Randomised controlled trial of day patient versus inpatient psychiatric treatment

Abstract

Objective: To assess the proportion of acutely ill psychiatric patients who can be treated in a day hospital and compare the outcome of day patient and inpatient treatment.

Design: Prospective randomised controlled trial of day patient versus inpatient treatment after exclusion of patients precluded by severity of illness or other factors from being treated as day patients. All three groups assessed at three and 12 months.

Setting: Teaching hospital serving small socially deprived inner city area. Day hospital designed to take acute admissions because of few beds.

Patients: 175 Patients were considered, of whom 73 could not be allocated. Of the remaining 102 patients, 51 were allocated to each treatment setting but only 89 became established in treatment--namely, 41 day patients and 48 inpatients. 73 Of these 89 patients were reassessed at three months and 70 at one year.

Interventions: Standard day patient and inpatient treatment.

Main outcome measures: Discharge from hospital and return to previous level of social functioning; reduction of psychiatric symptoms, abnormal behaviour, and burden on relatives.

Results: 33 Of 48 inpatients were discharged at three months compared with 17 of 41 day patients. But at one year 9 of 48 inpatients and three of 41 day patients were in hospital. 18 Of 35 day patients and 16 of 39 inpatients were at their previous level of social functioning at one year. The only significant difference at three months was a greater improvement in social role performance in the inpatients. At one year there was no significant difference between day patients and inpatients in present state examination summary scores and social role performance, burden, or behaviour.

Conclusions: Roughly 40% of all acutely ill patients presenting for admission to a psychiatric unit may be treated satisfactorily in a well staffed day hospital. The outcome of treatment is similar to that of inpatient care but might possibly reduce readmissions. The hospital costs seem to be similar but further research is required to assess the costs in terms of extra demands on relatives, general practitioners, and other community resources.