AIMS To compare the change in illicit opioid users' risk of fatal drug-related poisoning (DRP) associated with opioid agonist pharmacotherapy (OAP) and psychological support, and investigate the modifying effect of patient characteristics, criminal justice system (CJS) referral and treatment completion.

DESIGN National data linkage cohort study of the English National Drug Treatment Monitoring System and the Office for National Statistics national mortality database. Data were analysed using survival methods.

SETTING All services in England that provide publicly funded, structured treatment for illicit opioid users.

PARTICIPANTS Adults treated for opioid dependence during April 2005 to March 2009: 151,983 individuals; 69% male; median age 32.6 with 442,950 person-years of observation.

MEASUREMENTS The outcome was fatal DRP occurring during periods in or out of treatment, with adjustment for age, gender, substances used, injecting status and CJS referral.

FINDINGS There were 1499 DRP deaths [3.4 per 1000 person-years, 95% confidence interval (CI) = 3.2-3.6]. DRP risk increased while patients were not enrolled in any treatment [adjusted hazard ratio (aHR) = 1.73, 95% CI = 1.55-1.92]. Risk when enrolled only in a psychological intervention was double that during OAP (aHR = 2.07, 95% CI = 1.75-2.46). The increased risk when out of treatment was greater for men (aHR = 1.88, 95% CI = 1.67-2.12), illicit drug injectors (aHR = 2.27, 95% CI = 1.97-2.62) and those reporting problematic alcohol use (aHR = 2.37, 95% CI = 1.90-2.98).

CONCLUSIONS Patients who received only psychological support for opioid dependence in England appear to be at greater risk of fatal opioid poisoning than those who received opioid agonist pharmacotherapy.