Trends in Rates of Opioid Agonist Treatment and Opioid-Related Deaths for Youths in Ontario, Canada, 2013-2021

Key Points Question How have rates of opioid agonist treatment (OAT) and opioid-related deaths in youths changed over the last 9 years and how does this compare to trends among adults? Findings This cross-sectional study of public health data for individuals aged 15 to 44 years in Ontario, Canada, found that rates of opioid-related deaths increased in both youths and adults between 2013 and 2021, yet rates of OAT among youths declined over this time despite rising rates of OAT among adults. Meaning These results suggest that a better understanding of prevalence of opioid use disorder in youths and barriers to accessing OAT is needed.


Introduction
The overdose crisis across North America continues to worsen. While opioid-related deaths concentrate among middle-aged adults, recent data suggests an acceleration in opioid-related mortality among youths. 1,2 Opioid agonist treatment (OAT) is a core response to this crisis as it is an evidence-based treatment shown to reduce opioid-related and all-cause mortality among individuals with opioid use disorder (OUD). Despite recommendation for its use, 3 there are barriers to accessing OAT among youths, including: stigma, burden of witnessed dosing, and lack of availability of youthoriented services and prescribers comfortable treating this population. [4][5][6] Given the rising burden of opioid-related mortality in younger demographics, and the known barriers to accessing OAT in this population, we sought to contrast rates of OAT and opioid-related mortality between youths aged 15 and 24 years and adults aged 25 to 44 years in Ontario, Canada.

Methods
The use of the data in this project was authorized under section 45 of Ontario's Personal Health Information Protection Act (PHIPA) and did not require review by a research ethics board. This study followed reporting guidelines for observational trials as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
We conducted a repeated cross-sectional analysis of rates of OAT (methadone, buprenorphine, or slow-release oral morphine [SROM]; per 1000 population) and opioid-related deaths (per 100 000 population) between 2013 and 2021 in Ontario, the most populous province in Canada.
Opioid-related deaths were obtained in aggregate form from the Public Health Ontario Interactive Opioid Tool. 7 The Public Health Ontario Interactive Opioid tool uses data obtained by the Office of the Chief Coroner and Ontario Forensic Pathology Services. 8 Confirmed and suspected opioidrelated deaths in Ontario are investigated by this office and opioid-related deaths are defined by toxicity caused by a consumed substance where 1 or more substances was an opioid. 9 All unexpected or unnatural deaths must be reported to the coroner for investigation of cause of death, which includes toxicologic testing, as per Canadian law. 10 We used the Narcotics Monitoring System to capture rates of OAT using methods applied in the Ontario Drug Policy Research Network (ODPRN) Ontario Opioid Indicator Tool. 11 The Narcotics Monitoring System was implemented in Ontario in May 2012 to track pharmacy dispensing information about all controlled drugs. OAT dispensing is monitored within this system, regardless of the payment type (eg, privately paid, public drug insurance). This data set was linked using unique encoded identifiers and analyzed at ICES. We used the Statistics Canada population estimates for Ontario by age and sex. 11

Statistical Analysis
We restricted the population for our primary analysis to youths aged 15 to 24 years and constructed a comparator cohort of adults aged 25 to 44 years. This comparator group was selected as the age 25 to 44 years group is the demographic with the largest burden of opioid deaths and access to treatment in Ontario. In secondary analyses, we stratified rates by sex and OAT type. Data on race and ethnicity are not available within the datasets used and therefore could not be reported. For each age-specific cohort and strata, we determined the relative percentage change from 2013 to 2021 for OAT use and opioid-related deaths. All analyses were performed in SAS Enterprise Guide version 7.1 (SAS Institute).

Secondary Analyses
The trends documented among youths were largely consistent when stratified by sex. In general, rates of OAT and opioid-related deaths were higher among males throughout the study period. By 2021, rates of OAT had declined to similar levels among males and females (1.6 vs 1.4 per 1000, respectively), but opioid-related death rates were much higher among males compared with females (15.3 vs 8.9 per 100 000 population, respectively) (Figure 3).
Trends in rates of opioid-related deaths were similar among adults aged 25 to 44 years when stratified by sex, although much higher than rates observed among youths (eFigure in Supplement 1).
Rates of OAT were increasing among both male and female adults (albeit leveling off from 2017 to

Discussion
In this population-based study of youths and adults across Ontario, we found that, despite increasing rates of opioid-related deaths, rates of OAT among youths declined considerably over the 9-year study period, in direct contrast to increasing rates of treatment among adults.
The decline in use of OAT specifically among youths warrants further attention and consideration as barriers in OAT access for youths have long been identified. Particularly for adolescents, there exists ongoing hesitance among clinicians to start OAT, and youths experience significant stigma related to using OAT to treat OUD. 12,13 For example, fewer than 25% of American youths diagnosed with OUD receive OAT (and fewer than 2% of those younger than 18 years). 5,14 There are low rates of OAT initiation even following emergency care for opioid overdoses in this population. 6 This is despite the recommendation for the use of OAT in youths with OUD by the American Academy of Pediatrics in 2016. 4 Challenges in access to care are not limited to youths with OUD but exist more broadly across mental health and substance use disorders. Youths may be more reluctant to seek or engage in care due to difficulties navigating the system, fear of breaches of their confidentiality, and a preference toward self-management of symptoms. 15 A number of strategies have been shown to improve the engagement of youths in care, including youth participation in program development, parental involvement, use of technology, and engagement of schools in screening and intervention. 16 The observed decline in rates of OAT in Ontario youths may also reflect lower underlying rates of OUD in this population. Data on the prevalence of OUD in North American youths are lacking; however, cross-sectional survey data suggest that the prevalence of opioid use is decreasing among

Limitations
This study had several limitations, which included our inability to determine rates of opioid use or OUD diagnoses among youths. We were also unable to determine whether OAT was sought out by youths or the extent of use of other substance use services. Duration on treatment could not be determined in this study; however, our measure provides an assessment of accessibility to treatment (ie, any exposure to OAT), which we believe is an instructive measurement when contrasting against harms. Data on naltrexone were not available as it is not commonly used in the treatment of OUD in Canada and in Ontario, the public drug benefit program only funds naltrexone for alcohol use disorder treatment. Finally, SROM data were not available for 2013.

Conclusions
These findings support the need for a multifaceted effort to change the trajectory of the overdose crisis and its harmful consequences, especially among youths. Specifically, more must be learned about optimizing access to treatment (including OAT) and harm reduction services for youths who use substances.