Longitudinal Prediction of Adaptive Behavior from Sensory Features and Intensity of Services in Children with ASD and Other DD

Friday, May 12, 2017: 12:00 PM-1:40 PM
Golden Gate Ballroom (Marriott Marquis Hotel)
K. L. Williams1, L. R. Watson2, A. V. Kirby3, J. Sideris4, J. C. Bulluck1 and G. T. Baranek1, (1)Department of Allied Health Sciences, University of North Carolina at Chapel Hill, Chapel Hill, NC, (2)University of North Carolina- Chapel Hill, Chapel Hill, NC, (3)University of Utah, Salt Lake City, UT, (4)Frank Porter Graham Child Development Institute, Chapel Hill, NC

Many behavioral therapies target sensory features in autism spectrum disorder (ASD) and other developmental disabilities (DD) throughout childhood. Little is known about the association of these sensory features with adaptive functioning over time in ASD, how these adaptive outcomes may vary in association with various treatments, and if these associations differ by diagnosis.


Our aims were to (1) examine the longitudinal association between early sensory features [hyporesponsiveness (HYPO), hyperresponsiveness (HYPER), and sensory interests, repetitions, and seeking behaviors (SIRS)] and adaptive functioning outcomes (daily living, socialization, communication, and an overall composite score); and (2) determine if these associations are moderated by services received or diagnostic category.


50 children with ASD [Time 1: 5.5(2.3) years] and 31 children with DD [Time 1: 5.7(2.6) years] participated across two time points approximately 3.6(1.5) years apart. Sensory measures at Time 1 included two parent-report measures: the Sensory Experiences Questionnaire Version 3.0 and the Sensory Profile; and two observational assessments: the Sensory Processing Assessment and the Tactile Defensiveness and Discrimination Test-Revised. Adaptive outcomes at Time 2 were measured using the Vineland Adaptive Behavior Scales-Survey Edition. We also collected data on amounts of therapy (occupational, speech-language, and physical) and special education services received between timepoints. Data were analyzed using a series of step-down regression models, with diagnostic group and services included as potential moderators. Covariates included child gender, age at baseline, IQ, household income, mother’s education, and time between data collection points.


Greater parent-reported HYPO behaviors were associated with lower Daily Living scores after controlling for therapy (F(1, 59) = 4.17; p = 0.05) and special education services (F(1, 59) = 4.52; p = 0.04), with a steeper decline for DD than ASD. The same trend was significant for ABC scores (F(1, 73) = 4.16; p = 0.05) and Socialization (F(1, 73) = 3.89; p = 0.05) when controlling for special education services but not significant when controlling for therapy services.

Greater parent-reported HYPER scores predicted poorer ABC (F(1, 76) = 5.65; p =0.02) and Daily Living outcomes (F(1, 62) = 14.02; p < 0.001) across groups and services amount. Three-way interactions between HYPER, group, and educational treatment (F(1, 70) = 6.20; p = 0.02) and Communication (F(1, 70) = 4.60; p =0.04) were also significant for ABC.

Higher parent-reported SIRS behaviors were associated with better adaptive outcomes for the DD group, and poorer adaptive outcomes for the ASD group on all outcomes of the VABS after controlling for both service types. These trends were in the opposite direction in observer-measured SIRS. As scores increased, ABC outcomes increased for the ASD group and decreased for the DD group (F(1, 73) = 4.74; p = 0.03).


Sensory features and therapy services were associated differentially with later adaptive outcomes, sometimes influenced by type of sensory measure (parent-report or observed) or diagnostic group. Interestingly, services were associated with poorer adaptive outcomes, possibly because children with more severe ASD receive more services regardless of effectiveness. Future treatment studies using RCTs are needed to address this question.