Abstract Background Opioid Agonist Treatment (OAT) is the most effective intervention for opioid use disorder (OUD), but retention has decreased due to increasingly potent drugs like fentanyl. This cohort can be used retrospectively to observe trends in service utilization, healthcare integration, healthcare costs and patient outcomes. It also facilitates the design of observational studies to mimic a prospective design. Methods This study used linked administrative data from ICES to create a cohort of 137,035 individuals who received at least one prescription of methadone or buprenorphine/naloxone between 2014 and 2022. Data were linked using de-identified personal health numbers. Variables included age, sex, rurality, income, homelessness, and mental health conditions. Regional differences in OAT use, retention, and mortality were analyzed. Results Of the cohort, 56.1% began OAT after 2014. Southern Ontario participants more often started on methadone (53.2%), while Northern Ontario patients favored buprenorphine/naloxone (62.7%). Northern patients were younger, more likely to be female, live in rural areas, and face homelessness. The death rate was higher in Southern Ontario (22.1%) than in Northern Ontario (13.2%). Retention declined over time, with 73.4% of patients remaining in treatment at the study's end. Conclusions The findings highlight regional disparities in OAT delivery and emphasize the need for region-specific strategies, particularly in rural areas, to improve retention and reduce mortality.