How many longstay patients could be equally served by less sheltered facilities, and what are the main reasons for staff members to judge patients suitable for such alternative care? These were two questions in a Dutch study (1984-1985) of five mental hospitals. The attending psychiatrists in these hospitals and members of the nursing staff, were asked to give an independent judgement for each of a random sample of 381 longstay patients. Thirty percent of the longstay patients were judged eligible for alternative care. They were generally younger, less frequently diagnosed as schizophrenic, better integrated both inside and outside the mental hospital, more likely to be resident of wards with high turnover rates, were less aggressive, had better social skills, and had shorter periods of inpatient stay. Compared with the rest of the patients no statistically significant differences were found with regard to suicidal behaviour, organic brain syndrome and formal admission status.