When to initiate methadone take-home dosing? A population-based target trial emulation

Importance Daily witnessed ingestion in community-based pharmacies is standard practice for methadone maintenance treatment for opioid use disorder in British Columbia (BC), Canada, but this practice may pose a barrier to sustained retention in treatment and its life-saving benefits. Objective To assess the comparative effectiveness of different initiation times of methadone take-home doses on the time-to all-cause mortality and methadone discontinuation. Design and setting We emulated a hypothetical target trial using observational data from BC, Canada, 2010-2022. We conducted both incident (no history of opioid agonist treatments [OAT]) and prevalent new-user (no OAT within the past month) designs. Participants We included individuals aged 18 years and older who completed OAT induction, were not pregnant or incarcerated and had no history of cancer or palliative care. Exposure Exposure strategies were: (i) no take-home dosing; and take-home dose initiation within (ii) 0-4 weeks; (iii) 5-12 weeks; (iv) 13-24 weeks; and (v) 25-52 weeks after completing induction. Main outcomes and Measures Outcomes were time-to- all-cause mortality and methadone discontinuation. Clone-censor-weight approach was used to estimate 78-week adjusted risk difference (aRD). 3Results A total of 9,788 incident users (median [Q1-Q3] age 34.1 [27.2-44.1] years; 32.8% female) and 31,658 prevalent new users (36.9 [29.7-46.2] years; 33.3% female) were included. Compared with ‘no take-home dosing’ in prevalent new-user analysis, aRD (95% CI) for the mortality outcome was -0.9 (-1.5, -0.2), -1.0 (-1.6, -0.3), -0.8 (-1.6, -0.1), and -0.3 (-1.1, 0.4) for the ‘0-4 weeks’, ‘5-12 weeks’, ‘13-24 weeks’, and ‘25-52 weeks’ strategies. For the discontinuation outcome, aRDs were -2.0 (-4.0, -0.1), -5.3 (-6.8, -3.8), -6.0 (-7.2, -4.8), and -4.4 (-5.4, -3.4). Similar patterns were observed among incident users. Multiple sensitivity analyses with sample restrictions, time-zero classifications, timeline restrictions, and addressing residual confounding bias supported the primary results. Conclusions and Relevance Initiating take-home methadone within 5-12 and 13-24 weeks after completing induction resulted in the greatest benefits in reducing mortality and treatment discontinuation, though any take-home dosing was consistently better than no take-home dosing. These client-level benefits warrant consideration against the threat of diversion and subsequent adverse public health outcomes when refining clinical guidelines on take-home dosing.