A Randomized Trial of a Modular Approach for Autism Programs in Schools (MAAPS)

Background: A number of empirically supported interventions exist for children with ASD. However, we know little about the feasibility and effectiveness of interventions when implemented in schools; barriers (e.g., limited buy-in, resources) often temper successful implementation by educators (Kasari & Smith, 2013). Modular Approach for Autism Programs in Schools (MAAPS) is a comprehensive and modular intervention, developed to address common barriers by providing educators with a flexible framework and in-vivo coaching to implement evidence-based interventions. Through MAAPS, educators identify goals for their student and select modules to address core and associated features of ASD. A coach uses behavioral skills training for educators to implement interventions. Iadarola et al. (2017) presented results of a MAAPS pilot, which provided preliminary support for the intervention: educators viewed the intervention as socially valid, implemented interventions with high fidelity, and favorably rated student outcomes.

Objectives: Expand findings by conducting an efficacy trial of MAAPS to evaluate (1) feasibility, (2) social validity, and (3) student outcomes.

Methods: We conducted an underpowered RCT across three sites with 28 students with ASD and their educators (student-teacher dyads), randomized to MAAPS or our control, enhanced services as usual (ESAU). For ESAU (n=14 dyads), educators were offered up to 12 hours of didactic trainings on module-related topics. For MAAPS (n=14 dyads), educators were assigned a coach (research personnel) and engaged in a systematic process of (1) selecting individualized goals and modules and (2) receiving up to 12 hours of in-vivo coaching per module. An independent evaluator administered the Developmental Disabilities modified Children’s Global Assessment Scale (DD-CGAS) on students at baseline, mid-intervention, and end of intervention. Social validity was assessed from the 29-item Usage Rating Profile (URP) (6=highest rating) and a 10-item scale of coaching quality (4=highest rating). Coach and teacher fidelity of implementation was also collected.

Results: Eleven dyads completed MAAPS and 13 dyads completed ESAU. Fidelity: Coach and teacher fidelity was high (above 90%), with high inter-rater reliability (97-99%). Social Validity: On the URP, teachers rated MAAPS highly for acceptability, M(SD) = 5.22(.31), usability, M(SD) = 5.61(.10), and feasibility, M(SD) = 5.47(.27). Mean coaching quality was rated 3.84(.07).Student Outcomes: Although not statistically significant, students receiving MAAPS showed gradual increases in mean DD-CGAS scores from baseline (51.5) to midpoint (54) and exit (61), with associated improvement in functioning from “moderate” to “slight” impairment. ESAU did not show change in mean functioning across timepoints (i.e., 57.5 at baseline, 58.5 at midpoint, 60 at post-treatment).

Conclusions: To our knowledge, this is the first comprehensive, modular intervention for educators of students with ASD. Educators rated MAAPS favorably; both educators and coaches implemented the intervention with high fidelity. Although we were underpowered to detect group differences on student outcomes and there were large differences in baseline scores between groups, greater within-group improvements were seen in MAAPS vs. control. Results indicate that a modular intervention approach with in-vivo coaching for educators may be more effective than providing didactic trainings for supporting students with ASD.