ABSTRACT Emerging adults (EAs; ages 18-25) have higher rates of substance use disorders than any other age group and have been hit particularly hard by the opioid crisis. EAs also demonstrate poor adherence to healthcare regimens associated with substance use services, with higher dropout rates and lower service utilization than any other age group. This poor adherence leads to devastating outcomes, including continued substance use, incarceration, and overdose. In addition, high dropout rates contribute to skyrocketing costs to treatment systems as a result of more acute service needs, expensive service utilization, and long waitlists. Cost- effective strategies that are aimed at improving treatment adherence to substance use services and tailored to meet the unique developmental needs of EAs are an imminent need. Further, little is known about risk factors for dropout specific to this age group, hindering effective system responses to this significant problem. At the same time, substance use service systems are increasingly using peer recovery supports (PRS; i.e., paraprofessionals who have “lived experience” with substance use problems) to bolster treatment outcomes without incurring considerable additional costs. However, services delivered by PRS have not been tailored specifically to reduce EA dropout, and few have been rigorously tested at all. The current study will evaluate an innovative EA-specific dropout prevention enhancement to usual treatment services, delivered by PRS in community-based substance use treatment clinics (Aim 1). We will employ a stepped-wedge cluster randomized design, resulting in each clinic having a longitudinal usual services phase and a longitudinal dropout prevention phase. The two phases will be compared on rates of EA dropout and service utilization using objective data from clinical charts. We will also evaluate cost-effectiveness and employ a qualitative approach to understanding the varied financial factors that influence potential sustainability of such a dropout prevention enhancement (Aim 2). In addition, we will leverage the stepped-wedge design to investigate factors purported to predict EA dropout from substance use services and preliminarily investigate whether factors moderate dropout prevention (Aim 3). These key variables include executive functioning, identity formation, motivation, substance use severity, comorbid mental health symptoms, social support, and treatment-related cognitions. In particular, this study will be the first to use a comprehensive assessment of executive functioning, including event-related potential and behavioral data collected during computerized tasks, as a predictor of dropout from substance use services. Results will greatly advance our knowledge of EA dropout and a potential enhancement specifically aimed at reducing EA dropout, which has high potential to be cost-effective and easily disseminated. Answering these key questions is a crucial next step in improving...