Using Participant Data to Extend Evidence Base for Intensive Behavioral Intervention for Children with Autism
Psychological, Educational, and Therapeutic Interventions
Applied Behavior Analysis (ABA)
Eldevik, S., Hastings, R.P., Hughes, J.C., Jahr, E., Svein, E., Cross, S. (2010).
Using participant data to extend the evidence base for intensive behavioral intervention for children with autism. American Journal on Intellectual and Developmental Disabilities, 115 (5), 381-405. Reviewed by: Lauren Pepa, Rutgers University
Why study this topic?
Much research supports the efficacy of intensive behavioral intervention (IBI) with individuals with autism for improving intellectual functioning, adaptive skills, and other core symptoms of autism. Most of this research has relied on group designs, wherein behavior changes are combined across groups of individuals and compared to changes observed in other treatments. However, individual change as a result of IBI is extremely variable, and little research has focused on the individual factors that may lead to better outcomes. One way to look at these factors is to examine the data for each individual participant in group studies and evaluate their level of change, and then to look at what factors predicted treatment success for these individuals.
What did the researchers do?
The authors looked at both individual change and predictors of change in IBI. The authors first identified research articles that evaluated IBI. This process yielded data for 453 participants from 16 studies, most of whom had been evaluated on standardized measures of intelligence and adaptive behavior. They then examined the degree of change in these measures from pre- to post-treatment and examined pre-treatment variables that predicted the amount of change made by individual participants. This was accomplished by calculating means (i.e., average) and standard deviation (i.e., variability) scores for all study measures based on the collected data from the samples. Reliable change was defined as an absolute change in index scores of 1.96 standard deviations, corresponding to a 27 point change in IQ and a 21 point change in adaptive functioning. Predictors of this change were examined using a multiple regression analysis which is a type of statistical analysis.
What did the researchers find?
Looking at individual change, the authors found that children who received IBI were more likely to show meaningful improvements in IQ scores and in adaptive functioning than those receiving no treatment and those receiving comparison treatments. Specifically, the proportion of children who achieved reliable change was 23% higher in the IBI group than in the comparison groups for IQ and 16% higher for adaptive functioning. Additionally, the number of individuals needed to observe a beneficial outcome in IBI was significantly lower than in the control group (i.e., those with no treatment) and comparison (i.e., those with alternate treatments) groups. Only five children were needed to achieve a reliable change in IQ with IBI and seven were needed for a reliable change in adaptive functioning. These findings extend previous reviews indicating that IBI is an evidence-based treatment for young children with autism. Looking at individual predictors, the authors found that higher intensity of intervention was associated with better outcomes for IQ and adaptive functioning. Lower adaptive functioning prior to treatment predicted greater gains in adaptive functioning post-treatment. Separately, higher IQ scores prior to treatment predicted more significant gains in adaptive functioning.
What were the strengths and limitations of the study? What do the results mean?
Overall, the authors demonstrated that children receiving IBI showed more significant individual gains than those receiving other treatments, or no treatment. There was considerable variability across studies in the size of the gains, but many of the studies contained small samples of participants. Further, the authors identified individual factors that may contribute to treatment success. Individuals receiving high intensity intervention, those with higher IQ, and those with lower adaptive functioning at baseline made the most significant improvements. However, by combining multiple studies to conduct a single analysis the authors cannot ensure the reliability and consistency of treatment implementation and data collection at each site. The possible variability between sites prevents the broader generalization of these findings without further investigation. Also, as the authors pointed out, many of the studies have methodological limitations that diminish confidence in the validity of findings. Nevertheless, the results contribute significantly to knowledge about the effectiveness of IBI by estimating the magnitude of effects and suggesting factors that predict response to intervention.