Randomized Controlled Trial of the Classroom Scerts Intervention Project for Students with ASD

Background: Current prevalence studies estimate that 1 in 68 children in the US is diagnosed with Autism Spectrum Disorder (ASD; CDC, 2014). The significant challenge and expense for schools to effectively educate this population makes the need for comprehensive, evidence-based interventions for students with ASD a priority. The Classroom SCERTS Intervention Project (CSI) is a classroom-based intervention using a cost-effective approach to achieve intensity of active engagement for young elementary students with ASD by training educators to implement SCERTS across the school day.

Objectives: The primary objective of this randomized controlled trial was to evaluate the efficacy of CSI for children with ASD in kindergarten through 2^nd grade classrooms. A total of 235 students with ASD in 140 classrooms nested within 64 schools participated in the study. The two treatment conditions were as follows: 1) CSI was provided with access to a 3-day SCERTS training and ongoing coaching during the school year; and 2) Autism Education Training (AET) was provided with access to a series of web-based training resources for educators.

Methods: A stratfied randomized design was used wherein school pairs were matched on demographic variables (e.g., race, % free and reduced lunch, number of students) and randomly assigned to either CSI or AET for a 9-month school year. Effectiveness of CSI and AET was compared using direct observation of student active engagement. The Classroom Measure of Active Engagement (CMAE; Wetherby, Morgan, & Sparapani, 2014) is a multicomponent observational tool designed to measure active engagement in students with ASD. In addition, both conditions were compared using teacher report measures (SSRS, BRIEF), parent report of adaptive behavior (Vineland), and standardized measures of language.

Results: Baseline comparisons indicated equivalency of groups on all measures with one exception on the CMAE (Productivity).  After 9 months, students in CSI made significantly greater gains than students in AET on two components of active engagement: (timeXcondition) productivity, F (1, 135)=4.32, p=0.042, and gaze to face, F (1,101)=7.19, p=0.009. On the SSRS, students in CSI made significantly greater improvement than AET on social skills, F (1,143)=12.38, p=0.001, and the problem behavior scale, F (1,143)=11.32, p=0.001. On the BRIEF, students in CSI also made significantly greater gains than AET on the initiate subscale, F (1,145)=6.59, p=0.011, the organization of materials subscale, F (1,145)=7.19, p=0.008, and the global executive functioning subscale, F (1,145)=7.52, p=0.007. On the Vineland, Students in CSI made significantly greater gains than AET on the communication subscale, F (1,116)=5.47, p=0.002, and on the Adaptive Behavior Composite, F (1,116)=4.07, p=0.046. Finally, students in both conditions showed significant improvement on standardized measures of receptive (PPVT) and expressive vocabulary (EOWPVT). These analyses are preliminary as we will use more sophisticated procedures (HLM) to confirm findings.

Conclusions: These findings support the effectiveness of CSI, classroom-based intervention with potential to be cost-effective. CSI demonstrated significantly greater efficacy than AET on measures of active engagement, social skills, executive functioning, and adaptive behavior.  These findings are important given the dearth of evidence on comprehensive interventions for students with ASD in educational settings.