Abstract
Background and Objectives
History of nonfatal overdose (NFO) is common among people who use opioids, but little is known about opioid agonist treatment (OAT) outcomes for this high‐risk subpopulation. The objective of this study was to investigate the relative effectiveness of buprenorphine/naloxone and methadone on retention and suppression of opioid use among individuals with opioid use disorder (OUD) and history of NFO.
Methods
Secondary analysis of a pan‐Canadian pragmatic trial comparing flexible take‐home buprenorphine/naloxone and supervised methadone for people with OUD and history of NFO. Logistic regression was used to examine the impact of OAT on retention in the assigned or in any OAT at 24 weeks and analysis of covariance was used to examine the mean difference in opioid use between treatment arms.
Results
Of the 272 randomized participants, 155 (57%) reported at least one NFO at baseline. Retention rates in the assigned treatment were 17.7% in the buprenorphine/naloxone group and 18.4% in the methadone group (adjusted odds ratio [AOR] = 0.54, 95% CI: 0.17–1.54). Rates of retention in any OAT were 28% and 20% in the buprenorphine/naloxone and methadone arms, respectively (AOR = 1.55, 95% CI: 0.65–3.78). There was an 11.9% adjusted mean difference in opioid‐free urine drug tests, favoring the buprenorphine/naloxone arm (95% CI: 3.5–20.3; p = .0057).
Conclusions and Scientific Significance
Among adults with OUD and a history of overdose, overall retention rates were low but improved when retention in any treatment was considered. These findings highlight the importance of flexibility and patient‐centered care to improve retention and other treatment outcomes in this population.
INTRODUCTION
Across North America, the toxic drug crisis is escalating with overdose mortality being a leading cause of unintentional death. 1 In Canada, this crisis is now primarily driven by the unpredictability of the unregulated drug supply and polydrug toxicity, including the combination of opioids with other substances such as stimulants and benzodiazepines. 2 , 3 Although many innovative public health responses have been implemented in recent years (e.g., take‐home naloxone, drug checking services, supervised consumption sites, etc.), 4 the changes in the composition of the drug supply have serious implications for overdose risk and complicated the overdose response interventions that primarily target opioids, 5 leading to a rising number of fatal and nonfatal overdoses (NFO) in many Canadian settings.
NFO are more common than fatal overdoses and are associated with an array of health‐related complications, including hypoxia‐related brain injury, seizures, nerve damage, kidney failure, and changes in cognitive and physical functioning, all requiring increased healthcare utilization. 6 , 7 Previous studies show a dose‐dependent relationship between nonfatal and fatal overdose, with individuals surviving an overdose being at higher risk of subsequent fatal overdose in the following year. 8 , 9 However, evidence shows that individuals enrolled in opioid agonist treatment (OAT) have a significantly lower risk of overdose while receiving treatment, highlighting the importance of engaging those with a history of NFO in OAT. 10 , 11 Additionally, qualitative evidence suggests that following a NFO event, individuals are often willing to engage in risk reduction behaviors, including accessing treatment for their substance use disorder. 12 Thus, it is critical to understand how to provide the most effective treatment for this population.
Numerous studies have evaluated the relative effectiveness of buprenorphine/naloxone versus methadone for retaining patients in treatment. 13 , 14 , 15 However, the majority of these studies have been conducted among individuals who primarily use heroin and have followed strict dosing regimens that do not account for the differing programmatic requirements for buprenorphine/naloxone (generally take‐home dosing) versus methadone (often supervised ingestions). It is unclear whether the results from these studies translate to those who use prescription and synthetic opioids (including pharmaceutical and illicitly manufactured fentanyl) under real‐world treatment conditions. Therefore, the objective of the current study was to determine the relative effectiveness of flexible take‐home dosing of sublingual buprenorphine/naloxone compared to standardized supervised methadone models of care on OAT retention and opioid use among individuals with opioid use disorder (OUD) and a history of NFO enrolled in a pragmatic trial.
METHODS
Study design
This is a secondary analysis from the OPTIMA trial, a multicenter, open‐label, two‐arm, randomized trial evaluating the relative effectiveness of flexible take‐home dosing buprenorphine/naloxone compared to supervised methadone models of care for individuals with OUD. 16 The trial recruited participants from seven hospitals and community‐based clinics in four Canadian provinces, including British Columbia, Alberta, Ontario, and Quebec, between October 2017 and March 2020. All participant follow‐up was completed in July 2020.
Eligibility
Study design, including eligibility criteria, are described in detail elsewhere. 16 Briefly, OPTIMA enrolled treatment‐seeking adults (18–64 years old) meeting DSM‐5 criteria for moderate to severe OUD attributed to prescribed or nonprescribed morphine, tramadol, oxycodone, hydromorphone, and/or fentanyl, who were willing to be randomized to either methadone or buprenorphine/naloxone, able to provide written and informed consent and comply with study procedures. Individuals were excluded if they reported heroin as the most frequently used opioid in the last 30 days, were pregnant or breastfeeding, were enrolled in OAT in the 30 days before initiation, were taking medications that interact with either of the study medications, had a previous adverse reaction to study medications or were pending legal action that would interfere with study completion.
The sample for the present analysis was restricted to all randomized participants who reported a lifetime history of NFO during screening. Data regarding history of NFO was missing for one participant and four were missing baseline urine drug test data resulting in a total analytic sample of 154 participants with a lifetime history of NFO. In a subanalysis, we further restricted the sample to randomized participants with a recent history of NFO (i.e., in the last 6 months before screening) resulting in a sample of 83 participants.
Procedures
Recruitment for the study primarily took place through referral from physicians and clinical staff, word of mouth, advertisement and other offsite methods like community outreach and local organizations that support people who use drugs. Eligible participants were randomized, stratified by site and the presence of lifetime heroin use, in a 1:1 ratio to either receive methadone or buprenorphine/naloxone via witnessed ingestion or flexible take‐home dosing, respectively. All randomized participants met with a study physician to discuss their treatment plan and initiation procedures. Treatment initiation took place within 2 weeks of randomization, or it was considered a treatment initiation failure. The first dose was supervised for all participants at the clinic or pharmacy and medications were dispensed at study sites or community pharmacies for the remainder of the study period. All participants attended follow‐up research visits every 2 weeks for 24 weeks and provided urine samples, self‐reported substance use information and updated demographic information. Participants were compensated up to $560 ($40/visit) over the 24 weeks. The study followed local and international ethical guidelines and each clinical site's Ethics Committee approved the study. The trial was registered to clinicaltrials.gov (NCT03033732).
As the trial was pragmatic in nature, treatment induction procedures, titration speed, dosages, and dispensation frequency were based on clinical judgment and followed national and provincial guidelines for OUD and Health Canada product monographs. 17 Generally speaking, participants in the methadone arm began with daily witnessed ingestion of a maximum oral dose of 30 mg on the first day. Dosages were gradually titrated (i.e., 5–10 mg/day every four or more days), with maintenance doses falling in the 60–120 mg/day range. In certain cases, take‐home doses were permitted after 2–3 months.
Most participants initiated buprenorphine/naloxone at 4 mg/1 mg dose with additional doses administered on the first day (up to 12 mg/3 mg) if needed. In the following days, titration continued up to a recommended daily dose of 24 mg/6 mg. For participants that were clinically stable, take‐home doses of 1‐week carries were recommended within 2 weeks of treatment initiation and 2‐week carries within 4 weeks of treatment initiation. Following treatment initiation, participants were permitted to switch OAT if indicated by clinical circumstances or patient preference. Participants who experienced symptoms of withdrawal before or during treatment initiation were treated with ancillary medications at the physician's discretion.
Measures
The primary exposure was the assigned treatment, either methadone or buprenorphine/naloxone. The primary outcome measure was retention in the assigned treatment, defined as having an active prescription for the assigned OAT at Week 24 and having a positive urine sample result for the assigned OAT (i.e., buprenorphine or EDPP) at Week 24. Opioid use, measured by the proportion of opioid‐free urine drug tests (UDT, excluding the assigned study medication) during the 24 weeks, was a secondary outcome. UDT data were collected biweekly after treatment initiation for a maximum of 12 samples per participant. We tested for the presence of the following opioids: morphine, oxycodone, fentanyl, tramadol, 6‐monoacetylmorphine and hydromorphone, methadone and buprenorphine/naloxone using a Rapid Response Multi‐Drug Once Step Screen Test Panel and single test strips. Missing UDT data were considered opioid positive in the primary analysis.
Based on existing literature describing factors that impact both type of treatment and retention, 18 , 19 a number of variables were considered for inclusion in bivariable analyses. The variables included were: age (in years); biological sex (male vs. female); ethnicity (white vs. Black, Indigenous and People of colour [BIPOC]); homelessness (yes vs. no); receiving income assistance in the month before enrollment (yes vs. no); geographical location (Ontario and Quebec [East Coast]) versus Alberta and British Columbia ([West Coast]); positive UDT for opioids, including morphine, oxycodone, heroin, hydromorphone, fentanyl, unprescribed methadone, or buprenorphine (yes vs. no); positive UDT for stimulants, including cocaine and amphetamine/methamphetamine (yes vs. no). All variables refer to screening or baseline assessments before randomization and treatment initiation.
Statistical analyses
Group differences between participants randomized to buprenorphine/naloxone or methadone were first assessed using Pearson's χ 2 test (or Fisher's exact test for small cell counts) for categorical variables and Mann–Whitney U test for continuous variables. Next, we used descriptive statistics to determine the median number of days spent in treatment, the number of individuals that dropped out of treatment in the first 30 days, OAT dosing and number of take‐home doses. We then investigated the effect of buprenorphine/naloxone treatment on retention relative to methadone using multivariable logistic regression. Multivariable logistic regression models were adjusted for the stratification variables, lifetime heroin use and clinical site, and other confounding variables were excluded based on the Akaike Information Criterion and effect size. We conducted a subanalysis, using the same approach as described above but restricted to participants with a recent NFO.
Finally, we calculated the unadjusted and adjusted mean difference in opioid‐free UDT between the buprenorphine/naloxone and methadone groups and its 95% confidence interval (α = 0.05, two‐sided), using an Analysis of Covariance (ANCOVA), adjusting for stratification variables clinical site, and lifetime heroin use. Shapiro–wilk and Brown–Forsythe tests were used to verify the assumptions of normality and variance homogeneity and Tukey's honest significant test was used to adjust for multiple comparisons. All statistical analyses were performed using R Studio version 1.4.1106. All p values are two‐sided.
Sensitivity analyses
We performed sensitivity analyses on the primary outcome using an alternative definition: having a prescription at the end of treatment and positive UDT for any OAT including methadone, buprenorphine/naloxone, diacetylmorphine, long‐acting morphine, or hydromorphone. In the secondary analysis, we analyzed the mean difference in opioid use (i.e., secondary outcome) using two alternative approaches: (1) excluding all missing UDT data, (2) only including data for participants that were retained in the assigned treatment at 24 weeks.
RESULTS
Sample characteristics
A total of 154 (57% of 272 randomized OPTIMA participants) participants reported at least one NFO in their lifetime and had valid baseline UDS results. Of these, 76 (49%) were randomized to methadone and 78 (51%) to buprenorphine/naloxone. Baseline characteristics of the entire sample, stratified by treatment arm are presented in Table 1. At baseline, the median age was 38 (interquartile range [IQR]: 32–46), 58 (38%) participants were female and 91 (59%) self‐identified as white. Seventy‐five (49%) participants reported experiencing homelessness and 80 (52%) participants received income assistance in the last month. The majority of participants were enrolled in sites located in British Columbia or Alberta [West Coast (109; 71%)] and were positive for opioids (148; 96%) and stimulants at baseline (122; 79%). Of participants that were positive for opioids, 111 (72%) were positive for fentanyl at baseline. There were no statistically significant differences between the two treatment groups at baseline for all variables examined.
Table 1.
Baseline characteristics of participants with a history of nonfatal overdose, stratified by assigned treatment arm.
| Opioid agonist treatment, n (%) | ||||
|---|---|---|---|---|
| Total, n (%) | Methadone | Buprenorphine | ||
| (n = 154) | (n = 76) | (n = 78) | p Value | |
| Age, median (IQR) | 38 (32–46) | 38 (31–44) | 37.5 (32–47) | .859 |
| Sex | ||||
| Male | 96 (62.3) | 47 (61.8) | 49 (62.8) | .999 |
| Female | 58 (37.7) | 29 (38.2) | 29 (37.2) | |
| Ethnicity | ||||
| White | 91 (59.1) | 44 (57.9) | 47 (60.3) | .893 |
| BIPOC | 63 (40.9) | 32 (42.1) | 31 (39.7) | |
| Homelessness | .875 | |||
| No | 79 (51.3) | 38 (50.0) | 41 (52.6) | |
| Yes | 75 (48.7) | 38 (50.0) | 37 (47.4) | |
| Income assistance | ||||
| No | 74 (48.1) | 40 (52.6) | 34 (43.6) | .336 |
| Yes | 80 (51.9) | 36 (47.4) | 44 (56.4) | |
| Clinical sitea | ||||
| East Coast | 45 (29.2) | 19 (25.0) | 26 (33.3) | .337 |
| West Coast | 109 (70.8) | 57 (75.0) | 52 (66.7) | |
| Lifetime heroin use | ||||
| Absence | 33 (21.4) | 17 (22.4) | 16 (20.5) | .933 |
| Presence | 121 (78.6) | 59 (77.6) | 62 (79.5) | |
| Positive UDS for opioids | ||||
| No | 6 (3.9) | 2 (2.6) | 4 (5.1) | .701 |
| Yes | 148 (96.1) | 74 (97.4) | 74 (94.9) | |
| Positive UDS for stimulantsb | ||||
| No | 32 (20.8) | 15 (19.7) | 17 (21.8) | .908 |
| Yes | 122 (79.2) | 61 (80.3) | 61 (78.2) | |
| Recent history of NFO | ||||
| No | 71 (46.1) | 34 (44.7) | 37 (47.4) | .862 |
| Yes | 84 (54.2) | 42 (55.3) | 41 (52.6) | |
Abbreviations: BIPOC, Black, Indigenous and People of Colour; IQR, interquartile range.
West Coast: British Columbia and Alberta, East Coast: Quebec and Ontario.
Positive UDS for stimulants includes: Methamphetamine/amphetamines and cocaine.
The median number of days in treatment for those randomized to methadone was 163 days (IQR: 30–168) and 161 days (IQR: 22–168) for those randomized to buprenorphine/naloxone (N = 120). A total of 32 (27%) participants dropped out in the first 30 days after treatment initiation [of these, 17 (53%) were in the buprenorphine/naloxone and 15 (47%) in the methadone arm].
OAT dosing and dispensation
A total of 116 participants (75%) with a history of NFO had OAT dosing data following randomization (i.e., initiated treatment). Within this subset of participants, the median methadone dose taken during induction was 30 mg/day (IQR: 25–34 mg) and five participants were instructed to return for additional dose adjustments on induction day. The median buprenorphine dose taken during induction was 2 mg/day (IQR: 0.5–4 mg) and 26 participants were instructed to return for dose adjustments on the day of induction. The median maximum dose taken during the study was 60 mg/day (IQR: 30–100 mg) for methadone and 24 mg/day (IQR: 24–40 mg) for buprenorphine/naloxone. Among participants that were retained in any OAT, the median maximum dose was 95 mg/day (IQR: 35–138) for methadone and 32 mg/day (IQR: 21–60) for buprenorphine.
The proportion of participants initiating treatment who received take‐home doses was 9 (11%) in the methadone group and 41 (55%) in the buprenorphine/naloxone group. Among those who received take‐home doses, the median number of doses prescribed was two (IQR: 1–4) for methadone and three (IQR: 0.5–5) for buprenorphine/naloxone. The median maximum number of consecutive days of take‐home doses was three (IQR: 3–13 days) for the methadone and seven (IQR: 6–14 days) for the buprenorphine/naloxone group.
Treatment retention
Table 2 displays results of the bivariable and multivariable logistic regression analyses of the effect of assigned OAT on retention in treatment at 24 weeks among participants with a lifetime history of NFO. For retention in the assigned treatment, 14 (17.7%) participants were retained in the buprenorphine/naloxone group and 14 (18.4%) in the methadone group (adjusted odds ratio [AOR] = 0.54, 95% CI: 0.17–1.54). When retention in any OAT was examined, 22 (27.8%) participants were retained in the buprenorphine/naloxone group and 15 (19.7%) in the methadone group (AOR = 1.22, 95% CI: 0.51–2.96).
Table 2.
Logistic regression analyses for the association between methadone vs. buprenorphine/naloxone and retention in treatment among participants with a lifetime history of nonfatal overdose and opioid use disorder.
| Retention in treatment (n = 154) | ||||
|---|---|---|---|---|
| No, n (%) | Yes, n (%) | Odds ratio (95% CI) | Adjusted odds ratioa (95% CI) | |
| Retention in assigned treatment | ||||
| Methadone | 62 (81.6) | 14 (18.4) | Ref | Ref |
| Buprenorphine/naloxone | 65 (82.3) | 14 (17.7) | 0.95 (0.42–2.18) | 0.54 (0.17–1.54) |
| Retention in any treatment | ||||
| Methadone | 61 (80.3) | 15 (19.7) | Ref | Ref |
| Buprenorphine/naloxone | 57 (72.2) | 22 (27.8) | 1.57 (0.75–3.37) | 1.22 (0.51–2.96) |
Abbreviation: CI, confidence interval.
Odds ratio is adjusted for lifetime heroin use and clinical site.
Table 3 presents results of our subanalysis restricted to the 83 participants with a recent NFO event. For retention in the assigned treatment, 3 (7.1%) participants were retained in the buprenorphine/naloxone group compared to 10 (23.8%) in the methadone group (AOR = 0.13, 95% CI = 0.01–0.64). When retention in any OAT was examined, 9 (21.4%) participants were retained in the buprenorphine/naloxone group and 10 (23.8%) in the methadone group (AOR = 0.71, 95% CI: 0.21–2.30).
Table 3.
Logistic regression analyses for the association between methadone vs. buprenorphine/naloxone and retention in treatment among participants with a recent history of nonfatal overdose and opioid use disorder.
| Retention in treatment (n = 83) | ||||
|---|---|---|---|---|
| No, n (%) | Yes, n (%) | Odds ratio (95% CI) | Adjusted odds ratioa (95% CI) | |
| Retention in assigned treatment | ||||
| Methadone | 32 (76.2) | 10 (23.8) | Ref | Ref |
| Buprenorphine/naloxone | 39 (92.9) | 3 (7.1) | 0.25 (0.05–0.88) | 0.13 (0.01–0.64) |
| Retention in any treatment | ||||
| Methadone | 32 (76.2) | 10 (23.8) | Ref | Ref |
| Buprenorphine/naloxone | 33 (78.6) | 9 (21.4) | 0.87 (0.31–2.44) | 0.71 (0.21–2.30) |
Note: Bolded values are statistically significant, cutoff value for significance is p < .05.
Abbreviation: CI, confidence interval.
Odds ratio is adjusted for lifetime heroin use and clinical site.
Opioid‐free urine drug tests
Figure 1 displays the proportion of opioid‐negative UDT during the study period, stratified by treatment arm. Among participants with a history of NFO, the buprenorphine/naloxone arm had a statistically significantly higher proportion of opioid‐free UDT compared to methadone across all analyses (Table 4). For the total sample (n = 154), including missing UDS data that were considered positive, the mean proportion of opioid negative UDT (±SD) was 19.3 (±32.2)% in the buprenorphine/naloxone group and 7.4 (±20.0)% in the methadone group (adjusted mean difference = 11.9%, 95% CI = 3.5–20.3, p = .0057). When missing UDT data were excluded, the mean proportion of opioid negative UDT were 36.2 (±41)% in the buprenorphine/naloxone group and 14.8 (28.7)% in the methadone group (adjusted mean difference = 21.4%, 95% CI = 12.4–30.4, p ≤ .001). Finally, when we restricted the sample to participants who were retained in the assigned treatment (n = 29), the mean proportion of opioid negative UDT were 73.8 (±20.1)% in the buprenorphine/naloxone arm and 30.0 (±43.8)% in the methadone arm (adjusted mean difference = 43.8, 95% CI = 26.3–61.3, p ≤ .001).
Figure 1.
Percentage of opioid‐free urine drug tests for participants receiving buprenorphine/naloxone or methadone across the 24‐week intervention period, including missing urine drug tests as positive for opioids.
Table 4.
Mean difference in the proportion of opioid‐free urine drug screens in the buprenorphine/naloxone and methadone treatment arms among participants with a history of nonfatal overdose and prescription‐type opioid use disorder.
| % negative UDS for opioids | ||||||
|---|---|---|---|---|---|---|
| Buprenorphine/naloxone | Methadone | |||||
| Analyses | Mean | SD | Mean | SD | Adjusted mean differencea | 95% CI |
| Total (i.e., missing considered as positive for opioids) | 19.3 | 32.2 | 7.4 | 20 | 11.9 | 3.5–20.3 |
| Based on available UDS | 36.2 | 41 | 14.8 | 28.7 | 21.4 | 12.4–30.4 |
| With participants retained on assigned OAT | 73.8 | 20.1 | 30 | 33.7 | 43.8 | 26.3–61.3 |
Abbreviations: CI, confidence interval; OAT, opioid agonist treatment; SD, standard deviation; UDS, urine drug screen.
OR is adjusted for clinical sites and lifetime heroin use.
DISCUSSION
The results indicate that the overall levels of retention among people with OUD and a history of NFO initiating supervised methadone or flexible take‐home dosing buprenorphine/naloxone as part of a pan‐Canadian pragmatic trial were low. There were no significant differences between buprenorphine/naloxone and methadone groups in retention in assigned or any OAT. The proportion of opioid‐free urine drug tests over the study period was significantly higher among participants randomized to buprenorphine/naloxone across all sensitivity analyses. These findings add to the limited evidence on OAT outcomes among individuals with a history of NFO within the context of the current toxic drug crisis.
The retention rates observed in this analysis are significantly lower than those reported in other settings and among all participants in the OPTIMA tral. 18 , 20 Several factors may explain the lower retention rates in this study. First, a history of NFO has been shown to predict treatment dropout at 90 and 120 days, aligning with our finding that nearly a third of participants dropped out within the first 30 days. 21 , 22 Second, most participants tested positive for opioids and stimulants during baseline assessments, and concurrent stimulant use may affect OAT success, as documented in other studies. 23 , 24 Most importantly, the pragmatic design, which reflected real life conditions, did not select participants most likely to be retained, undertake procedures to artificially boost retention, or exclude those with significant comorbidities. Initial stabilization on OAT often requires multiple attempts and individualized dosing protocols, complicating retention efforts. Interventions to improve retention in various stages of the treatment system exist in other clinical areas (e.g., HIV/AIDS, diabetes) and should be implemented and evaluated for persons with OUD. 19
In line with prior literature, 13 , 25 post hoc descriptive analyses revealed that participants with a history of NFO and retained in OAT had higher maximum median OAT doses compared to those who were not retained in treatment. These findings may suggest that insufficient OAT dosing is a clinically relevant factor that contributes to treatment outcomes, including retention. Although not explored in the present study, the rate of titration and take‐home doses may also play a critical role in retention among participants with a history of NFO and is a critical next step in this line of research.
The fact that we failed to find a significant difference in retention in any OAT between treatment arms are similar to those among the whole OPTIMA sample, 20 and confirm the potential benefits of stepped care strategies initially described in a study by Kakko and colleagues among people using heroin. 26 Specifically in the randomized trial, people who initiated buprenorphine/naloxone were allowed to switch to methadone if unsuccessful with buprenorphine, with findings indicating that both approaches (stepped care vs. initial methadone) resulted in similar treatment retention rates. Altogether, our findings support current guidelines for the management of OUD that transitions between OAT medications should be available for people with a history of NFO if indicated by clinical circumstances or for individuals unsatisfied with their current treatment. 17
We found that retention in the assigned treatment among those with a recent history of NFO (i.e., last 6 months) was higher in participants randomized to methadone, although these results should be interpreted cautiously as there were high attrition rates and only a small proportion of participants were retained in both buprenorphine/naloxone and methadone groups. Methadone may be superior for this high‐risk population given the ease of induction relative to buprenorphine/naloxone. 27 Low‐dose inductions of buprenorphine/naloxone are becoming more common in clinical practice and may improve retention as it reduces the risk of precipitated withdrawal and does not require the patient to abstain from opioid use for extended periods of time before OAT initiation. 28 Updated Canadian guidelines for the clinical management of OUD state that low‐dose inductions may be preferable for patients who use fentanyl and other synthetic opioid analogues, given the unpredictability of withdrawal. 17 , 29 , 30 Additional research is needed to determine optimal induction dosing, titration and maintenance dosing protocols for populations using potent opioids who are at the highest risk of future overdose.
This analysis found that participants with a history of NFO randomized to buprenorphine/naloxone had a higher proportion of opioid‐negative UDTs over the 24‐week study period than those on methadone, consistent across all sensitivity analyses. This aligns with previous studies in general OUD populations showing buprenorphine/naloxone's superiority in reducing unregulated opioid use. 31 , 32 This may again be explained by buprenorphine/naloxone being a partial agonist which better blocks the effects of opioids and may encourage abstinence from ongoing unregulated use. 17 However, many participants in both groups continued to use unregulated opioids, indicating current interventions may be insufficient for those with a history of NFO. Higher dosing regimens or alternative pharmacotherapies like injectable OAT and slow‐release oral morphine may be warranted for better treatment outcomes. 33 , 34 For individuals who continue to use opioids, evidence‐informed harm reduction strategies are essential to prevent opioid‐related morbidity and mortality in high‐risk populations.
A number of limitations should be noted when interpreting findings from the current study. First, participant history of NFO was self‐reported which may have been impacted by recall bias, social desirability or underreporting. Second, all missing or unavailable UDS data were classified as positive for the opioid positive UDS outcome which may have resulted in an over‐estimation of the proportion of opioid positive UDS. However, sensitivity analyses revealed that even when missing UDS data were excluded the results were consistent. Third, the study was restricted to participants with a history of NFO at seven Canadian sites, therefore findings may not be generalizable to those that fall outside these jurisdictions. Fourth, due to a small sample size we were only able to consider a small number of variables that could potentially impact the relationship between treatment type and retention.
In conclusion, this secondary analysis of a clinical trial for the treatment of OUD demonstrated that overall retention rates in the assigned or any treatment were alarmingly low among those with a history of NFO and did not differ between those randomized to buprenorphine/naloxone and methadone. Opioid use remained high in both treatment groups, but those randomized to buprenorphine/naloxone had a higher proportion of opioid‐free UDT. These findings highlight the importance of allowing patients to switch OAT medications to determine which one works best based on their circumstances and flexibility with OAT dosing and take‐home dosing. They also highlight the need for alternative interventions that target treatment retention among people with a history of NFO, particularly in the early stages of treatment. These interventions may include increasing the accessibility of alternative pharmacotherapies, concurrent psychosocial interventions provided alongside OAT and connection to additional supports to address other social determinants of health among this high‐risk population.
AUTHOR CONTRIBUTIONS
Hannah Crepeault: Conceptualization; methodology; data curation; formal analysis; investigation; writing—original draft; writing—review and editing. Lianping Ti: Conceptualization; methodology; validation; writing—review and editing. Paxton Bach: Validation; methodology, writing—review and editing Evan Wood: Funding acquisition; resources; validation; writing—review and editing. Didier Jutras‐Aswad: Funding acquisition; investigation; project administration; resources; validation; writing—review and editing Bernard Le Foll: Investigation; project administration; resources; validation; writing—review and editing. Ron Lim: Investigation; project administration; resources; validation; writing—review and editing. Maria E. Socias: Conceptualization; funding acquisition; investigation; methodology; project administration; resources; validation; writing—review and editing.
CONFLICT OF INTEREST STATEMENT
M. E. S. has also received partial support from Indivior's Investigator Initiated Study program for work outside this study. E. W. is a physician who works for Vancouver Coastal Health in the area of withdrawal management and undertakes work in the area of occupational addiction medicine. E. W. is also a professor of medicine based at the University of British Columbia (UBC), a position supported by a Canadian Institutes of Health Research (CIHR) Tier 1 Canada Research Chair, and has received salary support from an R01 from the US National Institute on Drug Abuse, paid to UBC. E. W.'s research lab is further supported by CIHR grants to the Canadian Research Initiative in Substance Misuse. E. W. has also undertaken consulting work in legal matters related to substance use disorders and for a mental health company called Numinus Wellness, where E. W. is former chief medical officer; E. W. has also received compensation in the form of equity in Numinus. E. W. reports receiving honoraria for non‐industry related academic lectures and conference presentations. E. W. has also received payment for expert reports and expert testimony in legal matters pertaining to substance use disorder, including from the Canadian Medical Protective Association and from trade unions representing workers with possible substance use disorder. E. W. has received travel support from the CIHR. B. L. F. has obtained funding from Pfizer Inc. (GRAND Awards, including salary support) for investigator‐initiated projects. B. L. F. has obtained funding from Indivior for a clinical trial sponsored by Indivior. B. L. F. has in‐kind donations of cannabis products from Aurora Cannabis Enterprises Inc. and study medication donations from Pfizer Inc. (varenicline for smoking cessation) and Bioprojet Pharma. He was also provided a coil for a Transcranial magnetic stimulation (TMS) study from Brainsway. B. L. F. has obtained industry funding from Canopy Growth Corporation (through research grants handled by the Centre for Addiction and Mental Health and the University of Toronto), Bioprojet Pharma, Alcohol Countermeasure Systems (ACS), Alkermes and Universal Ibogaine. Lastly, B. L. F. has received in kind donations of nabiximols from GW Pharmaceuticals for past studies funded by CIHR and NIH. He has participated in a session of a National Advisory Board Meeting (Emerging Trends BUP‐XR) for Indivior Canada and has been consultant for Shinogi. D. J. A. receives study materials from Cardiol Therapeutics and Tetra Bio Pharma for clinical trials funded by the Quebec Ministry of Health and Social Services and that are not related to the topic of the present manuscript. The remaining authors declare no conflict of interest.
ACKNOWLEDGMENTS
The study was supported by the Canadian Institutes of Health Research (CIHR) through the Canadian Research Initiative in Substance Misuse (CRISM; grant numbers CIS‐144301, CIS‐144302, CIS‐144303, CIS‐144304) The 4 nodes of CRISM received independent funding through a CIHR priority‐drive initiative (grant numbers SMN‐139148, SMN‐139149, SMN‐139150, SMN‐139151). M. E. S. is supported by a Michael Smith Health Research BC (MSHRBC)/St. Paul's Foundation Scholar Award. L. T. is supported by an MSHRBC Scholar Award. H. C. is supported by the Cordula and Gunter Paetzold Fellowship and the Canadian Graduate Scholarships Master's Program through the University of British Columbia. E. W. is supported by a Canada Research Chair by CIHR. B. L. F. is supported by CAMH, Waypoint Centre for Mental Health Care, a clinician‐scientist award from the department of Family and Community Medicine of the University of Toronto and a Chair in Addiction Psychiatry from the department of Psychiatry of University of Toronto. D. J. A. is supported by a research scholar award from the Fonds de Recherche du Québec en Santé. P. B. is supported by a Health Professional‐Investigator Award from Michael Smith Health Research BC, the British Columbia Centre on Substance Use, and the St. Paul's Foundation. All funders had no role in trial design, conduct, analysis or reporting. The authors thank the study participants for their contribution to the research, as well as current and past researchers and staff of the ‘OPTIMA Research Group. A special thank you to Omar Ledjiar and Michelle Cui for their support with the statistical analysis.
Crepeault H, Ti L, Bach P, et al. Opioid agonist treatment outcomes among individuals with a history of non‐fatal overdose: findings from a pragmatic, pan‐Canadian, randomized control trial. Am J Addict. 2025;34:50‐59. 10.1111/ajad.13635
REFERENCES
- 1.National Safety Council. Drug overdoses—Data details—Injury facts. Accessed December 13, 2023. https://injuryfacts.nsc.org/home-and-community/safety-topics/drugoverdoses/data-details/
- 2. Ciccarone D. The rise of illicit fentanyls, stimulants and the fourth wave of the opioid overdose crisis. Curr Opin Psychiatry. 2021;34(4):344‐350. 10.1097/YCO.0000000000000717 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Tori ME, Larochelle MR, Naimi TS. Alcohol or benzodiazepine co‐involvement with opioid overdose deaths in the United States, 1999‐2017. JAMA Netw Open. 2020;3(4):e202361. 10.1001/JAMANETWORKOPEN.2020.2361 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Milaney K, Passi J, Zaretsky L, et al. Drug use, homelessness and health: responding to the opioid overdose crisis with housing and harm reduction services. Harm Reduct J. 2021;18(1):92. 10.1186/S12954-021-00539-8/TABLES/3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Fischer B. The continuous opioid death crisis in Canada: changing characteristics and implications for path options forward. Lancet Reg Health Am. 2023;19:100437. 10.1016/j.lana.2023.100437 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Suffoletto B, Zeigler A. Risk and protective factors for repeated overdose after opioid overdose survival. Drug Alcohol Depend. 2020;209:107890. 10.1016/J.DRUGALCDEP.2020.107890 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. U.S. Department of Health and Human Services . Non‐fatal opioid overdose and associated health outcomes. Published online 2019. Accessed September 16, 2022. https://aspe.hhs.gov/reports/non-fatal-opioid-overdose-associated-health-outcomes-final-summary-report
- 8. Caudarella A, Dong H, Milloy MJ, Kerr T, Wood E, Hayashi K. Non‐fatal overdose as a risk factor for subsequent fatal overdose among people who inject drugs. Drug Alcohol Depend. 2016;162:51‐55. 10.1016/J.DRUGALCDEP.2016.02.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Moe J, Chong M, Zhao B, Scheuermeyer FX, Purssell R, Slaunwhite A. Death after emergency department visits for opioid overdose in British Columbia: a retrospective cohort analysis. CMAJ Open. 2021;9(1):242. 10.9778/CMAJO.20200169 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Larochelle MR, Bernson D, Land T, et al. Medication for opioid use disorder after nonfatal opioid overdose and association with mortality: a cohort study. Ann Intern Med. 2018;169(3):137‐145. 10.7326/M17-3107 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Keen C, Kinner SA, Young JT, et al. Periods of altered risk for non‐fatal drug overdose: a self‐controlled case series. Lancet Public Health. 2021;6(4):e249‐e259. 10.1016/S2468-2667(21)00007-4 [DOI] [PubMed] [Google Scholar]
- 12. Elliott L, Bennett AS, Wolfson‐Stofko B. Life after opioid‐involved overdose: survivor narratives and their implications for ER/ED interventions. Addiction. 2019;114(8):1379‐1386. 10.1111/ADD.14608 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Mattick RP, Breen C, Kimber J, Davoli M. Buprenorphine maintenance versus placebo or methadone maintenance for opioid dependence. Cochrane Database Syst Rev. 2014;2014(2):002207. 10.1002/14651858.CD002207.PUB4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Klimas J, Hamilton MA, Gorfinkel L, Adam A, Cullen W, Wood E. Retention in opioid agonist treatment: a rapid review and meta‐analysis comparing observational studies and randomized controlled trials. Syst Rev. 2021;10(1):216. 10.1186/S13643-021-01764-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Hser YI, Saxon AJ, Huang D, et al. Treatment retention among patients randomized to buprenorphine/naloxone compared to methadone in a multi‐site trial. Addiction. 2014;109(1):79‐87. 10.1111/ADD.12333 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Socias ME, Ahamad K, Le Foll B, et al. The OPTIMA study, buprenorphine/naloxone and methadone models of care for the treatment of prescription opioid use disorder: study design and rationale. Contemp Clin Trials. 2018;69:21‐27. 10.1016/J.CCT.2018.04.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. British Columbia Centre on Substance Use, BC Ministry of Health, and BC Ministry of Mental Health and Addictions . A guideline for the clinical management of opioid use disorder. Published online November 2023. Accessed November 29, 2023. https://www.bccsu.ca/opioid-use-disorder
- 18. O'Connor AM, Cousins G, Durand L, Barry J, Boland F. Retention of patients in opioid substitution treatment: a systematic review. PLoS One. 2020;15(5):e0232086. 10.1371/JOURNAL.PONE.0232086 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Tahsin F, Morin KA, Vojtesek F, Marsh DC. Measuring treatment attrition at various stages of engagement in opioid agonist treatment in Ontario Canada using a cascade of care framework. BMC Health Serv Res. 2022;22(1):490. 10.1186/S12913-022-07877-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Jutras‐Aswad D, Le Foll B, Ahamad K, et al. Flexible buprenorphine/naloxone model of care for reducing opioid use in individuals with prescription‐type opioid use disorder: an open‐label, pragmatic, noninferiority randomized controlled trial. Am J Psychiatry. 2022;179(10):726‐739. 10.1176/APPI.AJP.21090964 [DOI] [PubMed] [Google Scholar]
- 21. Au VYO, Rosic T, Sanger N, et al. Factors associated with opioid overdose during medication‐assisted treatment: how can we identify individuals at risk? Harm Reduct J. 2021;18(1):71. 10.1186/S12954-021-00521-4/TABLES/3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Gottlieb A, Yatsco A, Bakos‐Block C, Langabeer JR, Champagne‐Langabeer T. Machine learning for predicting risk of early dropout in a recovery program for opioid use disorder. Healthcare. 2022;10(2):223. 10.3390/HEALTHCARE10020223 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Troberg K, Bråbäck M, Isendahl P, Nilsson S, Dahlman D, Håkansson A. Malmö treatment referral and intervention study (MATRIS)—36‐month follow‐up on retention and substance use among patients referred from needle exchange to opioid agonist treatment—the role of stimulant use at baseline. J Eng Therm Energy Power. 2023;151:209036. 10.1016/J.JOSAT.2023.209036 [DOI] [PubMed] [Google Scholar]
- 24. Frost MC, Lampert H, Tsui JI, Iles‐Shih MD, Williams EC. The impact of methamphetamine/amphetamine use on receipt and outcomes of medications for opioid use disorder: a systematic review. Addict Sci Clin Pract. 2021;16(1):62. 10.1186/S13722-021-00266-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Kurz M, Min JE, Dale LM, Nosyk B. Assessing the determinants of completing OAT induction and long‐term retention: a population‐based study in British Columbia, Canada. J Subst Abuse Treat. 2022;133:108647. 10.1016/J.JSAT.2021.108647 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Kakko J, Grönbladh L, Svanborg KD, et al. A stepped care strategy using buprenorphine and methadone versus conventional methadone maintenance in heroin dependence: a randomized controlled trial. Am J Psychiatry. 2007;164(5):797‐803. 10.1176/AJP.2007.164.5.797/ASSET/IMAGES/LARGE/R720F4.JPEG [DOI] [PubMed] [Google Scholar]
- 27. Bromley L, Kahan M, Regenstreif L, Srivastava A, Wyman J. Methadone Treatment for People Who Use Fentanyl: Recommendations . META:PHI; Published online 2021. www.metaphi.ca
- 28. Morris NP. Opioid use disorder treatment in the age of fentanyl. JAMA Intern Med. 2022;182(3):249‐250. 10.1001/JAMAINTERNMED.2021.8114 [DOI] [PubMed] [Google Scholar]
- 29. Varshneya NB, Thakrar AP, Hobelmann JG, Dunn KE, Huhn AS. Evidence of buprenorphine‐precipitated withdrawal in persons who use fentanyl. J Addict Med. 2022;16(4):e265‐e268. 10.1097/ADM.0000000000000922 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Shearer D, Young S, Fairbairn N, Brar R. Challenges with buprenorphine inductions in the context of the fentanyl overdose crisis: a case series. Drug Alcohol Rev. 2022;41(2):444‐448. 10.1111/DAR.13394 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Fischer G, Gombas W, Eder H, et al. Buprenorphine versus methadone maintenance for the treatment of opioid dependence. Addiction. 1999;94(9):1337‐1347. 10.1046/J.1360-0443.1999.94913376.X [DOI] [PubMed] [Google Scholar]
- 32. Hser YI, Saxon AJ, Huang D, et al. Treatment retention among patients randomized to buprenorphine/naloxone compared to methadone in a multi‐site trial. Addiction. 2014;109(1):79‐87. 10.1111/ADD.12333 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Fairbairn N, Ross J, Trew M, et al. Injectable opioid agonist treatment for opioid use disorder: a national clinical guideline. Can Med Assoc J. 2019;191(38):E1049‐E1056. 10.1503/CMAJ.190344/-/DC1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. BC Centre on Substance Use . Monthly drug checking reports. 2022. Accessed June 21, 2023. https://drugcheckingbc.ca/monthly-reports/