19046
A Longitudinal Examination of Adaptive Behavior in Youth with Autism Spectrum Disorder: Contributions of Executive Function

Friday, May 15, 2015: 5:30 PM-7:00 PM
Imperial Ballroom (Grand America Hotel)
C. E. Pugliese1, L. G. Anthony2, J. F. Strang2, K. M. Dudley2, G. Wallace3, D. Q. Naiman4, A. B. Ratto5 and L. Kenworthy6, (1)Children's National Medical Center, Washington, DC, (2)Center for Autism Spectrum Disorders, Children's National Medical Center, Rockville, MD, (3)NIMH Intramural Research Program, Bethesda, MD, (4)Applied Mathematics and Statistics, Johns Hopkins University, Baltimore, MD, (5)Children's National Health System, Silver Spring, MD, (6)Children’s Research Institute, Children's National Medical Center, Washington, DC
Background:  

Approximately half of all children with ASD do not have co-occurring intellectual disability (ID), yet outcome remains poor. Because outcome in ASD is more related to adaptive behavior skills than cognitive level, it is important to identify predictors of adaptive behavior. Prior cross-sectional research has demonstrated impairments in executive function (EF) abilities are associated with adaptive behavior deficits in youth with ASD. However, longitudinal research is needed to confirm predictors of adaptive behavior over time in ASD. 

Objectives:  

1) To characterize longitudinal change in adaptive behavior in children with ASD without ID, and 2) determine whether prior estimate of EF predicts future adaptive behavior scores. 

Methods:  

65 youth (13 females) with a DSM diagnosis of ASD were evaluated on multiple occasions (M=2.63, SD=0.80, range 2-5) separated by at least 6 months (M=3.40 years, SD=5.41) resulting in a total of 170 adaptive behavior observations. Participants had a prior assessment of EF for 92 subsequent adaptive behavior evaluations. Participants had a mean age of 8.01 years (SD=2.51) at first evaluation, possessed average IQ (M=107.03, SD=19.83), and met CPEA criteria for ‘broad ASD’ on the ADI-R and/or ADOS. 

Adaptive behavior and EF were assessed via parent-report using the Vineland Adaptive Behavior Scales (VABS) and Behavior Rating Inventory of Executive Function (BRIEF), respectively. To assess longitudinal change, consecutive observations on the VABS were characterized as Improved, Deteriorated, or Unchanged based on reliable change indices (RCI). A multiple regression was conducted to determine whether global EF abilities predicted subsequent VABS scores after accounting for baseline adaptive behavior, age, IQ, length of time between assessment, and age*length of time between assessment (interaction term). An exploratory analysis examined which specific EF skills predicted adaptive behavior.

Results:  

Most participants did not make significant improvements on VABS standard scores over time, despite intact cognitive abilities (see Figure 1). While daily living skills declined across time, the relationship between age and communication/socialization skills was qualified by an interaction between age and length of time between evaluations. IQ was not a significant predictor of later adaptive behavior in any domain. Prior global EF abilities predicted later VABS Daily Living and Socialization scores, but not Communication scores (see Table 1). Exploratory analyses determined higher monitoring skills were a robust predictor of better subsequent adaptive behavior across VABS domains. Additionally, better inhibition skills predicted better subsequent daily living and socialization skills. Finally, better shifting (e.g., flexibility) predicted better subsequent socialization skills. 

Conclusions:  

For the majority of youth in the sample, adaptive behavior standard scores did not improve over time. Higher EF skills, but not cognitive ability, predicted better subsequent adaptive behavior scores in daily living skills and socialization. Results support the notion that EF skills are important to real world outcomes. It will be important to target adaptive skills, and EF skills that contribute to them, in individuals with ASD across developmental periods from childhood through adolescence in order to improve outcomes.

*Note: We are currently in the process of collecting another wave of participant data to analyze with multi-level modeling techniques.