Modeling Implementation: A RE-AIM Hurdle Analysis of Self-Determination and Peer Support Interventions for Autistic Youth

Efforts to improve postschool outcomes require an understanding of what works to improve outcomes, and how to support the adoption, reach, implementation, and maintenance of evidence-based practices (EBPs) in schools. The Propel Project integrated two EBPs—the Self-Determined Learning Model of Instruction (SDLMI) and Peer Support (PS)—into a comprehensive delivery model (SDLMI + PS) spanning general and special education. This study used an implementation science lens to examine the feasibility of adoption and implementation. A sample of Midwestern and Southeastern U.S. high schools ( N = 32) were randomized to SDLMI Only, PS Only, or SDLMI + PS. Hurdle modeling indicated that school-level adoption was high across all groups, yet coordinated implementation across general and special education in the combined SDLMI + PS group proved challenging, with implementation frequently delayed or partial due to scheduling, training, and staffing constraints. On the other hand, cost-effectiveness threshold analysis based on cost data found that assuming only small additive effects for combining interventions ( d = 0.20) suffices for SDLMI + PS to be more cost-effective beyond the single-component arms. This suggests that implementation hurdles identified in this study need to be addressed before we can expect future implementation of coordinated models like SDLMI + PS to produce additive impacts for autistic students. Limitations and implications are discussed.