator, and ICC for inter-evaluator reliability was found to be 0.98. Photographic lesion count methodology is a reliable and accurate tool for objective measurement of ALCs and additionally, for training and/or qualifying evaluators.