PUBLIC HEALTH ABSTRACT Medications for Opioid Use Disorder (MOUD), such as methadone, buprenorphine (with or without naloxone), and extended-release naltrexone, have been shown to reduce overdoses and drug use. However, MOUD retention is essential to attaining these outcomes. Unfortunately, only 50% or less of patients remain on MOUD 6 months after initiating treatment, a minimum standard established by the National Quality Forum. One possible approach to improving retention would be using patient characteristics to select the MOUD with which patients are most likely to be successful. This study will utilize VHA data sets to identify patient characteristics associated with improved MOUD retention by MOUD type as well as patient characteristics associated with successful MOUD discontinuation without relapse. We will use a retrospective cohort design using the International Classification of Disease, 9th and 10th edition (ICD-9/10-CM) codes to operationalize OUD since the years of data used for this study cross over the time of ICD-10 implementation (2015). Aim 1 is to identify patient characteristics associated with MOUD retention and opioid overdose death by MOUD type. Our primary outcome for Aim 1 will be a binary measure of retention defined as continuous coverage with the type of MOUD the patient originally initiated for 180 days after the index date without a 28- day gap in treatment. Secondary outcome measures will include (1) a binary measure of retention similar to the primary outcome but over the 365 days after the index date and (2) opioid overdose death in the 365 days after the index date using the VHA Mortality Data Repository. Aim 2 is to quantify the association between probability of relapse after MOUD discontinuation and both length of MOUD therapy and patient characteristics. The outcome for Aim 2 will be discontinuation with vs. without relapse.