Predictors of Participation and Engagement in an Economic Support Intervention with Families at Risk for Child Maltreatment

Background and purpose: Child maltreatment prevention programs typically involve community-based, voluntary interventions either for families deemed to be at risk for child maltreatment or for families regardless of the presence of risk factors for child maltreatment.  However, an Achilles’ heel of such programming is whether they can successfully attract participation and engagement among families most in need of assistance, or whether they attract families likely to do well irrespective of their participation in prevention services.  This study focuses on the predictors of participation and engagement in a community-based, voluntary economic support intervention designed to prevent future child maltreatment among families reported to child protective services (CPS) in Milwaukee County, WI, but whose cases closed following an abuse and neglect investigation. 

Methods: The analysis sample involves 369 primary caregivers assigned to a treatment group following the administration of a baseline survey in 2016.  The survey collectedinformation on a number of known risk factors for child maltreatment, including depression, parenting distress, intimate partner violence, means-tested benefit receipt, and economic stress. Baseline controls included measures for race and ethnicity, respondent age, number of children, age of youngest child, marital/cohabiting status, education level, and current employment status.  Indicators of program contact with the caregiver (i.e., via staff outreach efforts), participation (i.e., completing an intake), engagement (i.e., progress toward one or more goals), and program completion (i.e., achieving one or more goals during the intervention window) were extracted from program records.

Results: Slightly different predictors mattered for each program outcome.  Having older children and a greater number of children predicted successful contact with caregivers by program staff.  Having older children, receiving SNAP benefits, and higher scores on a depression scale were positively associated with completing program intake, whereas higher scores on parental distress were associated with lower odds of program intake.  Conversely, higher parental distress was associated with increased odds of goal progress and program completion.  Economic stress was associated with higher odds of program completion, as were higher levels of self-efficacy at baseline. 

Conclusions and implications: Findings suggest that some family characteristics enhance contacts by program staff, such as having older and a greater number of children.  Such factors could indicate greater housing stability.  Some indicators of greater economic stress increase the odds of program engagement and completion, suggesting that the intervention is attracting the intended service population. Caregivers reporting higher levels of parenting distress were less likely to engage in services but more likely to complete the intervention, suggesting that some aspects of parenting may impede program delivery.  Caregivers with a greater sense of self-efficacy were also more likely to complete the intervention, pointing to personal characteristics that may matter for service persistence.  Findings from studies such as this help paint a more nuanced picture of the factors that play a role in prevention program participation and engagement, informing an understanding of whether service involvement is driven by need or selection.