OBJECTIVE: Because previous studies of day treatment as an alternative to inpatient treatment had major disadvantages or methodological shortcomings, the authors conducted a randomized controlled trial to estimate and predict the extent to which day treatment is feasible for unselected patients referred for inpatient treatment. METHOD: Of 160 patients, 57 were randomly assigned to the control condition and 103 were assigned to the experimental condition. Control patients received standard clinical care. In the experimental condition, day treatment was attempted as soon as the patient's condition permitted. The average number of nights per week that experimental patients spent away from the hospital was compared to the average number of nights away for patients under standard care. RESULTS: Day treatment was satisfactory for 40% of the experimental patients but was completely infeasible for another 40%. The level of surveillance needed in the first week, physical illness, number of previous admissions, depressive symptoms, and treatment by qualified psychiatrists versus registrars were variables predictive of these differences. CONCLUSIONS: In this unselected group of patients, no absolute contraindications against day treatment were found. This suggests that the selection criteria applied in nearly all other controlled studies on the subject were unwarranted. The approach used in this study facilitated treatment in the least restrictive environment possible.Abstract Teaser