Buprenorphine unobserved “home” induction: a survey of Ontario’s addiction physicians

Buprenorphine/naloxone is an effective treatment for opioid use disorder (OUD) [1] and is now considered a first-line treatment for OUD [2, 3] over methadone treatment because of its favourable safety profile and its feasibility in primary care settings. However, the majority of patients currently on opioid agonist treatment (OAT) in Ontario are still prescribed methadone. In 2017, there were 60,758 patients (4.3 per 1000 population) on OAT in Ontario and, of these, 44,375 were on methadone [4]. This is relevant considering that in 2017, methadone was involved in 211/1265 (16.7%) of opioid-related deaths and cited as one of the three opioids most commonly responsible in single-opioid deaths [5, 6].

Various facilitators and barriers have likely contributed to the current prevalence of methadone prescribing compared with buprenorphine in Ontario. Methadone has been available for the treatment of OUD in Canada for decades, whereas buprenorphine was only approved for use in Canada in 2007, and full public drug coverage¹ for buprenorphine has been in place in Ontario only since 2016. While the rate of buprenorphine prescribing increased from 0.23 to 0.85 per 1000 Ontarians between 2012 and 2016, the rate of methadone prescribing remained robust and increased from 2.0 to 2.6 per 1000 Ontarians in the same period [7]. This finding suggests that there are other potential barriers to buprenorphine being prescribed.

We hypothesized that one of the main barriers for both addiction physicians and primary care physicians to prescribing buprenorphine is the recommendation that the initial doses of buprenorphine be taken in the office under the observation of a clinician. This could play a role in some addiction medicine prescribers’ preference for methadone overall.

Most buprenorphine guidelines have recommended that the initial dose be taken in an observed clinical setting [8, 9] to prevent diversion, buprenorphine toxicity, and precipitated withdrawal. Yet, the logistics of observed office induction are onerous, as physicians or nurses are required to assess the patient several times over a period of a few hours and the initiation process has been described as a barrier to buprenorphine prescribing [10‐13]. Recent studies have demonstrated the safety of unobserved induction [14, 15] and that this is becoming a more widely accepted practice in some jurisdictions [16, 17].

We designed a survey of Ontario OAT providers, who are largely primary care physicians with focused addiction practices, to better understand their attitudes towards buprenorphine. In particular, we explored potential barriers to buprenorphine’s use, including attitudes and practices regarding home induction.

We received 88 responses to our survey. Three incomplete surveys were excluded from the analysis as the participants did not answer the likert scale questions. The remaining 85 completed surveys represent 39.5% of the total number of 215 attendees at the 2016 conference. It likely represents a greater percentage of the physician attendees as the 215 attendees include allied health professionals and policy makers.

The group of respondents was relatively new to OAT practice: 35% had only been practicing for 1–5 years and 37% had been practicing for 6–15 years (Table 1). The majority of the physicians identified their practices as being urban (58%) or suburban (20%). Almost half of the physicians (48%) prescribed OAT for more than 100 patients and 13% had more than 300 opioid agonist patients in their practice. A significant percentage of physicians (41%) indicated that they had initiated buprenorphine rather than methadone in more than 40% of their new opioid agonist starts in the past year. The survey explored three main approaches to initiating buprenorphine (Table 2). A large majority (74%) of respondents indicated that they usually or always have patients take their initial doses at the clinic; 31% indicated that they usually or always have patients take the first 1–2 doses observed at the clinic and allow the rest of the day’s dose to be taken unobserved. Only 4% indicated that they usually used unobserved induction for the first few doses of buprenorphine. Yet, when asked about the convenience of unobserved induction, 64% respondents agreed or strongly agreed that it was more convenient for patients and 35% agreed or strongly agreed that it was more convenient for physicians (Table 3).

When asked about barriers to buprenorphine induction, 51% of respondents indicated that they usually or sometimes had to prescribe methadone rather than buprenorphine for patients who were unable to come to the clinic in withdrawal. (Table 2).

When asked about risks of unobserved induction, 59% of physicians agreed or strongly agreed that unobserved induction was risky because it was against “the guidelines,” 66% agreed or strongly agreed that it increased the risk of diversion, and 61% agreed or strongly agreed that it increased the risk of precipitated withdrawal.

We explored other areas of interest related to the uptake of buprenorphine prescribing in primary care and the perceptions of addiction physicians related to the possible advantages and disadvantages of buprenorphine in a primary care setting. Thirty-six percent of respondents agreed or strongly agreed that primary care physicians without focused addiction medicine practices should not prescribe buprenorphine. However, only 19% of physicians disagreed or strongly disagreed with the statement that non-addiction physicians should prescribe buprenorphine when travel to a specialized clinic is very difficult for patients (Table 3).

The majority (82%) of respondents agreed or strongly agreed with the statement that buprenorphine was preferred for patients with benzodiazepine or heavy alcohol use, and most (84%) agreed or strongly agreed that buprenorphine was preferred for patients who were working or were not close to a pharmacy. The vast majority of respondents (83%) agreed or strongly agreed with the statement that buprenorphine is  the preferred treatment choice for youth.

Our finding that only 4% of surveyed Ontario addiction physicians involved in OAT routinely used unobserved “home” induction differs from similar studies in other jurisdictions which have found that unobserved induction is widely used: a survey of physicians in Massachusetts [16] found that 43% of patients were given take-home doses on initiation of treatment. A more recent survey [17] of New York providers found that unobserved buprenorphine induction was standard practice amongst 65% of respondents even though a significant number (38%) felt that concern about diversion was a barrier to buprenorphine prescribing. One possible explanation for the differences between Ontario physicians and those in other jurisdictions is that the former can prescribe both methadone and buprenorphine (methadone is only available in specialized clinics in the United States). Thus, the Ontario physician can prescribe methadone if the patient is unwilling or unable to present to the office in withdrawal for buprenorphine induction.

Our survey results support the hypothesis that many patients who were either eligible or intended for buprenorphine treatment ended up on methadone: 25% of respondents said that they usually ended up prescribing methadone because the patient did not want to be in withdrawal, and almost half said this often occurred. More than 50% indicated that they usually or sometimes had to prescribe methadone because patients would miss their induction appointment or had trouble arriving in withdrawal on the scheduled induction day. This may help explain, in part, the high use of methadone prescribing in Ontario despite the greater educational requirements for physicians prescribing methadone.

It is interesting that the Ontario addiction physicians we surveyed largely did not use unobserved home induction even though cohort studies have established the safety of unobserved induction, with a low incidence of adverse events and good treatment retention rates [14, 15]. Concerns were expressed around “going against the guidelines”: in Ontario, physicians who prescribe methadone have clinical guidelines that are often used as the standard of care to which they are held accountable by the province’s medical licensing authority. This may have created a culture in which addiction physicians are reluctant to go outside the parameters of clinical guidelines in prescribing OAT. The first set of clinical guidelines for buprenorphine in Ontario were published in 2011 [8] and were, until recently, the ones most clinicians likely used in their clinical practices. Those guidelines did not endorse unobserved induction. Newer guidelines in Canada and United States [18‐20] recommend caution with unobserved induction and recommend that only experienced providers attempt it.

The acceptance of home induction as a mainstream practice could decrease reliance on methadone and expand access to buprenorphine treatment. Physicians could consider home induction for patients who cannot attend the office in withdrawal for planned buprenorphine inductions rather than feeling that their only option was to initiate methadone. Emergency Department (ED) physicians would be able to prescribe buprenorphine to patients not yet in withdrawal. Primary care physicians similarly may be more amenable to prescribe buprenorphine if they do not have to arrange office induction. Aside from being hard to schedule and time consuming, office induction also creates the impression that buprenorphine induction involves considerable risk and may discourage primary care physicians from incorporating buprenorphine treatment into their practices. Our survey results suggest that addiction physicians are open to working with primary care physicians to expand access to OAT: respondents believed that prescribing physicians should be experienced but that if distance was a barrier for patients then their family doctor should provide OAT.

However, family doctors and addiction physicians may be hesitant to routinely use unobserved “home” induction until there are clinical guidelines clearly stating that it is an acceptable alternative to office induction. Clinical guidelines and practices in Ontario continue to represent a barrier to both its adoption into primary care as well as a barrier to treatment entry for many patients who are unable to present in withdrawal to an ambulatory appointment. The existing recommendations could be revised to more clearly state the acceptability of home induction, as the risks of a small limited supply of buprenorphine tablets are minimal.

There were limitations to our survey. The surveyed group was a convenience sample of physicians at the annual opioid primary care provincial conference. While we are able to capture a sizeable portion of active of Ontario methadone providers (almost half of the registered methadone prescribers in Ontario attended the conference and almost 40% of attendees responded, which is likely more than half the attending physicians), we do not have the demographics of Ontario methadone prescribers as a whole to which we can compare our sample. The demographics of our survey respondents indicated that many were relatively new to practice, were starting new patients on buprenorphine, and the majority had practice sizes of fewer than 200 patients, suggesting that we were likely surveying addiction physicians who were already involved in buprenorphine treatment and who were not very high-volume prescribers.

In our survey of Ontario addiction physicians who prescribe OAT, the majority use observed office induction. Only 4% were routinely performing unobserved “home” buprenorphine inductions. Many participants stated that office induction was a barrier to treatment for patients, but were apprehensive about possible precipitated withdrawal and about not following clinical guidelines that recommend observed induction. Initiatives to promote provider comfort with buprenorphine home induction may help improve access to opioid use disorder treatment.