The Early Learning Measure As a Predictor for 12- Month Adaptive Behavior Outcomes for Children with Autism and Intellectual Disabilities

Background:  Children with autism (ASD) and intellectual disabilities (ID) have a serious need for assistance in developing adaptive behaviours and what predicts this development is not well understood.  This emphasis on the development of adaptive behaviours is imperative for severely affected children because adaptive behaviours encompass daily living skills, socialization, and functional communication that are necessary in maximizing independence. Identifying variables that predict outcomes in adaptive behaviour can be useful in examining a child’s prognosis. 

Objectives:  The current study investigates whether the Early Learning Measure (ELM), an assessment tool of early cognitive skills, is useful in predicting adaptive functioning in children with ASD and ID. after 12 months.

Methods:  Thirty-eight children (mean age= 55.23 mos, SD=13.83) were enrolled in the study and had independent clinical diagnoses of Autism or PDD-NOS according to DSM-IV criteria. At baseline, children were given a cognitive assessment (Mullen Scales for Early Learning or Stanford Binet- 5^th edition), an adaptive functioning assessment (Vineland Adaptive Behaviour Scales; VABS), the Childhood Autism Rating Scale, and the ELM. The ELM was again administered at the end of every month for a 4- month period. The ELM assessed children’s mastery of receptive instructions, expressive labels, non-verbal imitation and verbal imitation. ELM “mastery” was defined as scoring 80% in each domain at baseline or any time point within the 4-month assessment period.

Linear regressions were performed to identify significant predictors of the 12- month VABS Adaptive Behaviour Composite (ABC) score. Tested independent variables were: age at baseline, CARS score, NCS (derived from the Mullen or Stanford Binet- 5^th edition nonverbal scales), ELM Mastery and entry VABS subscale standard scores. Variables that were found to be significant (at the p<.05 level) were then subjected to a hierarchical linear regression analysis.

Results:  Results indicated that a model incorporating baseline measures excluding the ELM (VABS ABC, VABS daily living, VABS socialization and NCS) accounted for 57% of the variance of the 12-month VABS ABC. ELM mastery accounted for an additional 17% of the variance  over and above all other tested baseline predictor variables (p < 0.01).

Conclusions:  The ELM is a useful measure for the prediction of adaptive functioning in children with ASD and ID after 12 months. Our findings have important implications in understanding adaptive functioning prognoses and determining next steps in treatment planning for children with intellectual disabilities functioning at the severe end of the autism spectrum. Using the ELM, clinicians will be better able to assess how successful a child will be in acquiring adaptive behaviours after 12 months. This will allow for the development of novel treatment plans that focus on adaptive functioning and independence.