BACKGROUND: Atopic eczema (AE) is a chronic disease with flares and remissions. Long-term control of AE flares has been identified as a core outcome domain for AE trials. However, it is unclear how flares should be defined and measured. OBJECTIVE: To validate two concepts of AE flares based on daily reports of topical medication use: (i) escalation of treatment and (ii) days of topical anti-inflammatory medication use (topical corticosteroids and/or calcineurin inhibitors). METHODS: Data from two published AE studies (studies A (n=336) and B (n=60)) were analysed separately. Validity and feasibility of flare definitions were assessed using daily global bother (scale 0 to 10) as the reference standard. Intra-class correlations were reported for continuous variables, and odds ratios and area under the receiver operator characteristic (ROC) curve for binary outcome measures. RESULTS: Good agreement was found between both AE flare definitions and change in global bother: area under the ROC curve for treatment escalation of 0.70 and 0.73 in studies A and B respectively, and area under the ROC curve of 0.69 for topical anti-inflammatory medication use (Study A only). Significant positive relationships were found between validated severity scales (POEM, SASSAD, TIS) and the duration of AE flares occurring in the previous week - POEM and SASSAD rose by half a point for each unit increase in number of days in flare. Smaller increases were observed on the TIS scale. Completeness of daily diaries was 95% for Study A and 60% for Study B over 16 weeks). CONCLUSION: Both definitions were good proxy indicators of AE flares. We found no evidence that 'escalation of treatment' was a better measure of AE flares than 'use of topical anti-inflammatory medications'. Capturing disease flares in AE trials through daily recording of medication use is feasible and appears to be a good indicator of long-term control. TRIAL REGISTRATION: Current Controlled Trials ISRCTN71423189 (Study A).