Flexibility Explains Adaptive Behavior in Youth with Autism Spectrum Disorder

Oral Presentation
Saturday, May 12, 2018: 3:16 PM
Willem Burger Zaal (de Doelen ICC Rotterdam)
J. R. Bertollo1, J. F. Strang2, S. Seese3, L. Kenworthy3, G. L. Wallace4,5 and B. E. Yerys1, (1)Center for Autism Research, Children's Hospital of Philadelphia, Philadelphia, PA, (2)Center for Autism Spectrum Disorders, Children's National Health System, Washington, DC, (3)Children's National Health System, Washington, DC, (4)National Institutes of Health- National Institute of Mental Health, Bethesda, MD, (5)The George Washington University, Washington, DC
Background: Adaptive behavior, the ability to function independently in everyday life, is an important measure of functional outcomes for individuals with autism spectrum disorder (ASD). Individuals with ASD tend to achieve poorer outcomes and exhibit worse adaptive behavior than groups with other developmental disorders, to a degree that cannot be explained by IQ alone. Prior research points to the importance of flexibility, an executive function skill, as an important factor predicting adaptive behavior in school-age youth with ASD. However, this research was limited to an 8-item informant-report measure of flexibility that was not designed to measure specific flexibility skills (e.g., flexibility in thought, routines, or social behavior). With our team’s recent publication of the Flexibility Scale, we now have a multi-dimensional parent-report measure. Identifying which flexibility skills relate to adaptive behavior will help us better tailor treatments to increase an individual’s ability to function independently in daily activities.

Objectives: To examine specific flexibility skills that relate to an individual’s level of adaptive behavior.

Methods: Participants included 216 individuals with ASD (mean age=10.67 years, 83.3% male). Adaptive behavior was measured by the Vineland Adaptive Behavior Scales, First and Second Editions. Flexibility was assessed by the Flexibility Scale-Revised, which includes 27 items pertaining to Social Flexibility, Routines and Rituals, Transitions/Change, Special Interests, and Generativity. Flexibility skills, captured on the 8-item Shift scale from the Behavior Rating Inventory of Executive Function (BRIEF), were also examined, to check for replicability from prior research that utilized this measure. To evaluate the role of flexibility above and beyond covariates of age, IQ, and gender, three stepwise regressions were conducted, one for each Vineland domain. The first step evaluated covariates; the second step added the five Flexibility Scale-Revised Subscales. We hypothesized replicating prior findings of BRIEF-Shift scale scores associating with Vineland Socialization skills, while further specifying this association as driven by Social Flexibility deficits, using the more granular Flexibility Scale-Revised.

Results: In line with prior literature, the BRIEF-Shift scale was significantly associated with Vineland Socialization skills (p<0.001). When examining the multi-dimensional Flexibility Scale-Revised, as hypothesized, Social Flexibility accounted for significant variance in Socialization skills, as did Generativity and Routines and Rituals. Together, the Flexibility Scale-Revised subscales contributed 26.8% of variance in Socialization skills (p<0.001). Additionally, Flexibility Scale-Revised subscales explained 9.3% of variance in Daily Living Skills (p<0.001), with Generativity reaching statistical significance. Flexibility Scale-Revised subscales, particularly Generativity, Routines and Rituals, and Transitions/Change, explained 15.9% of variance in Communication skills (p<0.001).

Conclusions: As predicted, a more granular measure of flexibility impairments replicated and specified prior relationships with social adaptive functioning skills. It also showed strong relationships with other areas of adaptive behavior that were not seen with the BRIEF-Shift scale alone. Future studies should examine both the predictive value of specific flexibility skills with longitudinal outcomes, and convergence with performance-based measures at a single time point. Furthermore, dismantling the construct of flexibility into its subcomponents is critical for supporting a personalized treatment approach, so that children with flexibility deficits can receive more targeted supports to improve outcomes.