Predictors of Using Community Mental Health Services After the Mobile Crisis Team Intervention

Background and Purpose:  As a community-based psychiatric crisis service, a Mobile Crisis Team (MCT) provides a variety of crisis services (e.g., crisis assessment / intervention, home / site visiting, temporary shelter, referral services, etc.) to people with a psychiatric crisis in a large urban area.  Previous studies demonstrated that MCT intervention was effective in facilitating clients' use of community mental health services (Dyches, et al., 2002; Guo, et al., 2001).  However, few studies have simultaneously focused on the process itself and the post-phase of psychiatric crisis intervention (e.g., MCT intervention) in spite of the importance of Continuum of Care (CoC); MCT seeks to facilitate immediate access to community mental health services for clients.  This study aims to identify the predictors of clients' use of community mental health services after MCT intervention.

Methods:  This study merges four administrative datasets (Multi-Agency Community Services Information System, Patient Care Systems, Criminal Justice Data of Cuyahoga County, and Intake Data from MCT Authority).  Based on six criteria (service year, age, geographic area, diagnosis, referral, and legal issue), this study selects 1,771 adults in Cuyahoga County (Ohio) who received MCT intervention in the fiscal year 2008.  Because a crisis period varies by clients' severity and referral procedures, this study applies a 7-day interval to specify the last date of MCT intervention; the successive crisis services (MCT interventions, hospitalizations, emergency room services, and crisis stabilization units) provided within 7 days are regarded as one episode.  Measuring the length between the last date of MCT intervention and the first date of using a community mental health service, this study identifies the factors that affect the timing and occurrence of using community mental health services within 30 days after MCT intervention.  This study uses the Cox proportional hazards regression.

Results:  Overall, 44.2% of the sample is exposed to the event, which is the use of community mental health services within 30 days after MCT intervention.  Cox proportional hazards model identifies significant predictors of the event: (1) clients who experienced community mental health services (within 12 months) before or during the period of MCT intervention have a higher likelihood of the event (p<.001); (2) clients referred to hospital settings (hospitalizations and emergency rooms) have a higher likelihood of the event (p<.01); (3) clients with  schizophrenic disorders or other non-organic psychoses have a higher likelihood of the event (p<.01); and (4) clients with substance abuse issues have a lower likelihood of the event (p<.01).

Conclusions and Implications:  The findings highlight the vital roles of MCT intervention such as linking resources and referral services.  MCT intervention should not be seen as a stand-alone program, but as embedded in a CoC.  This study shows when MCT is integrated with other parts of community mental health services the CoC is more likely to be maintained.  As this study simultaneously observes the process and post-phase of psychiatric crisis intervention, its findings not only assist in improving social work practice for people with psychiatric crisis but also support social policy and programs that strengthen the CoC.