Prevention-Based Intervention with Foster Families: Proximal Outcomes from and Experimental Trial of the Connecting Study

Background & Purpose
While most adolescents are susceptible to engaging in risky behaviors, teenagers in foster care are even more vulnerable. Caregivers need tools to meet the challenges of providing environments that buffer teens’ risks. Yet few cost-effective, evidence-based prevention programs exist to support these families. We present proximal outcomes on teens in foster care from a randomized controlled trial of the Connecting program, a self-administered, family-based, prevention-focused intervention based on the Social Development Model. Connecting was adapted for use within the child welfare system from the Staying Connected with Your Teen (SCT) program, which has been demonstrated to prevent substance use, risky sexual behavior, and violence 2 years following the intervention and prevent substance use into young adulthood

Method
Data & Sample:
Two-hundred-twenty foster teen (age 11-15) and caregiver dyads were recruited from across Washington State and randomized to either receive the Connecting program (110) or into a no intervention control condition (110). Baseline equivalence was strong, showing only 1 significant difference (more Hispanic youth in the program condition). Average age of the foster youth was 12.3 at baseline; gender was balanced (54% female); youth represented diverse race/ethnicity (30% Hispanic, 12% Native, 15% Black). Families in the intervention condition received the Connecting materials (videos and handbook with activities and activity checklist) and were contacted weekly by a family consultant for about 12-14 weeks. Baseline, and posttest, assessments were conducted to determine whether Connecting has short-term effects on improving proximal family outcomes. These are expected to lead to longer term behavioral outcomes at 1 and 2 year follow-up.

Analysis:
Intent-to-treat analyses were conducted for teen reported proximal family scales related to the intervention. Intervention assignment effects on each outcome were analyzed separately using appropriate regression techniques (logistic, ordered logistic, or OLS) controlling for baseline levels of the outcome as well as gender, time in current placement, ethnicity, and region within the state.

Results
Eighty-nine percent of youth completed the posttest survey. Of the 110 families assigned to the program condition 86 (78%) completed some of the program. Fifty-eight (52%) complete 60% or more of the 94 program tasks. Three of ten post-test analyses revealed that foster youth in the program condition (n=93) were significantly more likely than controls (n=104) to be have talked with their caregivers about substance use and other risky behaviors (d=.43, p=.007), to report being involved in making family rules (OR=2.38, p=.01), and to report high levels of caregiver recognition for positive behavior (OR=3.02, p=.03). Less pronounced improvements were observed in problem solving, refusal skills, and pro-bonding communication with caregivers. Scales showing no effect included family conflict, inconsistent discipline, monitoring and antisocial norms about substance use and violence.

Discussion
Overall the foster teens report positive trends in strengthening their involvements with caregivers (communication, rulemaking, positive rewards). This is encouraging as these may be precursors to stronger future connections. The findings will be discussed in light of the need to provide resources to foster caregivers during this important developmental period.