Holden, E.W., Rousseau O’Connell, S., Liao, Q., Krivelyova, A., Connor, T., Blau, G., & Long, D. (2008). Outcomes of a randomized trial of continuum of care services for children in a child welfare system. Child Welfare, 86, 89-114.
Restrictive residential treatment programs for children with mental health problems have been found to be both more expensive and linked to poorer outcomes than those in less restrictive community-based systems of care. Non-experimental studies indicate that community-based continuum of care may successfully reduce residential placements. Although they might be more expensive, they appear to offset juvenile justice and child welfare costs. Methodological challenges in past work have made it difficult to draw firm conclusions.
This study presents the results of a rigorous approach: a randomized controlled trial in which community agencies were given financial incentives to see if outcomes for children in an integrated continuum of care could be improved, and their time in residential programs reduced, compared to the usual state services.
Children involved with child welfare services with at least moderate mental health (see methodology) were randomized in either the usual state child welfare services or a system of integrated community-based continuum of care services. A greater percentage of children in the continuum of care group received crisis stabilization, family therapy, family preservation, family support services, behavioral aid, respite care, transportation services and flexible funds. The usual state services provided more residential treatment, inpatient hospital stays and medication monitoring. Children in the continuum of care group were more likely to stay at home, and spend more days at home, than children receiving state services. Both groups improved significantly in well-being over time. However, children at home at 12 months showed the most improvement and children in the continua of care group were more likely to be at home (38% versus 14%). This suggests that systems of care better utilize least restrictive treatment options to maximize children’s outcomes. The authors also found that the continua of care services cost less than services as usual.
This study was conducted in Connecticut with 155 children (7-15 year olds). Interviews (at initiation, 6 and 12 months) were conducted with children and their caregivers. Financial data for comparison were obtained from the management information system, but some costs were missing. To be included, children had to have at least “moderate mental health” (263 children were excluded because of poor daily living skills, relationships, self-care skills, school behaviour and participation; suicide attempts; assaults; firesetting; drug abuse; IQ < 65; or medical conditions requiring specialized medication), 10 children and their caregivers refused to participate and 2 children dropped out of the study. The sample mean age was 12.1 years and 53% were male. Approximately half of the sample was White, while Black and Hispanic participants each made up about one quarter of the sample. Primary problems for participants were most commonly aggression, self-injury, depression, and included diagnoses of mood disorders, ODD, PTSD and ADHD. Just over half of participants were in state custody and most had utilized various mental health services in the previous 12 months. After randomization the groups did not differ from one another on any observed variables.
This study does not generalize to children with more severe mental health.