Examining Differential Mediation in Brief and Comprehensive Parenting Training

Introduction:Parent Management Training Oregon Model (PMTO) is a family of evidence-based parenting programs aimed at preventing and reducing conduct problems. PMTO has been adapted for a wide array of contexts, ranging from short-term to comprehensive formats. In a recent review (Forehand et al., 2014), PMTO was cited as the program with the most support for parenting practices as mediators. This study extends previous research by examining mediation in two samples from Norway, and by examining whether the effects on child conduct problems are differentially mediated in short-term (Brief Parent Training; BPT, 5 sessions) or comprehensive PMTO (30 sessions). First, we examined whether BPT and PMTO produced direct effects on child conduct problems at follow-up (6 months post-intervention). Second, we examined whether the effects were mediated through reduced harsh parenting and increased positive parenting at post-intervention (immediately after intervention). Third, we tested for differential mediation in BPT and PMTO.

Method:This study was a secondary analysis of 2 randomized effectiveness trials: 1) The PMTO parent group trial that consisted of 137 families randomized to PMTO or regular services, and 2) the BPT trial that consisted of 216 families randomized to BPT or regular services. Assessments were completed pre-intervention, post-intervention, and at follow-up. Parent reports were used to estimate latent and observed constructs, and autoregressive, multi-group models were estimated in Mplus.

Results: First, we found positive direct effects of BPT (Cohen’s d = .32, p=.01) and PMTO (Cohen’s d = .58, p=.000) on conduct problems. Second, when including parenting practices in an overall model (including both samples) a significant indirect effect emerged via harsh discipline only, γ = −.04, p < .01, SE = .02, 95%, CI: −.7 to −.01. When we ran multi-group comparisons, this indirect relationship did only hold true in the PMTO sample, γ= −.07, p < .02, SE = .03, 95%, CI: −.14 to −.01.  

Conclusions: PMTO produced larger direct effects on conduct problems at follow-up than BPT. Change in harsh discipline emerged as a mediator in PMTO, but not in BPT, indicating differential mediation in the two interventions. This finding may be due to the difference in comprehensiveness in BPT and PMTO. That is, increased comprehensiveness may be related to sustained effects on parenting practices which again is related to long-term positive effects on conduct problems. Thus, the effectiveness of BPT may be better sustained if families are followed up after the initial intervention period.