Assessing intermediate outcomes: Increasing protective factors and decreasing risk factors

BACKGROUND AND PURPOSE. The Family Connections (FC) intervention was designed with a prevention science theory of change suggesting that child maltreatment is prevented by first enhancing protective factors and reducing risk factors. The purpose of this study was to examine the outcome trajectories of families served by eight agencies replicating FC. It was hypothesized that FC families would show greater improvement in outcomes than non-FC families and that families that received the FC program for longer would show greater improvement over time. METHODS. Families were randomly assigned to treatment conditions which varied along three dimensions: provision of FC versus typical services, duration of FC services, and presence of a local enhancement to the FC program. A total of 802 families were recruited into the FC Replication studies with an average of 100 families per site. Families were assessed on all outcomes at intake, program completion, and six months following completion with a common set of standardized measures to assess child behavior, parenting stress and depressive symptoms, parenting attitudes, and need for support. A count of areas in which clinical judgment indicated the need for intervention was collected by caseworkers through the Family Assessment Form (FAF). Using an intent-to-treat approach, two sets of repeated measures linear mixed models were conducted, one for sites with no-FC control conditions to assess the impact of the FC program and the other set to assess the effect of longer receipt of FC services. Mixed model methodology was used to maximize the use of available data. To control for variability among the sites, Site was included in the models as a fixed effect along with interactions of Site with all other model terms. RESULTS. A consistent finding across all sites was improvement in outcomes from baseline to post-test. Significant variability in outcomes across sites was also found. Treatment group analyses indicated that FC caregivers reported a significant decrease in their need for support from baseline to post-test. Non-FC caregivers showed no change over this period. Program duration analyses indicated that receiving FC for longer periods was associated with decreased parenting stress and depressive symptoms and greater decreases in the number of FAF areas in need of clinical intervention. CONCLUSIONS and IMPLICATIONS. Results indicate that the FC program leads to enhanced protective factors and decreased risk factors. Participants will be engaged to discuss the implications of results for future social work practice with similar target populations. In particular, implications will be considered related to the findings that suggested that longer service delivery was associated with greater improvements in parenting stress, caregiver depressive symptoms, and fewer areas of family functioning in need of clinical intervention. These results were not moderated by the substantial variability among the service populations of the replication sites or the local adaptations to the standard FC program model. Discussion will also focus on the complexity of conducting multi-site evaluations, even when programs have proposed to replicate the same social work intervention using a core set of outcome measures.