BackgroundDepression can be treated in an outpatient, inpatient or day hospital setting. In the German health care system, episodes of inpatient or day hospital treatment are common, but there is a lack of studies evaluating effectiveness in routine care and subgroups of patients with a good or insufficient treatment response. Our study aims at identifying prognostic and prescriptive outcome predictors as well as comparative effectiveness in psychosomatic inpatient and day hospital treatment in depression.Methods/DesignIn a naturalistic study, 300 consecutive inpatient and 300 day hospital treatment episodes in seven psychosomatic hospitals in Germany will be included. Patients are assessed at four time points of measurement (admission, discharge, 3-months follow-up, 12-months follow-up) including a broad range of variables (self-report and expert ratings). First, the whole sample will be analysed to identify prognostic and prescriptive predictors of outcome (primary outcome criterion: Quick Inventory of Depressive Symptoms QIDS-total score, expert rating). Secondly, for a comparison of inpatient and day hospital treatment, samples will be matched according to known predictors of outcome.DiscussionNaturalistic studies with good external validity are needed to assess treatment outcome in depression in routine care and to identify subgroups of patients with different therapeutic needs.Trial registrationCurrent Controlled Trials ISRCTN20317064