Umbrella Review: Systematic Reviews of Psychosocial Interventions for Children with Autism Spectrum Disorder

Background: Autism spectrum disorder (ASD) now affects approximately one in 68 children, making it one of the most prevalent neurodevelopmental disorders of childhood (Center for Disease and Control [CDC], 2016). Although there is no cure for ASD, there is a diverse range of interventions available that are designed to target the core deficits of the disorder and can improve a child’s developmental trajectory (CDC, 2016). Over the past 20 years there has been an explosion of ASD intervention research, and as such, it is no longer feasible for individuals to read and retain the vast body of available information in this area (Interagency Autism Coordinating Committee, 2012). Indeed, it is an overwhelming and challenging task for physicians, clinicians, teachers, and policy-makers to determine which specific interventions should be recommended for each individual child. In response to this challenge, there has been a rise in production of systematic reviews (SRs). An umbrella review, which involves collating information from multiple SRs, makes it possible to better determine the effectiveness of ASD interventions and will facilitate evidence-informed decision-making.

Objectives: The present study is an umbrella review of SRs of psychosocial interventions for ASD. The review aims to summarize SRs from 2006 to 2016, evaluate their quality, identify predictive factors associated with quality, and determine best practice recommendations for intervention.

Methods: Studies were identified through comprehensive searches of six electronic databases. The search strategy consisted of keywords and medical subject headings for autism and related disorders and various psychosocial interventions. Eligible studies met the following criteria: systematic search description; participants aged 0-12 with a diagnosis of ASD; and review of a psychosocial intervention. The methodological quality of SRs was assessed using the AMSTAR tool (Shea et al., 2007) and descriptive data were extracted.

Results: The comprehensive search resulted in the inclusion of 159 SRs published between 2006 and 2016. In terms of participants, 77% of SRs included both preschool and school age children, 12% included only preschool age, and the remaining 11% included only school age. Sixty-five different journals published these SRs. Six dissertations were included. Additional characteristics of the SRs will be summarized descriptively. Evidence tables will be produced to synthesize the clinical findings and recommendations of the SRs. A backward elimination multivariable regression analysis will be conducted post-hoc to examine predictors of the methodological quality of reviews, as measured by the overall score on the AMSTAR. Review characteristics associated with the quality of reviews and several independent variables will be explored in the regression models.

Conclusions: The methodological strength of the SRs included in this umbrella review was greater than those included in Seida and colleagues’ (2009) umbrella review, indicating that they are less vulnerable to bias. Further, Seida and colleagues’ review found 30 SRs that met eligibility criteria and were published between 1996 and 2007 while the current umbrella review included more than five times as many SRs, indicating that there has indeed been an explosion of SRs focused on ASD interventions in recent years.