Pain and treatment outcomes after initiating methadone vs buprenorphine among medicare patients with opioid use disorder and comorbid chronic pain: A target trial emulation

by Yu-Jung Jenny Wei, Almut G. Winterstein, Roger B. Fillingim, Stephan Schmidt, Siegfried Schmidt

Background

Methadone and buprenorphine, effective treatments for opioid use disorder (OUD), also provide analgesia for managing pain, which is commonly experienced by patients with OUD. Limited population-based evidence exists comparing pain-related and treatment outcomes for methadone versus buprenorphine among patients with OUD and comorbid pain. The study aims to examine pain-related and treatment outcomes among Medicare patients with comorbid pain and OUD who initiated methadone or buprenorphine.

Methods and findings

We conducted a retrospective cohort study with target trial emulation using the 100% Medicare data from 2020 to 2023. Participants included patients with comorbid chronic pain and OUD who initiated methadone or buprenorphine. The key dependent variables were pain-related outcomes that included hospitalization and emergency department (ED) visit due to pain, and treatment outcomes that included opioid overdose and all-cause mortality. Outcomes were assessed 1 year following treatment initiation. Intention-to-treat and per-protocol analyses were conducted to estimate incidence rate ratios (IRRs) for pain-related outcomes and opioid overdose and hazard ratios (HRs) for all-cause mortality. For each outcome, we also calculated the adjusted risk difference (aRD) between the methadone and buprenorphine groups. We identified 49,727 eligible Medicare patients (mean [SD] age, 59.0 [11.6] years; 24,538 [49.3%] female and 25,189 [50.7%] male). Of the identified patients, 16,174 (32.5%) initiated methadone solely administered at opioid treatment programs, and 33,553 (67.5%) initiated buprenorphine primarily prescribed at office-based clinics. Compared with buprenorphine, initiation of methadone was associated with lower adjusted incidence rates of pain-related hospitalization (IRR, 0.64 (95% CI [0.58, 0.70]; P < .001); aRD, −7.2 (95% CI [−8.8 to −5.7]) per 1,000 person-years) and ED visit (IRR, 0.87 (95% CI [0.82, 0.92]; P < .001); aRD, −10.2 (95% CI [−14.4, −5.9]) per 1,000 person-years) in per-protocol analyses, with no difference in opioid overdose (IRR, 1.02 (95% CI [0.93,1.10]; P = .72); aRD, 0.33 (95% CI [−1.5, 2.1]) per 1,000 person-years) and all-cause mortality (HR, 1.06 (95% CI, [0.81–1.39]; P = .66); aRD, 1.1 (95% CI [−1.3, 1.0]) per 1,000 person-years) rates. Similar results were observed in intention-to-treat analyses. Main study limitations included unmeasured confounders and limited generalizability.

Conclusions

This population-based cohort study of Medicare patients with comorbid chronic pain and OUD found that methadone administered at opioid treatment programs is associated with reduced hospitalizations and ED visits for pain-related visits while offering treatment outcomes similar to buprenorphine primarily prescribed at office-based clinics. The favorable pain-related outcomes in patients with methadone should be interpreted with caution, as the finding may reflect differences in the underlying patient population, treatment dosing practices, pharmacological properties, and treatment practice settings, which cannot be measured in Medicare data and merit further investigations.