Background. There has been a widespread development of community multi-disciplinary teams aimed to deliver coordinated comprehensive mental health care, yet there is little published evidence on the quality of care and economics of providing such care for people with severe mental illness.

Method. This is a clustered randomized controlled economic comparison of the quality of care for patients with chronic schizophrenia by a multi-disciplinary community team with close links with primary care, and a traditional psychiatric service in a district general hospital psychiatric unit.

Results. Two years after it was established, patients with access to the community team had more of their needs met; they had fewer unmet needs; and they were more satisfied with the care they had received. They had more service contacts and received more interventions. The community team resulted in savings in the use of some hospital resources but these were not sufficient to offset the cost of the new service. The community team successfully directed care to patients with more needs, whereas no such relationship was evident for the traditional hospital-based service. Four years after the team was established, it met a greater proportion of needs for underactivity, daily living skills, use of public amenities and managing finances.

Conclusions. Better quality care was provided at 2 and 4 years after its establishment by the multi-disciplinary community service than the traditional hospital-based service. Resources were targeted more efficiently by the community service.