Abstract Adolescents who receive treatment for substance use disorders (SUD) have high rates of resuming substance use after they leave treatment. Although relapse prevention is widely incorporated within SUD treatment for adolescents, tools for sustaining recovery are frequently lacking or inadequate. Smartphones have the potential to address this need by delivering Ecological Momentary Assessments (EMA); EMA is a self-monitoring intervention that can be used to help people identify and better understand the antecedents, patterns, and consequences of substance use. In addition, smartphones can provide immediate access to a range of recovery supports delivered via Ecological Momentary Interventions (EMI). As part of the P.I.’s MERIT award (R37 DA011323-14), the research team developed the Smartphone Addiction Recovery Coach for Adolescents (SARC-A) mobile application, which is based on a dynamic model of relapse prevention and utilizes smartphones to deliver recovery-oriented interventions. The applicant team completed two pilot studies in which EMAs and EMIs were delivered to adolescents via smartphones, both separately and combined. Results indicated that EMAs were reliably completed (5 to 6 per day) and accurately predicted the risk of use in the next 7 days. In addition, participants completed an average of 20 or more EMIs per day and EMI utilization was associated with reduced substance use in the next 7 days. Building on these findings, the applicants propose the following experimental study. Adolescents (N=400) will be recruited upon their discharge from 1 of 4 outpatient treatment programs and randomly assigned to 1 of 4 conditions: (1) recovery support as usual control, (2) EMA-only, (3) EMI-only, or (4) EMA+EMI experimental conditions. Interventions will be delivered for 6 months post-discharge with quarterly assessments through 9 months. Data include standardized assessments, urine tests, mobile phone meta data, EMA responses, and EMI utilization. Study hypotheses will be tested using survival analysis, multi-level modeling and structural equation modeling of longitudinal data. The study has the following aims: Aim 1: Test the effects of EMA, EMI, and EMA+EMI (v control) on the frequency of substance use; Aim 2: Evaluate the extent to which the above effects are moderated by baseline substance use frequency; Aim 3: Test the extent to which the frequency of substance use mediates the effects of EMA, EMI, and EMA+EMI on other aspects of recovery including SUD symptoms, HIV risk behavior, quality of life, mental wellness, days of school; and Aim 4: Determine the degree to which EMA responses and EMI utilization predict the duration of abstinence.