2007-08best_abstract
- 1. HONOURS BACHELOR OF SCIENCE (2008) MCMASTER UNIVERSITY Hamilton, Ontario TITLE: The Early Learning Measure as a predictor of outcome following Intensive Behavioural Intervention in young children with severe autism AUTHOR: Jacqueline Best SUPERVISOR: Dr. Jo-Ann Reitzel NUMBER OF PAGES: ix, 72
- 2. ABSTRACT The Early Learning Measure as a predictor of outcome following Intensive Behavioural Intervention in young children with severe autism Jacqueline Best1 , Jo-Ann Reitzel1,2 , Jane Summers1,2 , Peter Szatmari1,2 , and Lonnie Zwaigenbaum3 1 McMaster University Department, 2 Hamilton Health Sciences, 3 Capital Health, Alberta Intensive Behavioural Intervention (IBI) for young children with autism is an extensively researched intervention and has resulted in significant developmental gains in many (Lovaas, 1987; McEachin Smith, & Lovaas, 1993; Smith et al., 2000), but not all children (Sallows & Graupner, 2005). To date, little attention has been paid to identifying predictors of good outcome for young children with severe autism in a prospective rather than post-hoc manner. Moreover, intensity of IBI and pre-treatment static measures such as intellectual functioning (Smith et al., 2000) tend to be poor predictors of outcome and do not account for a child’s ability to learn once in treatment. In Ontario, IBI is provided for children with severe autism, but demand for treatment outweighs the limited resources, and many children remain on waitlists. An understanding of the relationship between early learning characteristics and developmental outcomes would be useful for maximizing available IBI resources and preventing long-term exposure to ineffective treatment for children with autism if they are not benefiting from IBI. The objective of this study is to determine if the Early Learning Measure (ELM) is a useful predictor of 6- and 12-month outcomes on measures of cognitive functioning and adaptive behaviour for children with severe autism following IBI, and if the ELM is a better predictor of these outcomes than baseline characteristics, which include entry IQ, age at entry to IBI, and autism severity.
- 3. A sample of 29 children with Autistic Disorder or PDD-NOS was administered the ELM monthly for the first four months of IBI. The ELM is a measure of skill acquisition in four domains: receptive instructions, expressive labels, non-verbal imitation, and verbal imitation. Children were grouped into mastery or non-mastery groups, with mastery defined as a correct response rate of 80% in all domains in the first 4 months of IBI. The Vineland Adaptive Behaviour Scales (VABS) and a cognitive measure (either the Mullen Scales of Early Learning or the Stanford-Binet Intelligence Scales-5th Ed.) were used to assess adaptive behaviour and cognitive functioning respectively at entry to IBI and at 6 and 12 months during IBI. Results demonstrated that the ELM is a useful predictor of outcome following IBI. Mastery of the ELM was correlated with significantly higher outcomes in adaptive behaviour and cognitive functioning. The ELM domains that served as the best predictors of adaptive behaviour outcomes were those involved in using or understanding language: Expressive Labels, Verbal Imitation, and Receptive Instructions. Regression analyses completed to determine whether mastery of the ELM served as a better predictor of outcome than baseline characteristics demonstrated that, for some outcomes (12-month Vineland Adaptive Behaviour Composite, 12- month Vineland Communication standard score), mastery of certain ELM domains or the total ELM does predict outcome better than baseline characteristics. However, the ELM alone was never the best predictor of outcome. The best predictor for each outcome was a combination of baseline characteristics and mastery of the ELM or the Expressive Labels domain of the ELM. This study demonstrates that while the ELM is a useful predictor of adaptive behaviour and cognitive functioning outcomes, a better prediction of outcome can be obtained when combining ELM mastery category with baseline characteristics. The study has also shown that the ability to expressively label objects early in IBI may be an important skill that helps
- 4. distinguish children who are likely to benefit from IBI and those who are not. Determining predictors of outcome in IBI has clinical implications and can help maximize IBI resources and prevent exposure to IBI for children who do not stand to benefit from the treatment.