OBJECTIVE: This study evaluated the impact of a community-based mobile crisis intervention program on the rate and timing of hospitalization. It also explored major consumer characteristics related to the likelihood of hospitalization. METHODS: A quasi-experimental design with an ex post matched control group was used. A community-based mobile crisis intervention cohort (N=1,696) was matched with a hospital-based intervention cohort (N=4,106) on seven variables: gender, race, age at the time of crisis service, primary diagnosis, recency of prior use of services, indication of substance abuse, and severe mental disability certification status. The matching process resulted in a treatment group and a comparison group, each consisting of 1,100 subjects. Differences in hospitalization rate and timing between the two groups were assessed with a Cox proportional hazards model. RESULTS: The community-based crisis intervention reduced the hospitalization rate by 8 percentage points. A consumer using a hospital-based intervention was 51 percent more likely than one using community-based mobile crisis services to be hospitalized within the 30 days after the crisis (p<.001). Treating a greater proportion of clients in the community rather than hospitalizing them did not increase the risk of subsequent hospitalization. Those most likely to be hospitalized were young, homeless, and experiencing acute problems; they were referred by psychiatric hospitals, the legal system, or other treatment facilities; they showed signs of substance abuse, had no income, and were severely mentally disabled. CONCLUSIONS: Results indicate that community-based mobile crisis services resulted in a lower rate of hospitalization than hospital-based interventions. Consumer characteristics were also associated with the risk of hospitalization.
OBJECTIVE: This study evaluated the impact of a community-based mobile crisis intervention program on the rate and timing of hospitalization. It also explored major consumer characteristics related to the likelihood of hospitalization. METHODS: A quasi-experimental design with an ex post matched control group was used. A community-based mobile crisis intervention cohort (N=1,696) was matched with a hospital-based intervention cohort (N=4,106) on seven variables: gender, race, age at the time of crisis service, primary diagnosis, recency of prior use of services, indication of substance abuse, and severe mental disability certification status. The matching process resulted in a treatment group and a comparison group, each consisting of 1,100 subjects. Differences in hospitalization rate and timing between the two groups were assessed with a Cox proportional hazards model. RESULTS: The community-based crisis intervention reduced the hospitalization rate by 8 percentage points. A consumer using a hospital-based intervention was 51 percent more likely than one using community-based mobile crisis services to be hospitalized within the 30 days after the crisis (p<.001). Treating a greater proportion of clients in the community rather than hospitalizing them did not increase the risk of subsequent hospitalization. Those most likely to be hospitalized were young, homeless, and experiencing acute problems; they were referred by psychiatric hospitals, the legal system, or other treatment facilities; they showed signs of substance abuse, had no income, and were severely mentally disabled. CONCLUSIONS: Results indicate that community-based mobile crisis services resulted in a lower rate of hospitalization than hospital-based interventions. Consumer characteristics were also associated with the risk of hospitalization.
Authors: Mary-Anne Cotton; Sonia Johnson; Jonathan Bindman; Andrew Sandor; Ian R White; Graham Thornicroft; Fiona Nolan; Stephen Pilling; John Hoult; Nigel McKenzie; Paul Bebbington Journal: BMC Psychiatry Date: 2007-10-02 Impact factor: 3.630