Project DATA: A Multisite Evaluation of a School-based Model for Preschoolers with Autism
Co-Principal Investigator: Ilene Schwartz (University of Washington)
Purpose: The prevalence of autism spectrum disorders (ASD) has risen dramatically in the past decade and children are being identified at earlier ages, putting pressure on school districts to provide effective interventions for these young children. One comprehensive treatment model that aims to address this need for effective early intervention is Project DATA (Developmentally Appropriate Treatment for Autism). This model blends practices from the fields of applied behavior analysis, early childhood education, and early childhood special education.
The purpose of this research is to evaluate Project DATA for preschool children using a randomized controlled trial. The research team will examine the impact of the intervention on child cognition, language, social skills, and behavior. They will also investigate potential child-level moderators as well as family functioning and satisfaction as potential mediators of intervention effects.
Project Activities: The research team will evaluate Project DATA, a comprehensive preschool intervention for children with ASD, using a two-arm randomized trial. Twelve schools will participate, each with at least 10 children with clinical levels of ASD symptoms. Classrooms will be randomly assigned to the experimental or comparison (standard care) condition, and children will be randomized to classrooms. Outcomes of the intervention include child cognition, language, social skills, and behavior. The investigators will also examine whether various child-level variables (e.g., gender, age, severity of ASD symptoms) moderate the impact of the intervention, and whether family functioning and satisfaction with the intervention serve as potential mediators.
Products: The products of this project will include evidence of the efficacy of Project DATA for preschool children with ASD, peer-reviewed publications, and presentations.
Setting: The research will take place in preschool classrooms in Oklahoma and Washington.
Sample: Preschool children with autism spectrum disorders in approximately 12 schools will participate. Each participating school will have a minimum of 10 preschoolers enrolled annually with clinical levels of ASD symptoms.
Intervention: Project DATA is a comprehensive early intervention for preschoolers. The model blends practices from the fields of applied behavior analysis (ABA), early childhood education, and early childhood special education, enabling elements of ABA treatment to be more feasible for use in real-world child settings. Model components include (1) an integrated early childhood experience in a natural setting, including teaching children to interact successfully with other children; (2) extended day, intensive instruction (one-on-one and small group); (3) technical and social support for families; (4) collaboration and coordination across services; and (5) the quality of life influenced curriculum, which teaches children skills for independence and successful integration across school and community settings.
Research Design and Methods: The study will use a two-arm randomized clinical trial with children randomized to classrooms and classrooms randomized to the experimental (Project DATA) or control condition. Randomization will be stratified by state and follow a dynamic allocation procedure that balances groups on key substrata, including gender, age, ASD severity, and IQ. Teachers in the treatment classrooms will receive a week-long intensive training with hands-on experience and monitoring with in vivo coaching. Data will be collected for up to 2 years on the primary cognitive, language, and social outcomes of interest at baseline (pre-randomization) and quarterly (every 3 months) thereafter. Executive functioning and adaptive behavior outcomes will be measured at baseline and at the end of the school year. Fidelity checks will occur monthly, along with visits to provide teachers coaching.
Control Condition: The control condition will consist of usual practices provided in local school district preschool classrooms.
Key Measures: This study will use a variety of standardized and nonstandardized measures. The PDD Behavioral Inventory will be used as a diagnostic assessment. Outcomes will be measured using the Mullen Scales of Early Learning and Preschool Language Scale–4thEdition, Vineland Adaptive Behavior Scale–2nd Edition, Preschool Kindergarten Behavior Scales–2nd Edition, and Behavior Rating Inventory of Executive Function–Preschool. The researchers will also administer surveys of family variables, observational assessments of child engagement and toy play, treatment fidelity checklists, and consumer satisfaction surveys. Finally, interviews and focus groups will be used to obtain data on intervention feasibility and acceptability.
Data Analytic Strategy: Multilevel modeling will be the primary analytic test of intervention efficacy, examining group differences in change over time and structured testing of mediating and moderating influences. For ordinal variables, generalized linear mixed modeling will be used. Qualitative data on feasibility and acceptability of the intervention will be transcribed and coded by thematic category.
Related IES Projects: Evaluation of a Comprehensive Community-based Intervention for Toddlers with ASD (R324A110353)
Schwartz, I. S.; McBride, B. J. (2016). Project DATA (Developmentally Appropriate Treatment for Autism Manual: Blending Approaches to Meet Individual Needs. Paul Brookes Publishing Co.
Journal article, monograph, or newsletter
Schwartz, I.S., Thomas, C.J., McBride, B.J., and Sandall, S.R. (2013). A School-Based Preschool Program for Children With ASD: A Quasi-Experimental Assessment of Child Change in Project DATA. School Mental Health, 5(4): 221–232. doi:10.1007/s12310–013–9103–7