Abstract
OBJECTIVE: Ecological momentary assessment (EMA) is increasingly used to monitor depressive symptoms in clinical trials, but little is known about the comparability of its outcomes to those of clinical interviews and questionnaires. In our study, we administered EMA and questionnaires to measure change in depressive symptoms and repetitive negative thinking (RNT) in a clinical trial and investigated (a) the size of intervention effects associated with both techniques and (b) their validity in predicting clinical interview outcomes (i.e., global functioning). MATERIALS AND METHODS: Seventy-one depressed patients were randomly assigned to one of three psychological interventions. The EMA comprised a concise item set (four items per scale) and was administered three times per day during a 7-week intervention period. Conversely, questionnaires were assessed weekly (WQA), encompassing their full sets of items of depressive symptoms and RNT. RESULTS: While EMA excelled in detecting significant intervention effects, WQA demonstrated greater strength in predicting clinician ratings of global functioning. Additionally, we observed significant differences in time effects (slopes) between the two techniques. WQA scores decreased steeper over time and were more extreme, e.g., higher at baseline and lower postintervention, than EMA scores. CONCLUSIONS: Although clinical interviews, questionnaires, and EMA outcomes are related, they assess changes in depression differently. EMA may be more sensitive to intervention effects, but all three methods harbor potential bias, raising validity and reliability questions. Therefore, to enhance the validity and reliability of clinical trial assessments, we emphasize the importance of EMA approaches that combine subjective self-reports with objectively measured behavioral markers. This trial is registered with osf.io/9fuhn.