BACKGROUND: Numerous inclusion criteria and baseline severity assessments are used in clinical trials of atopic dermatitis (AD), which may limit comparison of results. OBJECTIVE: We sought to characterize the inclusion criteria and baseline severity assessments used in randomized controlled trials (RCT) of AD internationally. METHODS: We performed a systematic review of RCT with a pharmacological intervention from 2007-2016. Cochrane Library, EMBASE, GREAT, LILACS, MEDLINE and Scopus were searched. Two authors independently performed study selection and data extraction. RESULTS: Overall, 212 RCT met inclusion/exclusion criteria. Target population and inclusion criteria based on AD severity were not documented in 78 (36.8%) and 25 (18.7%) studies, respectively. Thirty and 58 severity assessments were used for inclusion criteria and baseline severity, respectively, with only 60.3% concordance between their use. Global assessments were most frequently used for both inclusion criteria and baseline severity assessment in North America (39.5% and 32.1%), while SCORing AD (SCORAD) or objective SCORAD index was most frequently used in Europe (23.5% and 23.0%) and Asia (34.2% and 43.5%). Minimum and maximum thresholds of severity assessments were inconsistently used between studies for inclusion criteria, even within similar target populations. SCORAD, global assessments and body surface area were most frequently used for both inclusion criteria and baseline severity assessment. IGA was particularly used in trials of topical agents. CONCLUSIONS: There was considerable variability and poor documentation of inclusion criteria and baseline severity assessments in RCT for AD. These differences may limit interpretation of a study and comparison of results between studies. This article is protected by copyright. All rights reserved.