Abstract
OBJECTIVE: The aim of this study is to perform the first meta-meta-analysis on the effectiveness of parent-based interventions for children with externalizing behavior problems. Even though parent-based interventions are considered as effective treatments the effects reported in meta-analyses are heterogeneous and the implementation in clinical practice is suboptimal. Recapitulative valid effect predictions are required to close the still existing gap between research findings and clinical practice. The meta-meta-analytic results on changes in child behavior shall result in a clear signal for clinical practice. METHODS: This meta-meta-analysis encompasses 26 meta-analyses identified via search in electronic databases (PsycINFO, Medline, PubMed). Meta-analyses had to report effects of parent-based interventions on child behavior and focus on children under the age of 13 years with externalizing behavior problems in a clinical setting. Analyses were based on random-effects models. To combine results, the effect estimates of the meta-analyses were transformed to SMD and weighted to correct for primary study overlap. The meta-meta-analysis is registered on PROSPERO, registration number CRD42016036486 and was conducted in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses statement (PRISMA). RESULTS: The results indicate a significant moderate overall effect for child behavior (SMD = 0.46) as well as for parent reports (SMD = 0.51) and observational data (SMD = 0.62). Further analyses focusing on child externalizing behavior yielded significant and moderate effects (SMD = 0.45). All effects remained stable to follow-up. Considerable heterogeneity was observed within results. CONCLUSION: Parent-based interventions are shown to be effective in improving behavior in children with externalizing behavior problems, as assessed using parent reports and observational measures. The present results should encourage health care providers to apply evidence-based parent-based interventions.