Aims The study estimated serious adverse event (SAE) rates among entrants to pharmacotherapies for opioid dependence, during treatment and after leaving treatment.
Design A longitudinal study based on data from 12 trials included in the Australian National Evaluation of Pharmacotherapies for Opioid Dependence (NEPOD).
Participants and settings A total of 1244 heroin users and methadone patients treated in hospital, community and GP settings.
Intervention Six trials included detoxification; all included treatment with methadone, buprenorphine, levo-alpha-acetyl-methadol (LAAM) or naltrexone.
Findings During 394 person-years of observation, 79 SAEs of 28 types were recorded. Naltrexone participants experienced 39 overdoses per 100 person-years after leaving treatment (44% occurred within 2 weeks after stopping naltrexone). This was eight times the rate recorded among participants who left agonist treatment. Rates of all other SAEs were similar during treatment versus out of treatment, for both naltrexone-treated and agonist-treated participants. Five deaths occurred, all among participants who had left treatment, at a rate of six per 100 person-years. Total SAE rates during naltrexone and agonist treatments were similar (20, 14 per 100 person-years, respectively). Total SAE and death rates observed among participants who had left treatment were three and 19 times the corresponding rates during treatment.
Conclusions Individuals who leave pharmacotherapies for opioid dependence experience higher overdose and death rates compared with those in treatment. This may be due partly to a participant self-selection effect rather than entirely to pharmacotherapy being protective. Clinicians should alert naltrexone treatment patients in particular about heroin overdose risks. Duty of care may extend beyond cessation of dosing.