Child Abuse Prevention for System Versus Non-System Families and Youth through Intensive in-Home Intervention

Introduction: Child abuse prevention has traditionally focused on delivering individualized home-based intervention to families who have been referred to the program because of substantiated child maltreatment to prevent new cases of abuse. There is promise in implementing such intervention more broadly with a goal of preventing high-risk families and youth from entering the system. However studies have not yet evaluated the effectiveness of individualized home-based intervention for child abuse prevention in system and non-system families, particularly in applied service settings. This study tested differences in intake-discharge change between system versus non-system families receiving an in-home intervention offered by a national service provider. Because the intervention is individualized and families were not expected to change on all outcomes, a technique for testing group differences on maximum individualized change was implemented.

Methods: Data for this study were obtained from clinical records during routine administration of the home-based family intervention, and included families that were referred to the program either by Child Protective Services (N = 98) or recruited in schools (N = 62) in 2015. Family functioning was assessed in five domains at intake and discharge: environment, family interactions, parental capabilities, family safety and child wellbeing. Paired-samples t-tests, repeated measures ANOVAs, and Maximum Individualized Change Analysis (MICA) was used to examine 1) within-group differences from intake to discharge; 2) between-group differences in gains in each domain; and 3) between-group differences in the MICA score, which captures each family’s domain of greatest change.

Results: Families recruited through schools made the most improvements in domains of child wellbeing (t = -9.75, p < .001; d = 1.39) and family interactions (t = -5.29, p < .001, d = .66), while Child Welfare families had the most improvement in child wellbeing (t = -10, p < .001; d = .84) and parental capabilities (t = -9.42, p < .001; d = .83). Child Welfare families also made significantly more improvements, compared to the school families in the domain of parental capabilities F (1, 151) = 4.163, p = .043. Gains on the MICA score did not differ between the two groups F (1, 158) = 1.42, p = .235.

Conclusions: The two types of families had different domains of change, suggesting the value of individualized intervention. Both groups displayed significant improvements on their maximum individualized change score, perhaps reflecting that the intervention is effective in both contexts. Limitations and implications will be discussed.