The effects of extended-release injectable naltrexone and incentives for opiate abstinence in heroin-dependent adults in a model therapeutic workplace: A randomized trial

•Neither XR-NTX nor opiate abstinence incentives affected opiate abstinence among heroin-dependent unemploye adults.•Some participants who received XR-NTX continued to use opiates while under XR-NTX blockade.•Combining XR-NTX and opiate abstinence incentives decreased opiate use. To determine whether extended-release injectable naltrexone (XR-NTX), incentives for opiate abstinence, and their combination reduce opiate use compared to a usual care control and whether the combination reduces opiate use compared to either treatment alone. Randomized 2 × 2 single-site controlled trial conducted from November 2012 through May 2016. After a detoxification and oral naltrexone induction, participants were assigned to a Usual Care, Abstinence Incentives, XR-NTX, or XR-NTX plus Abstinence Incentives group for a six-month intervention period. A model therapeutic workplace where participants could work on automated computer programs that targeted job-skills training for 4 h every weekday for 24 weeks and earn about $10 per hour. 84 heroin-dependent adults who were unemployed and medically approved for naltrexone. Most participants were male (71.4%), African American (80.1%), and cocaine dependent (71.4%). The primary outcome measure was the percentage of urine samples negative for opiates that were collected at once weekly assessments (24 per participant) that were not part of the intervention and for which participants were paid $10 for completing. Participants who attended the workplace provided thrice-weekly urine samples. Abstinence Incentives participants had to provide opiate-free urine samples to maintain maximum pay. XR-NTX participants received one injection every 4 weeks and were required to take injections in order to work and to maintain maximum pay. Usual Care participants were not offered XR-NTX and opiate urinalysis results did not affect pay. A large percentage (65 of 149; 43.6%) of individuals failed the induction protocol required for randomization and to be eligible to receive XR-NTX. When missing urine samples were considered positive, there was no significant interaction between XR-NTX and Abstinence Incentives. XR-NTX plus Abstinence Incentives participants provided significantly more opiate-negative samples (81.3%, SD 39.0%) than XR-NTX participants (64.5%, SD 47.9%; aOR 10.4, 95% CI 1.3–85.5; P = .030). When urine samples were not replaced, there was a significant interaction between XR-NTX and Abstinence Incentives (aOR 77.0, 95% CI 1.3–4432;P