Parent and Child Risk Profiles As Predictors of Response to a Conduct Problems Preventive Intervention

Introduction: Externalizing problems in youth (e.g., aggression) represent a major public health concern with substantial costs to individuals, their families, and larger society. Current evidence-based prevention programming targeting externalizing problems demonstrates modest overall effect sizes and are largely ineffective for a sizable proportion of youth who participate (McCart, Priester, Davies, & Azen, 2006). However, our understanding of the youth and family characteristics associated with response to specific programming is quite limited.

Methods: A sample of 240 Kindergarten youth were identified due to displaying elevated levels of school-based aggression. Youth were randomized to participate in either the Early Risers conduct problems preventive intervention (August et al., 2001) or a control condition. Early Risers consists of parent- and youth-focused skills training as well as individualized case management delivered over a three-year period. The current study relied upon 9 parent- and teacher-reported measures evaluated at baseline to derive risk profiles, including 5 parent variables (positive parenting, ineffective discipline, involvement, depression, life stress) and 4 child variables (externalizing, internalizing, social skills, self-regulation). Outcomes included teacher-reported aggression at four annual assessments.

Results: A latent profile analysis (LPA) was used to identify common youth/parent risk profiles in the sample across 9 baseline measures. The LPA revealed a best-fitting 5-class solution, including the following classes 1) Uninvolved parents with normative children; 2) Depressed/stressed parents with low positive parenting with children with high externalizing and poor self-regulation; 3) Parents with strong parenting with normative children; 4) Parents with strong parenting with children with high externalizing; 5) Parents with high ineffective discipline with children with high externalizing and internalizing. Intervention condition was evaluated within each class as a predictor of aggression slope in an overall latent growth model over four annual assessment points. Classes 2 and 4 demonstrated the strongest intervention response. Classes 1 and 3 were not responsive to the intervention, and Class 5 demonstrated evidence of iatrogenic effects.

Conclusions: Parent/child risk profiles were associated with variability in response to the Early Risers intervention. These profiles may be used to tailor preventive interventions by targeting those youth/families most likely to respond as well identifying those not likely to benefit or even demonstrate iatrogenic effects in order to provide alternative programming. This approach has considerable potential to increase preventive intervention effectiveness and efficiency.