Monitoring and Managing Outcomes in Residential Treatment: Practice-Based Evidence in Search of Evidence-Based Practice

Abstract

In 1995, the Illinois Department of Children and Family Services (IDCFS) spent about $350 million on longterm residential treatment and psychiatric hospitalization, accounting for more than three fourths of the total funds spent on mental health services for the more than 50,000 children in state custody. At that time, no community-based treatment options existed for these children other than outpatient clinic visits (Lyons et al., 1998). The problem was especially acute in poor urban neighborhoods, where even outpatient clinic services were less available and used (Lyons et al., 1998). A more balanced system of care was lacking, leading to disproportionate use of residential treatment, a level of care both costly and incongruent with the leastrestrictive treatment approach (Stroul, 1993). In response to the perceived problems with residential care, the IDCFS established a project to determine whether a subset of children in residential care could be effectively served in the community. Lyons and colleagues (1998) conducted a clinical needs assessment of 333 youths, all in state custody, randomly selected from 17 randomly selected residential treatment settings of various sizes and types (i.e., residential placement programs, group homes, supervised independent living facilities, and residential treatment programs for sexual offenders). The Childhood Severity of Psychiatric Illness (CSPI; Lyons, 1998), a standardized assessment tool comprising 27 items, each with four anchored levels, was used to assess children’s mental health needs across several domains, including psychiatric symptoms, risk behaviors, and functioning. Results indicated that although the majority of children in the sample had levels of mental health need consistent with residential services, 41.4% of the girls and 29.7% of the boys were judged, based on their level of risk, to be eligible for community-based treatment (Lyons et al., 1998). The proportion of children receiving residential treatment yet not appearing to require it varied according to the size and type of residential facility. In both large and small residential facilities, the proportion of low-risk children (i.e., no current suicidal ideation or history of suicide attempt, danger to others, runaway risk, delinquent behavior, or sexual aggression) was higher than in mid-sized residential facilities; in independent living facilities, the majority of children were at low risk, and in both sexual offender programs and mid-sized facilities, far more children were at high risk (Lyons et al., 1998). On the basis of these data, the IDCFS implemented a process of residential placement review and a system of residential facility gatekeepers to ensure more appropriate placement and to monitor progress in care (Illinois Department of Children and Family Services, 2000). This is an example of evidence-based practice: the finding that a proportion of children in residential treatment were at sufficiently low risk that they likely could be served in the community incited systematic changes regarding residential treatment. Between 1995 and 2003, the number of children in residential care in Illinois decreased by almost 60% from 4,015 children in 1995 to 1,683 in 2003 (Budde et al., 2004). As a result, far more Accepted September 1, 2005. The preparation of this article was funded in part by the Illinois Department of Children and Family Services through a contract with Northwestern University. The authors thank Julie Eisengart and Catherine Francis for their assistance with this project. Correspondence to Dr. John S. Lyons, Mental Health Services and Policy Program, Northwestern University, Feinberg School of Medicine, Abbott Hall, Suite 1205, 710 North Lake Shore Drive, Chicago, IL 60611; e-mail: jsl329@northwestern.edu. The authors are with the Mental Health Services and Policy Program, Northwestern University, Feinberg School of Medicine, Chicago. 0890-8567/06/4502–0247 2006 by the American Academy of Child and Adolescent Psychiatry. DOI: 10.1097/01.chi.0000190468.78200.4e