Session: Outcomes from a Supportive Housing and Child Welfare Experiment (Society for Social Work and Research 23rd Annual Conference - Ending Gender Based, Family and Community Violence)

In 2012, the Children's Bureau funded five sites nationally to design, implement, and evaluate supportive, permanent housing models for homeless, child welfare involved families. The goal of the five-year demonstration project was to test a “housing first” approach to safety reducing the need for foster care, increasing housing stability, and improving child and family well-being. This session presents a selection of impact findings from two local site evaluations and the cross-site evaluation.

The first presenter will briefly describe the theory and aims of the intervention, and the target population. Each site developed its own model under the federal directive, although all sought to implement a “housing first” approach that included permanent housing coupled with supportive services. The models followed from a developing literature on typologies characterizing some of the highest-need child-welfare involved families as those with are not only homeless but who also struggle with comorbidities that impede a family's ability to remain safely intact. All five sites used a randomized controlled trial design.

The first presenter will also discuss findings from the cross-site evaluation. Housing impact findings relied on family surveys at baseline and one year later, showing that the program group reported more stable and higher quality housing than the service-as-usual group, although it was not necessarily in better neighborhoods. Child welfare outcomes included probability of removal (for intact families), probability and speed of reunification (for families with a child in foster care), probability and speed of case closure, and probability of repeat maltreatment. Removal and reunification outcomes were promising but uneven across the five sites.

The second paper involves child welfare reunification outcomes in the San Francisco site. The intervention there sought, in part, to promote reunification for children who had been removed. The hypothesis was that coupling stable housing with supportive services would better enable families to complete child welfare case plans and would expedite that process. After following families from enrollment through 2017, initial analyses suggest that children in treatment families were somewhat more likely to reunify compared to control group children, although both groups took a similarly long time to do so.

The third paper deals with child well-being outcomes in the San Francisco site. If families are stably housed, the theory goes, children should exhibit fewer behavioral problems and better daily functioning. The Child and Adolescent Needs and Strengths (CANS) assessment measured baseline functioning and changes in functioning over time. The treatment group showed small but greater improvements.

The fourth paper examines the intersection of child welfare and housing outcomes among supportive housing participants in the Cedar Rapids, Iowa site. Using a typology of child welfare and housing outcomes, the study examined initial and intervening factors that differentiate families that remained intact/reunified as well as stably housed, those that became stably housed though not unified, and families that were neither unified nor stably housed. Study findings offer implications for improving supportive housing services with child welfare-involved families.