To What Extent Are Parenting Interventions for Conduct Problems Likely to Affect Social Disparities in Youth Outcomes? Pooling Data from Multiple Trials of Parenting Interventions

Introduction: Parenting interventions are a key strategy to prevent and treat child conduct problems. Although beneficial main effects are well documented, the extent of subgroup effects is less clear. Moderator analyses are important for targeting interventions, and establishing whether there is ‘inequity’ of effects- i.e., if socially disadvantaged groups benefit less than more advantaged subgroups. Unfortunately, many randomized trials lack adequate power for moderator analyses. Systematic reviews have identified moderator effects, but produce conflicting results (Lundahl et al., 2006; Leijten et al., 2013), and by relying solely on aggregate trial level moderators (e.g. mean SES in a sample), fail to make use of rich within-trial variability in participant characteristics. In order to understand equity effects of programs for preventing conduct disorder, which is highly patterned by social class, it is vital to conduct well-powered moderator analyses. Methods: This we attempted by pooling individual-level data across a complete set of 14 randomized trials in Europe (1800 families) of one parenting intervention, the Incredible Years. All trials were conducted independently of the program developer, in UK, Netherlands, Ireland, Norway, Sweden, and Portugal. Where different measures of the same constructs were used, we harmonized data across trials using norm deviation scores. We examined key putative moderators of intervention effects on conduct problems, drawn from prior reviews: family SES (i.e., income, educational level, lone parent, teen parent), ethnic minority, baseline child problem severity and co-morbidity, and child age. Moderator analyses used random effects modeling (fitted using maximum likelihood) to separate individual- from trial-level variation. We used multiple imputation to account for missing data. Preliminary analyses found a pooled intervention effect size on conduct problems of d=.50. Results: Moderator analyses found that families with socioeconomic disadvantage and with ethnic minority backgrounds benefit as much from the intervention as families without those characteristics. Having severe levels of child conduct problems, comorbid ADHD or emotional problems did not diminish intervention effectiveness.  Conclusions: This study is the first to synthesize data from multiple independent trials of a parenting intervention, and suggests that the intervention, if scaled up, would not only have (well-known) average benefits across a population, but would also have no detrimental effects on social disparities in conduct problems.