The major examination of services for homeless families that has occurred since 1998 is the Center for Mental Health Services (CMHS) and Center forSubstance Abuse Treatment (CSAT) Homeless Families Program, a five-year, two-phased, multisite study initiated in 1999 to advance knowledge about the effectiveness of interventions for homeless mothers with psychiatric and/or substance abuse disorders who are caring for their dependent children(Rog, Buckner al., in press). It was specifically designed to be the first multisite evaluation of the effectiveness of innovative interventions, compared to services as usual or alternative interventions, in addressing the particular treatment and service needs of homeless families.
The focus of the target intervention in each site in the CMHS/CSAT Homeless Families Program was to be a time-limited (i.e., no more than a nine-month period of intensive services) intervention aimed at meeting the psychiatric, substance abuse, and/or trauma services needs of homeless women with children.The interventions were to be existing programs in the community, but could be enhanced. All target interventions were to be multifaceted, involving a combination of services focused on mental health treatment, substance abuse treatment, trauma recovery, securing and maintaining housing, parenting skills, household and money management, and goal setting.
Despite having a common set of core parameters, the target intervention models varied widely across the eight sites. Most sites involved some form of intensive case management, but combined that approach with other services in various settings. Three sites used more comprehensive service approaches, including (1) a multidimensional family assistance intervention in which families were provided with multisystemic therapy both in the shelter and in their residence (Henggeler et al., 1998), (2) multiple services (i.e., family-centered case management, home-based parent support, education and skills training, and child-focused interventions such as primary care) through a Comprehensive Family Health Practice within a community health center, and (3) a family therapeutic community in a residential substance abuse treatment program that was enhanced with trauma recovery and aftercare components.
Results from the CMHS/CSAT initiative are currently being analyzed and reported. Overall, the study did not find any effects of the target interventions on a range of outcomes for the homeless families compared to services as usual (e.g., Pearson et al., in press; Rog, Buckner et al., in press; Sacks et al., in press). However, for substance abuse and mental health outcomes, having more on-site services in these areas was associated with greater improvements for all families, and especially for families with clinical-level need for substance abuse and mental health services (Pearson et al., in press; Rog, Buckner et al., in press) Homeless mothers in programs that provided more on-site mental health services, such as having a psychiatrist or psychologist on site and having designated mental health providers who could provide an array of mental health services in the shelter or other setting where the families resided (e.g., residential treatment), experienced a greater decrease in their mental health symptoms than mothers who were in programs that had fewer on-site services and/or relied on referral services. Similarly, homeless mothers in programs with on-site substance abuse services reported less substance use over time than mothers in programs with fewer on-site services.
In addition, because of the multisite studys longitudinal design, it was possible to examine trajectories of change over time and to examine the role of other time-varying conditions on families outcomes. On most outcomes, families in both the target and comparison conditions on average had a positive rate of change. However, for each outcome, there was a substantial segment of families who started with mental health problems severe enough to warrant treatment and who did not show improvement over 15 months. Across outcomes, reports of ongoing trauma and network conflict were associated with less improvement, whereas employment was associated with more improvement (e.g., Pearson et al., in press; Rog, Buckner et al., in press; Sacks et al., in press). These findings suggest the need for not only understanding the history of problems families have as they enter shelter and other settings, but also the struggles they continue to experience that may be interfering with their ability to progress.