Distinct patterns of treatment response using the Early Start Denver Model with young children with autism spectrum disorder: a cluster analysis

Evidence-based early intervention approaches have been shown to significantly improve many areas of development in young children with autism. Despite having an overall positive impact for most children, there is high inter-individual heterogeneity of response to treatment, and it is currently not possible to predict which child will benefit from which specific intervention. In this study, we analyze developmental trajectories of young children with autism who received two-years of Early Start Denver Model intervention in Geneva, Switzerland (n=47). Using cluster analyses, we distinguish between 3 subgroups based on cognitive level at baseline and rates of change over time. The first subgroup of children had cognitive scores at baseline only slightly below what would be expected for their age and were found to have nearly no cognitive delay by the end of treatment (High at Baseline: HB). The children in the two other subgroups all presented with severe cognitive delay at baseline, with the second groups cognitive scores improving significantly in the first year of intervention, along with increased communication and daily living skills (High Responders: HR). The third subgroup showed little to no cognitive level change over the course of treatment (Minimal Responders: MR). Furthermore, the rates of change in cognition and adaptative behavior in the HR subgroup were already significantly higher compared to the MR subgroup within the first 6 months of intervention. A rapid and early response to intervention might thus represent a predictor of optimal outcome. Inversely, slower progress by the sixth month of intervention may predict that the child will have a slower response to the treatment overall and may need alternative supports to reach their learning objectives quickly.