An experimental design was used to assess the effectiveness of assertive community treatment (ACT) compared to traditional mental health services for individuals with severe mental illness (SMI) in four rural communities. A total of 160 clients (40 per site) were randomly assigned. Two-year findings were mildly encouraging. Experimental differences on staff rating of quality of life, level of functioning, and symptoms favored ACT clients. There were no experimental differences in hospital use. ACT clients exhibited less residential stability than control clients. We discuss problems implementing the model in rural communities and suggest adapting it for rural areas.