BACKGROUND: Retention in methadone maintenance treatment (MMT) is associated with reduced illicit drug use, criminal activity, and mortality; however, many clients move in and out of MMT. This study aims to identify determinants of time to dropout of MMT across multiple treatment episodes in specialist addiction services in Ireland.

METHODS: Cohort study of persons attending specialist addiction clinics between 2010 and 2015. MMT episodes were periods of continuous treatment if there were no interruptions to treatment lasting > 7days. Proportional hazards frailty models were used to assess factors associated with time to dropout from recurrent MMT episodes at 3 (90 days) and 12 months (91-365 days). MMT episodes were right- censored at time of death, transfer to prison or primary care, and study end.

RESULTS: A total of 2,035 individuals experienced 4,969 MMT episodes, with 2,724 dropout events during the six-year follow-up. Factors associated with dropout at 3 months included low dose methadone (<60 mg/day) (HR = 1.49, 95% CI 1.29-1.73) and previous dropout (HR = 1.65, 95% CI 1.41-1.92). Adherence was protective (HR = 0.91, 95% CI 0.90-0.92). Dropout at 12 months was associated with low dose methadone (HR = 1.44, 95% CI 1.23-1.68), previous dropout (HR = 1.37, 95% CI 1.16-1.61), males (HR 1.26, 95% CI 1.06-1.50), benzodiazepines (HR = 1.22, 95% CI 1.03-1.45) and number of comorbidities (HR = 1.12, 95% CI 1.05-1.20); adherence was protective (HR = 0.86, 95% CI 0.84-0.87).

CONCLUSIONS: Clients with a previous history of treatment dropout and those on low dose methadone should be identified as high risk for both early and later dropout. Inversely, adherence to treatment, not missing methadone doses, is protective.