Opioid agonist pharmacotherapy in New South Wales from 1985 to 2006: Patient characteristics and patterns and predictors of treatment retention

  • Lucinda Burns, National Drug and Alcohol Research Centre, UNSW, Australia
  • Dr Louisa Degenhardt, National Drug and Alcohol Research Centre, UNSW, Australia
  • Ms Deborah Randall, National Drug and Alcohol Research Centre, UNSW, Australia
  • Dr Wayne Hall, School of Population Health, University of Queensland, Australia
  • Dr Matthew Law, National Centre in HIV Epidemiology & Clinical Research, UNSW, Australia
  • Dr Tony Butler, National Drug Research Institute, Curtin University of Technology, Australia
  • A/Prof James Bell, Langton Centre, Australia

    • Objective
    To determine factors associated with leaving treatment prior to completion; and compare retention rates in methadone and buprenorphine treatment.
    Methods
    This study used treatment data from the NSW Pharmaceutical Drugs of Addiction System database between 1985 and 2006. Predictors of retention in treatment before the introduction of buprenorhpine were analysed using Cox regression. Differences in retention by medication type (methadone and buprenorphine) were analysed using Cox regression and a time-varying covariate for medication type.
    Results
    In 1985 to 2000: males were significantly more likely to leave their first treatment episode than females (HR: 1.16, 95% CI 1.13-1.19, p<.01); those aged >=40 years were less likely to leave than those <20 years of age (HR: 0.45, 95% CI 0.42-0.49, p<.01); those who started treatment in 1995-2000 were more likely to leave treatment than the first (1985-1989) cohort (HR: 1.24, 95% CI 1.20-1.28, p<.01). Those whose first treatment episode was in a correctional facility were significantly less likely to leave treatment in the first nine months (HR: 0.61 95% CI 0.56-0.65, p<.01) than those who first started in a clinic; but were more likely to leave treatment after nine months (HR: 1.18, 95% CI 1.09-1.28, p<.01). In 2001 to 2006, after controlling for age, sex and first administration point, the hazard of leaving treatment was 1.89 times higher (95% CI: 1.79-1.99, p<.01) for those on buprenorphine than those on methadone.
    Conclusions
    Males and younger clients were more likely to leave treatment early, and treatment retention varied over time most likely in response to heroin availability. There was better retention in first nine months among those entering treatment in prison and poorer thereafter, probably reflecting the higher likelihood of leaving treatment after release from prison. There was poorer retention in the first treatment episode when on buprenorphine compared with methadone.