Aims The aims of this study were to: examine the number and characteristics of patients entering and re-entering opioid replacement treatment between 1985 and 2006, to examine select demographic and treatment correlates of leaving treatment between 1985 and 2000, and to compare retention rates in methadone and buprenorphine maintenance treatment from 2001 to 2006.
Design A retrospective cohort study using register data from the Pharmaceutical Drugs of Addiction System.
Setting Opioid substitution treatment in New South Wales (NSW), Australia.
Participants A total of n = 42 690 individuals prescribed opioid replacement treatment between 1985 and 2006 in NSW.
Measurements Client characteristics over time, retention in days in first treatment episode, number of episodes of treatment and proportion switching medication.
Findings Overall, younger individuals were significantly more likely to leave their first treatment episode than older individuals. In 2001–06, after controlling for age, sex and first administration point, the hazard of leaving treatment was 1.9 times for those on buprenorphine relative to those on methadone. Retention in treatment varied somewhat across historical time, with those entering during 1995–2000 more likely to leave at an earlier stage than those who entered before that time.
Conclusions Retention in treatment appears to fluctuate in inverse proportion to the availability of heroin. Individuals in contemporary treatment are older users with a lengthy treatment history. This study has provided population-level evidence to suggest that retention in methadone and buprenorphine differ in routine clinical practice. Future work might investigate ways in which patient adherence and retention may be improved.