|Title:||Getting SMART about Social and Academic Engagement of Elementary Aged Students with Autism Spectrum Disorder|
|Principal Investigator:||Kasari, Connie||Awardee:||University of California, Los Angeles|
|Program:||Unsolicited and Other Awards: Special Education Research [Program Details]|
|Award Period:||18 months (7/1/15-12/31/16)||Award Amount:||$199,993|
Co-Principal Investigator: Amanda Gulsrud
Purpose: This purpose of this project is to conduct a small-scale SMART (Sequential, Multiple Assignment, Randomized Trial) to build an adaptive intervention for elementary school students with Autism Spectrum Disorders (ASD) as well as examine the feasibility and acceptability of the intervention. The adaptive intervention will be a sequence of decision rules that specify how and when the intervention practices should change based on student response, thereby adapting the intervention to each student's needs. In this study, four evidence-based practices will be implemented in various sequences, resulting in an adaptive intervention with a personalized set of practices designed to improve social, behavioral, and academic engagement skills for each of the participating students. The SMART research design enables investigators to examine whether and in which sequence the four evidence-based practices can be combined into an adaptive treatment. Information gained from this project will be critical for ensuring that the intervention will be replicable and for understanding the best way to incorporate the intervention into other elementary school settings. The results of this small-scale SMART will be the first step in preparing for a full-scale SMART, with a larger group of students, to build the best possible adaptive intervention.
Project Activities: Researchers will develop and assess the feasibility and acceptability of different intervention sequences using two classroom-level interventions, which target class recess (Remaking Recess) and class recess with in-the-classroom supports (Remaking Recess with Classroom Supports), and two individual student-level interventions, which target individual skills with the help of either parents (Parent-Assisted Home Intervention) or peers (Peer-Mediated School Intervention). In addition, researchers will measure how quickly students respond to practices to determine the best time to begin the implementation of student-level interventions. Specifically, the team will examine (1) progress during classroom-level interventions, (2) transitioning to student-level interventions, (3) identifying children as early (quick) versus slower responders, and (4) providing augmented student-level interventions to slower responders as warranted.
Products: The expected project outcomes will include timely, preliminary information regarding the feasibility and acceptability of the adaptive intervention inclusive of the four evidence-based practices in various sequences for students with ASD in elementary schools. In addition, products include peer review publications and presentations.
Setting: The research will take place in elementary school classrooms and recess settings, and in the students' homes (for one of the proposed intervention practices), in an urban area of California.
Population: The study will be implemented with 32 students with ASD, 5 to 12 years of age, who are fully included in general education classrooms and within the typical range of intellectual functioning. There will be five to six students per school, with six to eight schools participating. The students' teachers, parents, paraprofessionals, and classmates will also participate in this project.
Intervention: There are four practices to be investigated in this study; all are evidence-based but have never been combined or evaluated as part of one adaptive intervention. Remaking Recess (RR) targets behavior and focuses on peer engagement among students during class recess. Classroom Supports (CS) uses visual supports to manage behavior and transition for students in the classroom. Parent-Assisted Home Intervention (Parent) and >em>Peer-Mediated School Intervention (Peer) teach either peers or parents effective strategies to engage children with ASD during recess or during get-togethers or playdates.
Research design: The small-scale SMART includes three phases over the course of one school year. Children will be randomly assigned into either RR or RR with CS in Phase 1 of the study. Two months into treatment, in Phase 2, students will be re-randomized into either the Parent or the Peer intervention. In Phase 3, after 2 additional months, children will be assessed based on their response to treatment. Those categorized as early responders will continue receiving the intervention from Phase 2. Other students who are responding more slowly will receive an augmented intervention with combined parent and peer practices (Parent + Peer). Research assessments will occur at baseline, every 2 months during the school year, and at the 4-month follow up after each intervention sequence ends.
Key measures: The investigators will collect information regarding the acceptability, feasibility, and fidelity of the adaptations of the intervention from the perspective of teachers, other staff, and parents using interview protocols. Academic engagement will be assessed using an observation protocol, and time on task will be assessed with an adaptation of the Behavioral Observation of Students in Schools (BOSS). In addition, investigators will collect child social and engagement outcomes through three key measures: The POPE (Playground Observation of Peer Engagement) measures duration of peer engagement during recess, the Social Network Centrality (SNC) measures students' connections to peers, and the Quality of Play Questionnaire measures the number and quality of interactions.
Data analytic strategy: Primary analyses include a descriptive assessment of acceptability, feasibility, and fidelity of intervention components and sequences overall and by group (teaching staff and parents). In addition, outcomes of the intervention sequences will be assessed based on the responsiveness of the student (e.g., social and engagement outcomes). Exploratory analyses will make the following comparisons: (1) the main effect of each classroom-level intervention practice, (2) the main effects of each individual student-level intervention practice (added to the classroom-level practice received); and (3) the main effect of the augmented intervention for slower responders (added to the class-level and initial student-level intervention received). The data analysis model for all three exploratory comparisons will be a single, piecewise linear mixed model (LMM), with additional hierarchical linear models used to analyze data over time. There will also be descriptive analyses of potential moderators, including characteristics of the participating students and their families as well as the teachers and other staff.