Optimising the benefits of unobserved dose administration for stable opioid maintenance patients: follow-up of a randomised trial.

Background: The registration of combination buprenorphine/naloxone, a formulation designed to reduce risk of diversion, has led some Australian jurisdictional authorities to allow treatment without direct observation of dosing for stable, opioid-dependent patients.

Aim: To compare two approaches (1) initiating treatment with observed dosing, then allowing patients who demonstrate stability to change to unobserved dosing; or (2) initiating patients with unobserved dosing, subsequently requiring those who fail to stabilize to change to observed treatment.

Methods: This study builds on an RCT comparing efficacy of observed and unobserved treatment at 3 months. At the conclusion of the RCT, clinically "stable" subjects were allocated to continue without observed dosing, while those who did not demonstrate stability were allocated to observed dosing. Subjects were followed for a further 3 months. Primary end-point was retention in treatment.

Results: Of 119 subjects randomised, 70 were retained in treatment to 3 months. Forty-five stable subjects were allocated to unobserved dosing, 25 to observation. Unstable subjects allocated to observed treatment were more likely to drop out thereafter (OR 2.14, 95% CI 1.09-4.19). There was a non-significant trend for people initiated with observed dosing to be better retained during the allocation phase; at 6 months, 13 subjects (22%) from the original unobserved group, and 22 (34%) from the observed group, were retained in treatment (chi2=2.10, 1 df, p=0.15).

Conclusions: Withdrawal of unobserved doses led to marked attrition from treatment. If access to unobserved dosing is to be restricted to stable patients, it appears preferable to initiate dosing with observation and allow unobserved doses for people who successfully stabilize, than to initiate with unobserved doses and transfer unstable patients to observation.