Effects of incentives for naltrexone adherence on opiate abstinence in heroin-dependent adults

Aim: To test whether an incentive-based intervention that increased adherence to naltrexone also increased opiate abstinence. Design: Post-hoc combined analysis of three earlier randomized controlled trials that showed individually that incentives for adherence to oral and to extended-release injection naltrexone dosing schedules increased naltrexone adherence, but not opiate abstinence. Setting: Out-patient therapeutic work-place in Baltimore, MD, USA. Participants: One hundred and forty unemployed heroin-dependent adults participating from 2006 to 2010. Interventions: Participants were hired in a model work-place for 26 weeks and randomized to a contingency (n = 72) or prescription (n = 68) group. Both groups were offered naltrexone. Contingency participants were required to take scheduled doses of naltrexone in order to work and earn wages. Prescription participants could earn wages independent of naltrexone adherence. Measures: Thrice-weekly and monthly urine samples tested for opiates and cocaine and measures of naltrexone adherence (percentage of monthly urine samples positive for naltrexone or percentage of scheduled injections received). All analyses included pre-randomization attendance, opiate use and cocaine use as covariates. Additional analyses controlled for cocaine use and naltrexone adherence during the intervention. Findings: Contingency participants had more opiate abstinence than prescription participants (68.1 versus 52.9% opiate-negative thrice-weekly urine samples, respectively; and 71.9 versus 61.7% opiate-negative monthly urine samples, respectively) based on initial analyses [thrice-weekly samples, odds ratio (OR) = 3.3, 95% confidence interval (CI) = 1.7–6.5, P < 0.01; monthly samples, OR = 2.6, 95% CI = 1.0–7.1, P = 0.06] and on analyses that controlled for cocaine use (thrice-weekly samples, OR = 3.9, 95% CI = 3.3–4.5, P < 0.01; monthly samples, OR = 3.4, 95% CI = 1.1–11.1, P = 0.04), which was high and associated with opiate use. The difference in opiate abstinence rates between contingency and prescription participants was reduced when controlling for naltrexone adherence (monthly samples, OR = 1.1, 95% CI = 0.7–1.7, P = 0.84). Conclusions: Incentives for naltrexone adherence increase opiate abstinence in heroin-dependent adults, an effect that appears to be due to increased naltrexone adherence produced by the incentives.