Abstract
IMPORTANCE: In opioid treatment programs (OTPs), restarting methadone after missed doses is common. Current guidelines recommend significant dose decreases but do not account for maintained tolerance. OBJECTIVE: To evaluate the safety and treatment retention associated with a novel methadone restart protocol. DESIGN, SETTING, AND PARTICIPANTS: This cohort study included participants in a public, safety-net OTP in Colorado, comparing cohorts before (2021) and after (2023) a 2022 protocol change. Eligible participants were patients who restarted methadone after 4 or more missed dosing days. Data were analyzed August 2024 through June 2025. EXPOSURE: Restart protocol incorporating patient-reported nonprescribed opioid use and individualized assessments to determine restart doses. MAIN OUTCOMES AND MEASURES: Protocol implementation was measured by change in dose before vs after restart, comparing preimplementation and postimplementation periods. Primary outcome was patient safety, measured by emergency department (ED) visits within 7 days of restart (all-cause, opioid withdrawal, and overdose-related) and 7-day and 90-day all-cause mortality. The secondary outcome was 90-day treatment retention. Generalized estimating equations were used to calculate adjusted risk ratios (aRRs) comparing preimplementation and postimplementation periods, controlling for demographics and comorbid substance use. Patient and clinician satisfaction and perceived safety were surveyed. RESULTS: A total of 786 patients were included in the preimplementation cohort and 780 in the postimplementation cohort; 202 patients had methadone restarts before implementation (464 restart episodes; 124 male [61.4%]; 148 ages 25 to 44 years [73.3%]; 10 Black [5.0%], 72 Hispanic [35.6%], 112 White [55.4%]) and 195 patients after implementation (489 restart episodes; 116 male [59.5%]; 123 ages 25 to 44 years [63.1%]; 12 Black [6.2%], 60 Hispanic [30.8%], 114 White [58.5%]). Most patients were insured by Medicaid (preimplementation, 164 of 202 patients [81.2%]; postimplementation, 154 of 195 patients [79.0%]). In the preimplementation period, restart doses were 32.8% lower (95% CI, 30.8%-34.7%) than the last dose prior to restart. After protocol implementation, restart doses were 3.4% lower (95% CI, 1.5%-5.3%) than the last received dose. In the preimplementation period, 44 of 464 restarts (9.5%) were followed by ED visits within 7 days, compared with 30 of 489 restarts (6.1%) in the postimplementation period (aRR, 0.61; 95% CI, 0.37-0.98). There were 4 restarts followed by death within 90 days in both the preimplementation and postimplementation periods. Treatment retention was similar in preimplementation and postimplementation periods (aRR, 0.88; 95% CI, 0.73-1.05). Survey responses indicated high acceptance and satisfaction with the protocol. CONCLUSIONS AND RELEVANCE: In this cohort study of individuals with opioid use disorder restarting methadone after a gap in treatment, an individualized methadone restart protocol that considered interim opioid tolerance was associated with higher restart doses without compromising safety or treatment retention. Further research is needed to evaluate if this approach improves treatment outcomes.