Abstract “Unpacking Treatment Mechanisms: Combining Evidence from Three Early Intervention Models for ASD” Though randomized control trials (RCTs) of early intervention for children with autism spectrum disorders (ASD) have shown evidence for the effectives of various treatment models, we know little about the mechanisms of treatment. Specifically, many early intervention models include parent coaching as a core strategy to increase the use of effective interactive strategies in parents to promote social communication in children. However, our knowledge about links between changes in parents' behaviors and changes in children's behaviors are still very limited. Therefore, the primary aim of this project is to combine evidence for treatment effects of early intervention and mediation of changes in parent behaviors on changes in child behaviors across the three RCTs of Early Start Denver Model (ESDM), Early Social Intervention (ESI), and Joint Attention Symbolic Play Engagement and Regulation (JASPER) treatment using innovative statistical techniques. The behaviors of the children and parents over the course of treatment will be measured based on approximately 1200 parent-child interaction videos that are already available over the course of three RCTs. Core ASD symptoms in children will be rated blindly using a newly validated treatment outcome measure, the Brief Observation of Social Communication Change (BOSCC), and parents' use of interactive strategies will be coded using a newly developed coding system of parent strategy use. Leveraging the recent advancement of the Behavioral Signal Processing (BSP) approach, we will also implement novel, objective automated acoustic measures to rate changes in children's and parents' behaviors in response to treatment. Combining data from the three RCTs will result in a sample size of 286 toddlers and young preschoolers from 1 to 4 years of age followed over the course of the three RCTs. The large sample size will increase the power of statistical tests to examine the moderators of treatment and mediation effects. Analyses of the combined data from 3 different studies and 5 different sites will enhance the generalizability of the inferences that can be drawn from our findings. Through this study, we hope to provide insight regarding for whom, how, and why early interventions can change behaviors and trajectories in autism. Findings will inform future development and dissemination of cost-effective early treatment models as well as service and policy level decision making to make the best use of resources for young children with ASD and their families.

The proposed study attempts to examine the moderators and mediators of treatment for young children with autism spectrum disorders (ASD) using cutting-edge measurement and statistical approaches. Answering questions about the mechanisms of treatment for ASD will help us gain insight into the active ingredients of early interventions, and will ultimately inform the development of cost-effective treatment models and dissemination of treatment techniques. Answering the question regarding for whom interventions work will help us individualize treatment programs for children and families with certain characteristics.