Predictors of non-use of illicit heroin in opioid injection maintenance treatment of long-term heroin dependence

Aims: To investigate baseline and concurrent predictors of non-use of illicit heroin among participants randomized to injectable opioids in the North American Opiate Medication Initiative (NAOMI) clinical trial. Methods: NAOMI was an open-label randomized controlled trial comparing the effectiveness of injectable diacetylmorphine and hydromorphone for long-term opioid-dependency. Outcomes were assessed at baseline and during treatment (3, 6, 9, 12 months). Days of non-use of illicit heroin in the prior month at each follow-up visit were divided into three categories: Non-use; Low use (1 to 7 days) and High use (8 days or more). Tested covariates were: Sociodemographics, Health, Treatment, Drug use and illegal activities. Mixed-effect proportional odds models with random intercept for longitudinal ordinal outcomes were used to assess the predictors of the non-use of illicit heroin. Results: 139 participants were included in the present analysis. At each follow-up visit, those with non-use of illicit heroin represented 47.5% to 54.0% of the sample. Fewer days of cocaine use (p = 0.074), fewer days engaged in illegal activities at baseline (p < 0.01) and at each visit (p < 0.01), less money spent on drugs (p < 0.001), days with injection opioid or oral methadone treatment (p < 0.001) and total mg of injectable opioids taken (p < 0.001), independently predicted lower use of illicit heroin. Conclusions: The independent effect of several concurrent factors besides the injection of opioid dose suggests benefits from the clinic that go beyond the provision of the medication alone. Thus, this supervised model of care presents an opportunity to maximize the beneficial impact of medical and psychosocial components of the treatment on improving outcomes associated with non-use of illicit heroin. (C) 2014 The Authors. Published by Elsevier Ltd.