31344
Predictors of Response to Early Parent-Mediated Interventions for Young Children with Autism: A Systematic Review
Objectives: We conducted a systematic review to identify child and family characteristics associated with developmental outcomes for young children with/at risk of autism receiving parent-mediated early intervention.
Methods: Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and Covidence software platform, we identified relevant literature using Psych INFO, Medline, Pubmed, CINAHL and ERIC databases and manual review of reference lists from key papers. Eligibility for inclusion in the systematic review was the paper: (1) examined the efficacy/effectiveness of early parent-mediated behavioural interventions; (2) utilised a comparative design (e.g., RCT); (3) included participants with/or having high likelihood of developing autism, aged 0-6 years; and, (4) reported a statistical moderation between a child/family characteristic and a quantitative, established outcome measure of child development. Two independent reviewers screened titles and abstracts of all identified papers. Full-texts were accessed to confirm eligibility for papers meeting inclusion criteria. Relevant data were extracted for each paper by two independent reviewers. Discrepancies were identified and resolved through discussion and with the aid of an additional reviewer where necessary.
Results: Ten papers from eight individual studies met inclusion criteria. Each varied in the predictors, outcomes and interventions explored. There was some evidence that children who commenced interventions with lower skills made greater gains in language, adaptive behaviour and social communication, but not for all children. Two papers reported that higher parental stress and higher family income were associated with larger gains in child outcomes for the intervention group. Two papers (reporting on the same intervention study) also found associations between features related to higher quality parent-child interactions and better child outcomes following the intervention. However, a key drawback of the identified studies is the heterogeneity of predictors, outcomes and interventions, making it difficult to synthesise findings and draw strong conclusions. Additionally, risk of bias – particularly around lack of blinded outcome assessment – was an issue in some identified papers.
Conclusions: Identifying child and family characteristics that predict favourable intervention outcomes may help further our understanding of which children with autism may benefit most from which intervention. More tailored interventions could help optimise short and longer terms outcomes. Further well-designed studies, including larger sample sizes, are needed to identify differential predictors of intervention outcomes. Inclusion of additional parent-related predictors of parent-mediated interventions in future studies would also strengthen the literature