Meta-Analysis of Parent Skills Training for Parents of Children with Autism Spectrum Disorder

Background:  Parent skills training programs are a popular treatment for families of children with autism spectrum disorders. Given the lower resource output of parent skills training programs compared to comprehensive treatment models, such programs have the potential to be a widely used intervention especially in low-resource settings, were a majority of individuals with autism spectrum disorders reside.

Objectives: The primary objective of our review was to systematically review and meta-analyze evidence to determine if parent skills training programs for parents of an individual with an autism spectrum disorder produce greater benefits than comparator on child and parental outcomes 

Methods: We searched 11 databases through September 2014 with no language or publication restrictions and used snowball methods to locate relevant articles. We included studies that were conducted using randomized control trial design that had a sample size of at least 10 participants per pairwise comparison that conducted an experimental comparison of parent skills training programs to a no-treatment control group, including waitlist control, or a treatment as usual comparison group for parents or caregivers who had a child with an autism spectrum disorder. We extracted data on study characteristics, study level risk of bias using Cochrane methodology, sample characteristics, intervention content and delivery methods, outcome measures, and study results and effects. Two researchers independently extracted data, with discrepancies resolved through consensus.  We combined the effects of parent skills training programs using a random effects meta-analysis with an inverse-variance weighted mean effect size for six outcomes (adaptive behavior, child development, problem behavior, parent skills, parent psychological health, and interpersonal family relations).

Results: We screened 606 full-text records and identified 21 studies with 22 independent pairwise comparisons. Random effects meta-analyses showed better outcomes in reduction in problem behaviors (k = 8, g = 0.38; 95% CI 0.19 to 0.57), but not for child development (k = 11, g = 0.14; 95% CI -0.07 to 0.35). Parents show improvements in parenting skills (k = 11, g = 0.84; 95% CI 0.64 to 1.03), psychological health (k = 13, g = 0.34; 95% CI 0.19 to 0.50), and interpersonal family relations (k = 7, g = 0.43; 95% CI 0.23 to 0.64). We also examined adaptive behavior but only located two studies examining this outcome thus did not statistically synthesize the data due to the small sample.

Conclusions: This review and meta-analysis of parent skills training programs for children with autism spectrum disorders shows the strongest evidence to date that this intervention leads to positive child and family outcomes. Specifically, our findings show children have reduced frequency and/or intensity of problem behaviors and parents have improved parenting skills, psychological well-being, and interpersonal relations with other members of their family.