Innovating for real-world care: A systematic review of interventions to improve post-detoxification outcomes for opioid use disorder

Inpatient detoxification is a common health care entry point for people with Opioid Use Disorder (OUD). However, many patients return to opioid use after discharge and also do not access OUD treatment. This systematic review reports on the features and findings of research on interventions developed specifically to improve substance use outcomes and treatment linkage after inpatient detoxification for OUD. Of 6419 articles, 64 met inclusion criteria for the current review. Articles were coded on key domains including sample characteristics, study methods and outcome measures, bias indicators, intervention type, and findings. Many studies did not report sample characteristics, including demographics and co-occurring psychiatric and substance use disorders, which may impact postdetoxification OUD treatment outcomes and the generalizability of interventions. Slightly more than half of studies examined interventions that were primarily medical in nature, though only a third focused on initiating medication treatment beyond detoxification. Medical and combination interventions that focused on initiating medications for OUD generally performed well, as did psychological interventions with one or more reinforcement-based components. Research efforts to improve post-detoxification outcomes would benefit from clearer reporting of sample characteristics that are associated with treatment and recovery outcomes, including diagnostic comorbidities. Findings also support the need to identify ways to introduce medication for opioid use disorder (MOUD) and other effective treatments including reinforcement-based interventions during detoxification or soon after. •Relapse and treatment discontinuation are common after inpatient opioid detoxification.•Work to improve these outcomes has tested medical, psychosocial, and combined approaches.•Medications for opioid use disorder are effective but underrepresented in this work to date.•Many studies do not demonstrate that their study sample represents the treatment population.