Examining Curriculum-Based Assessment as a Measure of Early Intervention Outcome

Background:

As children with autism spectrum disorders (ASD) are increasingly identified at earlier ages, researchers and clinicians must adapt to develop effective, comprehensive interventions suitable for very young populations. Currently, the literature is lacking empirically supported intervention packages for children under age three (Rogers & Vismara 2008). Furthermore, standardized assessments often used to evaluate the effects of treatment may not be sensitive to progress in this population.

Objectives:

This study examined the utility of a curriculum-based measure to assess the efficacy and appropriateness of an early intervention for children under age three with risk for ASD. Each participant received an individualized program addressing communication, social, play, cognitive and developmental skills. Interventions included Strategies for Teaching Based on Autism Research Program (STAR; Arick, Loos, Falco, & Krug, 2004), systematically combined with the Teaching Social Communication (TSC) curriculum  (Ingersoll & Dvortcsak, 2010).

Methods:

To date twenty-eight children between 13 and 27 months of age (M=22.4) identified withrisk for, ASD have participated in the intervention as part of the UCSD Autism Center of Excellence. Children receive an average of 9.29 treatment hours per week. The curriculum is based on the STAR program, which utilizes behavioral interventions including Pivotal Response Training, Discrete Trial Training and Functional Routines to teach a range of skills. To target engagement and social interaction, including joint attention, TSC was used to supplement the STAR Program. Initially, and every 3 months thereafter, each child was assessed using the adapted Student Learning Profile (aSLP) from the STAR curriculum, designed to measure mastery level of the child’s skills across domains. The aSLP provides a consistent and comparable measure of skills mastered across participants during the treatment period. Children also received a battery of standardized cognitive and behavioral assessments every 6 months.

Results:

To date, 20 children have completed at least 6 months of treatment. As measured by the aSLP at post treatment, all children showed increases in the number of skills mastered across intervention domains (M=93.5; range=29-151).  As expected, aSLP results did not correlate with standardized assessments.  Likewise, standardized assessments of cognitive or communication skills at intake were not predictive of progress.  Results did identify a novel treatment response index, namely rate of skill mastery, in that the rate of learning after three months of intervention was indicative of outcome.

Conclusions:

All children benefited from the intervention as indexed by the acquisition of new skills.  A greater range of scores and progress was obtained with the curriculum assessments in comparison with standardized tests, and early scores were often predictive of outcome.  As such, curriculum-based assessments may be a more appropriate method for measuring intervention progress in young children with ASD.  Overall, these data provide preliminary evidence that early intervention programs designed for children with ASD over age three can be adapted to improve communication, cognitive, play and social skills in children under three. However, the variability in skill acquisition across children is high.