Comparison of community based service with hospital based service for people with acute, severe psychiatric illness.

OBJECTIVE: To compare the burden on relatives and outcome of people treated for severe acute psychiatric illness by a community service and a traditional hospital based service. DESIGN: Follow up of patients aged 16-65 who required admission to hospital or home treatment for psychiatric illness during January 1990 to February 1991. SETTING: Two Birmingham electoral wards, Sparkbrook and Small Heath; Sparkbrook has a community based service and Small Heath a traditional hospital based service. SUBJECTS: 69 patients from Sparkbrook and 55 from Small Health. MAIN OUTCOME MEASURES: Scores on present state examination, social behaviour assessment schedule, and general health questionnaire. RESULTS: 24 (35%) of Sparkbrook patients received some treatment in hospital during the initial episodes. Relatives of Sparkbrook patients were less distressed by their burden at the initial assessment than relatives of Small Health patients (mean score 0.11 v 0.29, p < 0.01). Relatives were also more satisfied with the support they received and the treatment received by patients. More patients from Sparkbrook than Small Health were in contact with a psychiatrist (81% (95% confidence interval 71% to 91%) v 62% (44% to 68%)) and community nurse (56% (44% to 68%) v 14% (13% to 24%)) one year after the initial episode. Sparkbrook patients spent significantly fewer days in hospital during the initial episode (8 days v 59 days) and the first year (20.6 v 67.9 days). CONCLUSION: The community based service is as effective as the hospital based service and is preferred by relatives. It is more effective in keeping people in long term contact with psychiatrists.