Friday, 14 January 2005 - 4:00 PM

This presentation is part of: Evaluating Child Welfare Intensive Services: Results from California's Title IV-E Waiver Child Welfare Demonstration Project

The Impact of Family Group Decision-Making on Child Welfare Outcomes

Purpose: Though family group decision-making (FGDM) has gained popularity in recent years, few studies have examined this intervention’s effectiveness in changing child welfare outcomes with the use of random assignment. This presentation will describe the results of an evaluation of family group decision-making (FGDM) programs administered under the Title IV-E Waiver Demonstration in Fresno and Riverside Counties in California.

Method: Children in the child welfare system in both counties were randomly assigned to participate in an FGDM intervention (Fresno n=39; Riverside n=31) or to receive traditional services (Fresno n=21; Riverside n=19) and then compared on child welfare outcomes. In Fresno, families in voluntary family maintenance participated in a one-time family conference, while in Riverside, families with children in out-of-home placement received an experimental intervention that consisted of several family conferences over a period from enrollment to case closure. The counties differed in their model of the intervention, with Fresno using a blended approach to FGDM and Riverside using the Family Unity Model. Since the counties utilized different models and targeted different populations, the data provides an important look at FGDM effectiveness in different contexts. Data for analysis was extracted from the California Children’s Services Archive, used by county child welfare workers to manage information related to children in the child welfare system. Bivariate and multivariate analysis, including linear and logistic regression, were used to examine group differences in child welfare outcomes related to child safety, placement stability, and permanence.

Results: Results in both counties indicated few group differences in child welfare outcomes around safety, placement stability, or permanence. Specifically, bivariate analysis of substantiated rates of maltreatment following enrollment showed no significant differences between groups in Riverside. In Fresno County, bivariate analysis showed the experimental group to have higher rates of maltreatment (p= .08); however, this should be interpreted cautiously, since there were few cases of maltreatment (experimental n=6; control n=0). Other measures of child safety, including removal from caretaker and declaration of court ordered dependency showed no group differences. Placement stability and permanency outcomes were only explored in Riverside County due to the target population. The number of placement moves, the likelihood of case closure, and time to permanency were not significantly different for the two groups. Multivariate analysis suggested that number of placement moves and case closure were related to the time a child was in the study (placement moves p= .005; case closure p=.0002), but not their group assignment.

Implications: These results suggest that child welfare outcomes may not be improved for youth receiving FGDM. However, the small sample size may have limited the ability to detect group differences or child welfare outcomes may not be the appropriate gauge of program effectiveness. This study provides an important step in increasing the knowledge base around this intervention, but suggests the need for more research, larger sample sizes, and additional measures to examine effectiveness. In a child welfare context where counties are being asked to choose between a myriad of new, interventions it becomes increasingly important to evaluate their effectiveness.

See more of Evaluating Child Welfare Intensive Services: Results from California's Title IV-E Waiver Child Welfare Demonstration Project
See more of Symposium

See more of Celebrating a Decade of SSWR (January 13 - 16, 2005)