The Development of Adaptive Skills in Young Children with ASD -- An Examination Across One Year Using Vineland-II

Background: The Vineland Adaptive Behavior Scale has been an important component for diagnostic assessments, treatment planning, progress monitoring, and research on autism spectrum disorders (Chawarska & Bearss, 2008). Mental age or IQ has been found to be significantly correlated with adaptive skills (Loveland & Kelley, 1991).  Researchers have consistently found that there is a significant decline in adaptive skills standard scores as children on the spectrum grow older (Fisch et al, 2002; Carter et al., 1998). In addition, levels of autism symptoms, cognitive, and language functioning affect the course of adaptive skills development (Perry, Flanagan, Geier, & Freeman, 2009). This study examines the development of adaptive skills measured by the Vineland-II within a large sample of young children on the autism spectrum.

Objectives: This study investigates the following questions: a) how do adaptive behaviors change in one year among young children with ASD? b) What are the relationships among initial mental age, calibrated autism severity score, and the magnitude of changes in adaptive behavioral skills?

Methods: Participants are 100 children on the autism spectrum who are participating in a longitudinal study on ASD. At Year 1 visit, participants (Mean Chronological Age = 31 months, SD = 4.37), received a battery of assessments, including the ADOS, ADI-R, Vineland-II (parent/ caregiver interview), and Bayley-III Cognitive Scale.  Approximately one year later, ADOS, developmental testing, and the Vineland-II were re-administered.

Multivariate repeated measure analysis was used to examine the within-subject changes on the domains of the Vineland-II for children with and without delays on initial Bayley-III Cognitive Scale (n = 46 and 50, respectively). Cognitive delay was defined as Bayley age equivalent / chronological age < .75. Partial correlation coefficients were used to examine the relationship among initial nonverbal ability, calibrated ADOS severity score (Gotham, Pickles, & Lord, 2009), and the magnitude of adaptive skill changes in the entire sample.

Results:  Significant between-group and interaction effects (Group x Year) were found with both standard scores and age equivalents (α =.05). Using domain standard scores, within-subject difference was found on Communication only (F (1, 94) = 9.59, p = .003). Using age equivalent scores, significant within-subject differences were found on Communication, Daily Living, and Socialization (ps = .000) and Year 2 scores > Year 1 scores. Simple correlations were significant among all variables at α = .01. When Year 1 Bayley age equivalent was partialled out, the magnitude of changes from Year 1 to Year 2 on the Vineland domains were not significantly correlated with the calibrated ADOS severity score.

Conclusions: Young children on the autism spectrum make gains in communication, daily living, and socialization skills, as reflected by the increases of the age equivalent scores. The rate of domain age equivalent score change is different for children with and without initial cognitive delay. When Year 1 cognitive scores are statistically controlled, the level of autism features is not significantly related to gains in adaptive skills.