Background The Rotterdam Eczema Study was an observational cohort study with an embedded pragmatic randomised controlled open-label trial. It was conducted in children with atopic dermatitis (AD) in the Dutch primary care system. The objective of the trial was to determine whether a potent topical corticosteroid (TCS) is more effective than a low-potency TCS. Objective We are aiming to communicate transparently about the poor recruitment for the trial part and to explore the reasons why recruitment was weak. Design We used a survey to find out what patients in the cohort did when they experienced a flare-up. Methods Descriptive statistics were used to present the baseline characteristics of participants in the trial and the results of the survey. Results In total, 367 patients were included in the cohort. Of these, 32 were randomly assigned to a trial treatment; they had a median age of 4.0 years (IQR 2.0-9.8). A total of 69 of the 86 children (80.2%) who could participate in the survey responded. 39 (56.5%) suffered a flare-up during the follow-up (making them potentially eligible for inclusion in the trial). 26 out of 39 (66.7%) increased their use of an emollient and/or TCS themselves. Only 12 of the 39 (30.7%) contacted their general practitioner (GP) as instructed in the study protocol, but 8 out of these 12 did not meet the inclusion criteria for the trial. Conclusion The main reason why cohort participants did not take part in the trial was that they did not contact their GPs when they experienced an AD flare-up. Furthermore, the majority of patients who contacted their GPs did not match the inclusion criteria of the trial. We expect that the lessons learnt from this study will be useful when developing future studies of children with AD in primary care.