In Australia, maintenance treatment for opioid dependence involves supervised daily administration of a dose of methadone or buprenorphine. A sublingual tablet combining buprenorphine and naloxone in a 4:1 ratio (Suboxone) has been developed, designed to deter diversion and intravenous misuse, and may be suitable for unsupervised administration. The aim of this study was to investigate the tolerability of Suboxone, and investigate whether unsupervised administration can be effective in stabilized patients. Employed patients on buprenorphine maintenance, who had ceased heroin use, were switched to Suboxone and provided with weekly supplies of medication to take without supervised administration. Subjects were monitored closely with weekly clinical reviews, and research interviews at baseline, 3 and 6 months. Only 11% of people receiving buprenorphine met eligibility criteria. Seventeen subjects were recruited. Fifteen were retained for the full 6 months. No subject appeared destabilized by unsupervised dosing. Suboxone was well tolerated. The current trial demonstrated that unsupervised administration with regular clinical monitoring can be effective in selected patients. However, using access to unsupervised dosing to promote abstinence from heroin probably limits the potential benefits of unsupervised administration to a very small proportion of patients.