Implementation of Interventions in Under-Resourced Public Schools: Teacher and Administrator Perceptions of Programs for Students with ASD

Background:  Moving evidence-based practice from research clinics into under-resourced community settings requires considerable effort and resources. ASD interventions rarely are developed with the skills and resources of end user in mind, and rarely take into account the organizational context that will support (or not support) intervention implementation. Working in partnership with community practitioners and the organizations in which they work could enhance the effectiveness and sustainability of autism interventions (Locke, Kratz, Reisinger, & Mandell, 2014). 

Objectives:  To determine, after one year of implementation without outside support, if administrators and implementers (teachers & aides) feelings change regarding a) whether the intervention sustained, b) how school staff feelings change regarding the acceptability of the intervention and c) whether staff identify new barriers to implementation.

Methods:  

The programs of interest are the Schedules, Tools, and Activities for Transitions (STAT) program, a behaviorally based classroom intervention for children with ASD, and the Remaking Recess (RR) intervention, a social engagement intervention for included children with ASD. The STAT and RR interventions were implemented in three urban school districts in Philadelphia, Rochester, and Los Angeles. Including all sites, 78 teachers and 157 students were enrolled in STAT and 106 teachers/aides and 95 students were enrolled in RR. Participating schools were randomized to immediate treatment or a waitlist control.  Study staff collected program fidelity by rating staff adherence to program components throughout treatment. Measures of implementation climate and buy-in were collected at post-treatment and follow-up.

One year after support ended, 7 school administrators and 35 program implementers (classroom aides or teachers) were interviewed using a semi-structured interview protocol.  One interview protocol was developed per intervention.  Separate forms were developed to target the experience of implementers or supervisors. Each protocol contained between 14- 23 questions and each interview was audio recorded and later transcribed for analysis.  Each interview averaged between 30- 60 minutes and was conducted at the participating school or by phone.

For the buy-in measure, teachers rated the degree to which they agreed with statements about the importance of the intervention on a Likert scale (1 = strongly agree; 7 = strongly disagree). For implementation climate, teachers used a Likert scale (1 = not true; 5 = true) on items designed to assess program quality, school support, and global perceptions of climate. Qualitative data are analyzed using NVivo.  Codebook was created through consensus across members of the study team.  Quantitative analysis included descriptive stats of measures of buy-in and implementation climate.

Results:  Data analyses are ongoing.  Teachers report that intervention fidelity and buy-in continue to be relatively high after one year.  Qualitative responses suggest ongoing and new barriers that emerge in the course of the year. 

Conclusions:  Study results suggest that targeted modular interventions for children with ASD can be implemented and sustained in under resourced public schools.  The data also suggest that barriers to implementation must be taken into account when looking to improve community practice.