Efficacy of the ASAP Intervention for Preschoolers with Autism Spectrum Disorder

Background: Behavioral interventions aimed at improving the social-communication and/or play skills of children with ASD have led to positive short- and longer-term outcomes. Yet, many of these interventions were implemented in clinic-based settings or by research staff. This presents a need to develop and examine the efficacy of interventions implemented by real-world practitioners in natural settings. The Advancing Social-communication And Play (ASAP) intervention targets these skill areas, with classroom educational teams implementing the intervention in public school-based, preschool classrooms.

Objectives: The objectives of this multi-site, four-year study were to examine the effects of the ASAP intervention for the (a) primary outcomes of social-communication and play, and (b) secondary outcomes of engagement and challenging behavior.

Methods: A cluster randomized trial design was used to assign classrooms to ASAP (n=40) or a “business-as-usual” (BAU) control (n=38) group. Eligible preschool-aged children within a classroom were consented prior to informing educational teams of their group assignment. For children in the ASAP group (n=85), their educational team had to consist of at least one teacher, teaching assistant, and related service provider. To support ASAP implementation, the team received intervention materials, didactic training and ongoing coaching throughout one school year. Children in the BAU control group (n=76) continued to receive their typical classroom instruction. Blinded observational coding was used to assess the primary outcomes of social-communication and play as well as the secondary outcome of engagement. Teacher report on the Caregiver-Teacher Rating Form (CTRF) was used to measure challenging behavior. Primary outcome measures were collected at three time points, while secondary outcome measures were collected at two time points.

Results: Random intercept mixed models or hierarchical linear models (HLM) were used to assess primary and secondary outcomes. The primary outcome models were three-level HLM with observations nested within child and child within classroom. The secondary outcomes, with two time points, were run as pretest controlled ANCOVA. These were two-level models with child nested in classroom. Based on preliminary findings, no significant group differences were found for the primary social-communication or play outcomes. However, significant differences in favor of the ASAP group were found for both engagement and challenging behavior. There were statistically significant differences and moderate to large treatment effects found for: unengaged (d=.71), some engagement (d=.50), and overall engagement (d=.59). By post-test, the ASAP group spent more time in the some engagement state, less time unengaged, and blind observers rated them higher for overall engagement compared to the BAU group (Table 1). Further, children in the ASAP group received significantly lower scores on the CTRF measure for externalizing behavior and total challenging behavior (Table 2).

Conclusions: When delivered in authentic settings by real-world practitioners, ASAP led to improvements in children’s ability to appropriately engage in classroom activities and decreased their level of challenging behavior. Although no group differences have been found for social-communication or play, from a theoretical standpoint it may be that changes in engagement (or challenging behavior) must be achieved first in order to lead to downstream changes in these core deficit areas.