Effect of Community Treatment Orders on Mental Health Service Usage, Emergency Visits, and Violence: a Systematic Review and Meta-Analysis.

Introduction: Community treatment orders (CTOs) enable patients to actively engage in mental health services while being supervised in the community outside the hospital setting. However, the efficacy of CTOs remains controversial in terms of mental health services usage or service contacts, emergency visits, and violence.

Methods: The databases PsychINFO, Embase, and Medline were searched on 11 March 2022 by 2 independent reviewers through the Covidence website (www.covidence.org). Randomised or non-randomised case-control studies and pre-post studies were included if they examine the effect of CTOs on service contacts, emergency visits, and violence in individuals with mental illnesses by comparing with control groups or pre-CTO conditions. Conflicts were resolved by consultation of the third independent reviewer.

Results: Sixteen studies provided sufficient data in the target outcome measures and were included in analysis. Variability in the risk of bias was high among studies. Meta-analyses were conducted separately for case-control studies and pre-post studies. For service contacts, a total of 11 studies with 66,192 patients reported changes in the number of service contacts under CTOs. In 6 case-control studies, a small non-significant increase in service contacts was observed in those under CTOs (Hedge's g = 0.241, z = 1.535, p = 0.13). In 5 pre-post studies, a large and significant increase in service contacts was noted after CTOs (Hedge's g = 0.830, z = 5.056, p < 0.001). For emergency visits, a total of 6 studies with 930 patients reported changes in the number of emergency visits under CTOs. In 2 case-control studies, a small non-significant increase in emergency visits was noted in those under CTOs (Hedge's g = -0.196, z = -1.567, p = 0.117). In 4 pre-post studies, a small significant decrease in emergency visits was noted after CTOs (Hedge's g = 0.553, z = 3.101, p = 0.002). For violence, a total of 2 pre-post studies reported a moderate significant reduction in violence after CTOs (Hedge's g = 0.482, z = 5.173, p < 0.001).

Conclusion: Case-control studies showed inconclusive evidence, but pre-post studies showed significant effects of CTOs in promoting service contacts and reducing emergency visits and violence. Future studies on cost-effectiveness analysis and qualitative analysis for specific populations with various cultures and backgrounds are warranted.