Association Between Increased Dispensing of Opioid Agonist Therapy Take-Home Doses and Opioid Overdose and Treatment Interruption and Discontinuation.

Importance: During the COVID-19 pandemic, modified guidance for opioid agonist therapy (OAT) allowed prescribers to increase the number of take-home doses to promote treatment retention. Whether this was associated with an increased risk of overdose is unclear.

Objective: To evaluate whether increased take-home doses of OAT early in the COVID-19 pandemic was associated with treatment retention and opioid-related harm.

Design, Setting, And Participants: A retrospective propensity-weighted cohort study of 21 297 people actively receiving OAT on March 21, 2020, in Ontario, Canada. Changes in OAT take-home dose frequency were assessed between March 22, 2020, and April 21, 2020, and individuals were observed for up to 180 days to assess outcomes (last date of follow-up, October 18, 2020).

Exposures: Exposure was defined as extended take-home doses in the first month of the pandemic within each of 4 cohorts based on OAT type and baseline take-home dose frequency (daily dispensed methadone, 5-6 take-home doses of methadone, daily dispensed buprenorphine/naloxone, and 5-6 take-home doses of buprenorphine/naloxone).

Main Outcomes And Measures: Primary outcomes were opioid overdose, interruption in OAT, and OAT discontinuation.

Results: Among 16 862 methadone and 4435 buprenorphine/naloxone recipients, the median age ranged between 38 and 42 years, and 29.1% to 38.2% were women. Among individuals receiving daily dispensed methadone (n = 5852), initiation of take-home doses was significantly associated with lower risks of opioid overdose (6.9% vs 9.5%/person-year; weighted hazard ratio [HR], 0.73 [95% CI, 0.56-0.96]), treatment discontinuation (51.0% vs 63.6%/person-year; weighted HR, 0.80 [95% CI, 0.72-0.90]), and treatment interruption (19.0% vs 23.9%/person-year; weighted HR, 0.80 [95% CI, 0.67-0.95]) compared with no change in take-home doses. Among individuals receiving daily dispensed buprenorphine/naloxone (n = 662), there was no significant difference in any outcomes between exposure groups. Among individuals receiving weekly dispensed OAT (n = 11 010 for methadone; n = 3773 for buprenorphine/naloxone), extended take-home methadone doses were significantly associated with lower risks of OAT discontinuation (14.1% vs 19.6%/person-year; weighted HR, 0.72 [95% CI, 0.62-0.84]) and interruption in therapy (5.1% vs 7.4%/person-year; weighted HR, 0.69 [95% CI, 0.53-0.90]), and extended take-home doses of buprenorphine/naloxone were significantly associated with lower risk of interruption in therapy (9.5% vs 12.9%/person-year; weighted HR, 0.74 [95% CI, 0.56-0.99]) compared with no change in take-home doses. Other primary outcomes were not significantly different between groups.

Conclusions And Relevance: In Ontario, Canada, during the COVID-19 pandemic, dispensing of increased take-home doses of opioid agonist therapy was significantly associated with lower rates of treatment interruption and discontinuation among some subsets of patients receiving opioid agonist therapy, and there were no statistically significant increases in opioid-related overdoses over 6 months of follow-up. These findings may be susceptible to residual confounding and should be interpreted cautiously.