Efficacy of a School-Based Treatment Model for Adolescents with Autism: Effects on School Quality

Background: The increased prevalence of autism spectrum disorder (ASD) has created a demand for efficacious treatment services for children and youth with ASD, and one context in which most participate is the public schools. While much research has been conducted on comprehensive treatment models (CTM) for preschool and early school age children with ASD, few programs have been developed and evaluated for adolescents with ASD. This has created a gap in the field between needed effective services and treatment approaches that are evidence-based.

Objectives: The purpose of this study was to examine the efficacy of a comprehensive treatment model designed to improve the quality of educational programs for adolescents with ASD attending high school.

Methods: The Center for Secondary Education for Students with ASD (CSESA) designed a CTM for high school students with ASD that consisted for four program features: Academics, Independence, Social Competence, and Transition/Families. After training on the model, an “autism team” at the school implemented the program over a two year period. A primary feature of the CSESA model was to improve the overall quality of the program, which would serve as the foundation for the four features of the CSESA model. The Autism Program Environment Rating Scale (APERS) was the primary measure used to assess program quality. The APERS is a 66-item, five-point Likert-type scale that provides a total quality rating score and also four composite scores that reflect the quality of program features noted previously.
Design: Following a randomized clinical trial design, sixty high schools located in North Carolina, California, and Wisconsin were randomly assigned to the CSESA and a Services as Usual (SAU) condition. A total of 546 high school students with ASD participated in the study. The APERS was administered at the beginning of the school year and again at the end of the subsequent school year (~18-20 months later), with CSESA schools implementing the CTM and SAU schools following the practices they typically employed.

Results:

Examinations of reliability yielded Cronbach alpha scores of .94 for the total rating and an average of .76 for domain scores. Inter-observer agreement, collected on 20% of the sample, was 85% for agreement within one point and 76% for exact agreement. An ANCOVA was used to create adjusted posttest means, with the pretest as the control, for the two groups. The total mean item rating (d=.99) and the four composite scores [academic (d=.71), social (d=1.32), independence (d=.69), and transition/families (d=.83)] were significant at the p> .05 level (with Bonferonni adjustment), with mean rating favoring the CSESA group (see figure below).

Conclusions: This study demonstrated the efficacy of the CSESA model for improving the quality of the program environment for high school students with ASD. The next steps in this research will be to examine the relationship of program quality to the fidelity with which program features and evidence-based practices are implemented and the impact on the adolescents with ASD.