Abstract
Background:
In 2016, the College of Physicians and Surgeons of British Columbia released a legally enforceable opioid prescribing practice standard for the treatment of chronic noncancer pain (CNCP); it was revised in 2018 in response to concerns that it was misinterpreted. We aimed to test the effects of the practice standard on access to opioids for people treated for CNCP, living with cancer, or receiving palliative care.
Methods:
We used comprehensive administrative health data from Oct. 1, 2012, to Mar. 31, 2020, and multiple baseline interrupted time-series analysis to evaluate the effects of the 2016 practice standard and 2018 revision in cohorts of people treated for CNCP, living with cancer, or receiving palliative care.
Results:
The 2016 practice standard accelerated pre-existing monthly trends in morphine milligram equivalents (MME) dispensed per person treated for CNCP (−0.1%, 95% confidence interval [CI] −0.2% to 0.0%), but also for people living with cancer (−0.7%, 95% CI −1.0% to −0.5%) and those receiving palliative care (−0.3%, 95% CI −0.5% to 0.0%). The proportion of people with CNCP prescribed a daily dose greater than 90 MME (−0.3%, 95% CI −0.4% to −0.2%), coprescribed a benzodiazepine or other hypnotic (−0.6%, 95% CI −0.7% to −0.5%), and aggressively tapered (−0.1%, 95% CI −0.2% to 0.0%) also decreased more quickly after the practice standard. Although we observed null or decreases in level effects overall, the proportion of people aggressively tapered increased 2.0% (95% CI 0.4% to 3.3%) immediately after implementation of the practice standard. Trends slowed or reversed after the 2018 revision.
Interpretation:
The 2016 practice standard was associated with an immediate and long-lasting effect on physicians’ opioid prescribing behaviours, including inadvertently increasing aggressive tapering (observed level effect) and reducing access to opioids for people living with cancer or receiving palliative care.
Between January 2014 and September 2024, around 16 721 drug-related deaths occurred in British Columbia, with opioids present in more than 80% of events.1 As part of the effort to lower the risk of opioid-related deaths, the College of Physicians and Surgeons of British Columbia (henceforth “College”) released the Safe Prescribing of Drugs with Potential for Misuse/Diversion practice standard in June 2016 (Box 1 and Appendix 1, available at www.cmaj.ca/lookup/doi/10.1503/cmaj.250167/tab-related-content).2 This legally enforceable standard was developed to “prevent an increasing toll of prescription drug misuse and overdose deaths” by limiting access to opioid prescriptions for chronic noncancer pain (CNCP) management.2 Previous treatment guidelines in Canada and abroad (e.g., the CDC Guideline for Prescribing Opioids for Chronic Pain – United States, 2016,3 the College’s 2012 Prescribing Principles,4 and the National Opioid Use Guideline Group’s 2010 Safe and Effective Use of Opioids for Chronic Non-Cancer Pain)5 recommended courses of action and allowed physicians to “exercise reasonable discretion in their decision to act on guidance provided.”2,6 The 2016 practice standard required specific prescribing practices. Physicians found noncompliant could be disciplined or fined under the Health Professions Act, RSBC 1996, c.183 and College bylaws.2,8
Box 1: Safe Prescribing of Drugs with Potential for Misuse/Diversion Practice Standards, June 2016 * 3 .
Physicians must:
Review patients’ current medications before prescribing opioids, sedatives, or stimulants.
Base long-term treatment with medications with known risks, including opioids, sedatives, and stimulants, upon clinical diagnosis and objective evidence. Continuing to prescribe medication solely on the basis that they have been previously prescribed is not acceptable.
Document discussion with patients that nonpharmacologic therapy and nonopioid analgesics are preferred for chronic noncancer pain, and that the potential benefit of long-term opioid treatment (LTOT) is modest and risk significant.
Advise patients that LTOT is not indicated for certain medical conditions including headache disorders, fibromyalgia, and axial low back pain.
Always prescribe the lowest effective dosage of opioid medication. Doses > 50 morphine milligram equivalents (MME) per day warrant careful reassessment and documentation. Doses > 90 MME per day warrant substantive evidence of exceptional need and benefit. (This advice excludes treatment with methadone.)
When treating patients with acute pain conditions, prescribe only immediate-release opioids in quantities that the patient will need before community follow-up will be resumed (3 to 7 days is often adequate).
When discharging patients from acute-care settings, or postoperatively, prescribe only the quantities of opioids, sedatives, or stimulants that the patient will need before community follow-up will be resumed.
Base decisions to prescribe long-term psychoactive medications, including LTOT, on well-documented, comprehensive initial assessments and frequent (at least every 3 months) reassessments. These assessments and reassessments must include documented history and physical examination of the patient. There must also be documentation that the patient has been screened regularly for the presence or emergence of mental health and substance use disorders and risk factors and advised about safety-sensitive occupational risks, child care responsibilities and driving.
Document the offer of a take-home naloxone prescription to all patients who are at risk of respiratory depression as a consequence of receiving opioid medications.
Document having directed and regularly reminded patients for whom they are prescribing LTOT to abstain from alcohol and nonprescription sedatives.
Order at least annual random urine drug testing and/or random pill counts for all adult patients on long-term opioids, benzodiazepines, sedative hypnotics, or stimulants.
Further, physicians must not:
Prescribe benzodiazepines or sedative hypnotics to patients on LTOT, other than as a documented taper.
Prescribe combinations of opioids with benzodiazepines and/or sedative hypnotics.
Provide prescriptions allowing dispenses of opioids, sedatives, and stimulants, which exceed a 3-month supply or 250 pills, whichever is less.
Initiate treatment with drugs with a high risk-profile such as methadone and fentanyl without relevant training and experience.
Reproduced with permission from the College of Physicians and Surgeons of British Columbia (CPSBC). These practice standards have been revised since 2016 as part of the CPSBC standard cyclical review process and to reflect new information as it has become available.
Across Canada and the United States, the effect of prescribing guidelines on access to opioids appears mixed. A study on the College’s 2016 practice standard found no change in total opioid consumption but observed a decline in prescription renewals.9 However, the study was not specific to CNCP because it included opioids dispensed to patients exempt to the standard (i.e., those with cancer or palliative care). Three US studies on opioid prescribing after guideline implementation reported an increase in prescribing prevalence,10–12 while 2 studies found a decrease in prescribing incidence,13,14 and 2 others reported lower dosages.15,16 Notably, these studies also did not focus on CNCP. Additionally, evidence suggests that misinterpretation of guidelines led to nonconsensual aggressive tapering (i.e., the rate of change in dose exceeded the recommended threshold), debilitating pain, and increased overdose risk.17–23 In June 2018, the College released the Safe Prescribing of Opioids and Sedatives revision to address concerns from physicians, patient groups, and key partners about the misinterpretation of the 2016 standard leading to more conservative prescribing to all patients, and not just those living with CNCP (Appendix 1).24 Changes included careful consideration of coprescribing opioids with sedatives (e.g., registrants “must not prescribe benzodiazepines or sedative hypnotics to patients on long-term opioid treatment, other than as a documented taper” in 2016 version2 v. “If long-term treatment is considered for [coprescribed benzodiazepines, sedatives, and opioids], registrants must taper and discontinue one of them after making all efforts to involve the patient in this decision and providing a thorough explanation” in 2018 version24), removal of stimulants from the standard, and elimination of the limit on supply.
We tested the effects of the 2016 practice standard and its 2018 revision on the morphine milligram equivalents (MME) dispensed to patients with CNCP managed with long-term opioid treatment (LTOT), and the proportion of these patients with a reassessment by their prescribing physician every 3 months, a co-occurring benzodiazepine or other sedative hypnotic prescription, prescriptions exceeding a 3-month supply or 250 pills, a daily dose exceeding 90 MME, or an opioid taper exceeding a 10% decrease.2 We also tested the effects of the practice standard and its revision on MME dispensed to patients living with cancer or receiving palliative care.
Methods
Study design
We used a multiple intervention time-series study design to test the effects of the practice standard and its revision on opioid prescribing to adults living in BC between Oct. 1, 2012, and Mar. 31, 2020. Interrupted time-series analysis is a quasi-experimental study design that estimates population-level effects of policy interventions before and after implementation in contexts where randomized controlled trials are not feasible or ethical.25 This before–after comparison assumes selection bias and within-group characteristics that change slowly over time, secular changes, random fluctuations from 1 time point to the next, and regression to the mean are sufficiently controlled.
Data
We linked patient-level data from the BC Ministry of Health’s PharmaNet,26 the Discharge Abstract Database (DAD),27 Medical Services Plan (MSP) billings and consolidation files,28 and the National Ambulatory Care Reporting System (NACRS),29 with the BC Cancer Agency’s cancer registry30 and Vital Statistics’ Mortality31 data to determine the primary purpose of each opioid prescription dispensed and when to censor patients. PharmaNet included all prescriptions filled by community pharmacies to BC residents, irrespective of payer. Data from DAD, MSP, and NACRS captured hospital and outpatient physician visits. The consolidation file and mortality data sets informed when patients first entered and exited the cohort. The cancer registry included all residents diagnosed with cancer.
Cohort
Our primary cohort included all adult residents (aged ≥ 18 yr) living in the community with CNCP on LTOT. To create the cohort, we identified all unique opioid prescriptions dispensed between October 2012 and March 2020. We excluded opioid formulations prescribed for the management of opioid use disorder, cough, or diarrhea; prescriptions dispensed after patients met the exclusion criteria of cancer, palliative care, long-term care, death, or the end of registration with provincial health insurance; medications dispensed within 7 days of trauma or major surgical intervention; and prescriptions dispensed by a non-BC physician or other health professional (e.g., nurse practitioners, out-of-province physicians) or missing the prescribing physician’s identifier. Using the remaining prescriptions, we defined people on LTOT as those with an opioid analgesic dispensation renewed at least once with at least 60 days’ supply dispensed during a 90-day period. We repeated the process to create our standard-exempt cohorts, retaining prescriptions dispensed to patients living with cancer and prescriptions dispensed to patients receiving palliative care to create mutually exclusive cohorts.
Outcome
Our unit of analysis was person-month. We converted the total dose of opioids dispensed across prescriptions to daily MME using each prescription’s start date, days’ supply, and dispensed quantity (allowing for multiple co-occurring prescriptions). We separately tested the effect of the practice standard on MME dispensed per patient treated with LTOT for CNCP, living with cancer, or receiving palliative care. We also examined the proportion of patients on LTOT for CNCP who did not visit the opioid-prescribing physician within 90 days after the prescription was dispensed; those who had a co-occurring benzodiazepine or other hypnotic prescription at time of fill; and those who had a prescription that exceeded 3-month supply or 250 pills (whichever was less), a daily dose of 90 MME, or a dose reduction of 10% or more (i.e., patients who were aggressively tapered) from the previous 2-week period (including abrupt cessation not due to moving out of province, death, receiving a diagnosis of cancer, or starting palliative care).
Statistical analysis
We divided the study into 3 periods: preintervention (January 2013 to June 2016), post-2016 practice standard (July 2016 to June 2018), and post-2018 practice standard revision (July 2018 to December 2019). Prescription data from October 2012 to December 2012 and from January 2020 to March 2020 served as wash-in and wash-out periods for our measures, respectively.
We tested the effects of the practice standard using generalized linear regression models (Poisson) for interventions j and k (Outcomejkt = β0 + β1 timet + β2 levelj + β3 leveljtimet + β4 levelk + β5 levelktimet + ɛjkt) where β0 was the expected outcome at the first time point (i.e., intercept); β1 was the pre-existing trend in the outcome; β2 was the change in the outcome level between the preintervention period and immediately following the practice standard’s implementation (June 2016); β3 was the change in outcome trend from the preintervention period to after the implementation of the 2016 practice standard; β4 was the change in the outcome level between the post-2016 practice standard period and its revision in June 2018; and β5 was the change in the outcome trend between the post-2016 practice standard period and its 2018 revision. We used the 2-sided Durbin–Watson test and residual plots to test for autocorrelation (e.g., first order) and moving averages to inform the models’ correlation structures. We used Newey–West standard errors with a lag of 3 to correct for the autocorrelation detected. The data were prepared using SAS 9.4, and analysis was executed using R version 4.0.
Ethics approval
This study was approved by McGill University’s Institutional Review Board (no. A11-M55–19A).
Results
We identified 46 629 134 unique opioid prescriptions dispensed in the province between October 2012 and March 2020 (Figure 1). To identify patients receiving LTOT for CNCP, we excluded opioid formulations exclusively prescribed for the management of opioid use disorder, cough, or diarrhea (n = 26 541 369). Further, we removed all prescriptions dispensed after the date patients met cancer, palliative care, or long-term residence exclusion criteria (n = 3 711 326), and after the date of patients’ death or end of insurance registration (n = 9006). We considered and excluded opioids first dispensed within 7 days of trauma or major surgical intervention, and with less than 14 days’ supply as prescriptions for acute pain management (n = 437 007). We identified 921 033 prescriptions dispensed without a physician identifier, most of which (n = 862 738) were prescribed by a non-BC physician. The remaining 58 295 were prescribed by BC physicians but were missing the prescribing practitioner’s identifier. We assigned a prescriber to 4685 prescriptions by assuming it was the same physician as that prescribing the opioid analgesic immediately before and after. After removing prescriptions dispensed by dental surgeons (n = 154 334), we defined people on LTOT for CNCP as those with an opioid analgesic dispensation renewed at least once and with at least 60 days’ supply dispensed within a 3-month period (n = 9 611 066 dispensations for primary analysis).
Figure 1:
Flowchart of opioid prescriptions captured in PharmaNet data set and retained to identify cohort of patients treated for chronic noncancer pain who were dispensed long-term opioid treatment in British Columbia.
During the observation period, we included between 38 796 (February 2019) and 57 276 (October 2014) BC residents with CNCP in our study. We also included between 2652 (February 2013) and 7945 (October 2018) patients living with cancer, and between 1840 (November 2014) and 2231 (October 2019) patients receiving palliative care, after applying our exclusions. At the start of the observation period (January 2013), among patients receiving LTOT for CNCP, 25.4% did not have a visit with the prescribing physician within 90 days, 29.3% had a co-occurring benzodiazepine or other hypnotic prescription dispensed, 5.7% had a supply that exceeded 90 days or 250 pills, 32.7% had a daily dose that exceeded 90 MME, and 39.2% were aggressively tapered.
Our analysis revealed that, among patients with CNCP, the 2016 practice standard was associated with acceleration of pre-existing declining trends in the MME dispensed per person by 0.13% (95% confidence interval [CI] −0.21% to −0.04%) (Table 1 and Figure 2) and in the proportion of patients (Table 2 and Figure 3) with co-occurring benzodiazepine or other hypnotic dispensation (−0.62%, 95% CI −0.74% to −0.51%), those with daily doses exceeding 90 MME (−0.29%, 95% CI −0.36% to −0.22%), and those who were aggressively tapered (−0.13%, 95% CI −0.21% to −0.04%).
Table 1:
Population-level changes in morphine milligram equivalents (MME) dispensed per person following the implementation of the 2016 practice standard on opioid prescribing and the 2018 revision, by pain type
| Variable | % (95% CI) | ||
|---|---|---|---|
| Patients with CNCP on LTOT | Patients living with cancer | Patients receiving palliative care | |
| Preintervention trend, β1 | −0.64 (−0.67 to −0.61) | −0.09 (−0.16 to −0.02) | −0.60 (−0.69 to 0.51) |
| 2016 practice standard level change, β2 | −5.10 (−6.48 to −3.73) | 1.34 (−3.61 to 6.28) | −0.07 (−3.31 to 3.16) |
| 2016 practice standard trend change, β3 | −0.13 (−0.21 to −0.04) | −0.74 (−0.98 to −0.50) | −0.26 (−0.49 to −0.03) |
| 2018 revision level change, β4 | 1.26 (0.18 to 2.36) | 2.58 (0.14 to 5.02) | 2.67 (−2.54 to 7.88) |
| 2018 revision trend change, β5 | 0.57 (0.44 to 0.70) | 0.21 (−0.11 to 0.53) | 0.18 (−0.21 to 0.57) |
| Absolute change post-2016 practice standard | −8.2 (−11.4 to −4.7) | −15.0 (−24.7 to −5.1) | −6.1 (−14.4 to 2.4) |
Note: CI = confidence interval, CNCP = chronic noncancer pain, LTOT = long-term opioid treatment.
Figure 2:
Changes in morphine milligram equivalents (MME) dispensed per person in British Columbia, before and after the implementation of the 2016 practice standard and the 2018 revision, by pain type among (A) people treated for chronic noncancer pain, (B) people living with cancer, and (C) people receiving palliative care. The vertical dashed lines represent the implementation of the 2016 practice standard and the 2018 revision. Grey dots represent the total MME dispensed per person each month. The black solid line represents the trend in prescribing per period. The green ribbon represents the 95% confidence interval (CI) around the observed trend lines. The solid red line represents the expected trend had the 2016 practice standard not been implemented, with the pink ribbon capturing the 95% CI. The solid blue line represents the expected trend had the 2018 revision not been implemented, with the light blue ribbon capturing the 95% CI. Supporting data are presented in Table 1.
Table 2:
Effects of 2016 practice standard and the 2018 revision on measures of care for patients on long-term opioid treatment for chronic noncancer pain
| Variable | % (95% CI) | ||||
|---|---|---|---|---|---|
| No visit with prescribing physician within 90 d | Co-occurring benzodiazepine or other hypnotic | Supply exceeding 90 d or 250 pills | Daily dose exceeding 90 MME | Aggressive taper | |
| Preintervention trend, β1 | −0.07 (−0.17 to 0.04) | −0.67 (−0.71 to −0.63) | −0.65 (−0.83 to −0.47) | −0.53 (−0.57 to −0.50) | −0.31 (−0.34 to −0.28) |
| 2016 practice standard level change, β2 | −1.38 (−6.55 to 3.79) | −5.24 (−7.67 to −2.82) | −11.32 (−16.93 to −5.72) | −3.56 (−4.66 to −2.46) | 1.85 (0.41 to 3.29) |
| 2016 practice standard trend change, β3 | −0.20 (−0.49 to 0.09) | −0.62 (−0.74 to −0.51) | −0.24 (−0.60 to 0.12) | −0.29 (−0.36 to −0.22) | −0.13 (−0.21 to −0.04) |
| 2018 revision level change, β4 | 1.76 (−3.82 to 7.33) | 2.59 (0.25 to 4.94) | 2.71 (−4.36 to 9.77) | −0.05 (−1.25 to 1.35) | 0.50 (−2.77 to 3.78) |
| 2018 revision trend change, β5 | 0.22 (−0.25 to 0.69) | 0.74 (0.62 to 0.86) | 1.22 (0.77 to 1.67) | 0.48 (0.37 to 0.59) | −0.09 (−0.45 to 0.27) |
| Absolute change post-2016 practice standard | −6.1 (−17.7 to 6.0) | −19.1 (−24.0 to −14.4) | −16.9 (−30.4 to −2.8) | −10.3 (−13.0 to −7.6) | −1.2 (−4.5 to 2.3) |
Note: CI = confidence interval, MME = morphine miligram equivalent.
Figure 3:
Effects of 2016 practice standard and the 2018 revision on proportion of adult patients on long-term opioid treatment for chronic noncancer pain with (A) no visit with prescribing physician within 90 days, (B) co-occurring benzodiazepine or other hypnotic prescription, (C) supply exceeding 90 days or 250 pills, (D) daily dose exceeding 90 morphine miligram equivalents, and (E) rapid taper. The vertical dashed lines represent the implementation of the 2016 practice standard and the 2018 revision. Grey dots represent the monthly proportion as measured. Black solid line represents the observed trend in prescribing per period. The green ribbon represents the 95% confidence interval (CI) around the trend lines. The solid red line represents the expected trend had the 2016 practice standard not been implemented, with the pink ribbon capturing the 95% CI. The solid blue line represents the expected trend had the 2018 revision not been implemented, with the light blue ribbon capturing the 95% CI. Supporting data are presented in Table 2.
The analysis also revealed that, compared with the preintervention period, the 2016 standard was associated with an immediate decline (level effect) in the MME dispensed per person (−5.10%, 95% CI −6.48% to −3.73%; Table 1) and the proportion of patients with a co-occurring benzodiazepine or other hypnotic prescription (−5.24%, 95% CI −7.67% to −2.82%), whose supply exceeded 90 days or 250 pills (−11.32%, 95% CI −16.93% to −5.7%), and whose daily dose exceeded 90 MME (−3.56%, 95% CI −4.66% to −2.46%; Table 2). However, we also observed a 1.85% (95% CI 0.41% to 3.29%) increase in the proportion of patients who were aggressively tapered. We observed no level or trend effects on the proportion of patients who did not have a visit to the prescribing physicians within 90 days.
When the College released the June 2018 revision, we observed an increase (level effect) in the MME dispensed per person with CNCP (1.26%, 95% CI 0.18% to 2.36%) and the proportion of patients with concurrent prescriptions for benzodiazepine or other hypnotics (2.59%, 95% CI 0.25% to 4.94%). We also observed month-to-month increases (trend effect) in the MME dispensed per person (0.57%, 95% CI 0.44% to 0.70%) and the proportion of patients with co-occurring benzodiazepine or other hypnotic prescription (0.74%, 95% CI 0.62% to 0.86%), a supply exceeding 90 days or 250 pills (1.22%, 95% CI 0.77% to 1.67%), and a daily dose greater than 90 MME (0.48%, 95% CI 0.37% to 0.59%).
For our other cohorts, the 2016 practice standard was also associated with decreased prescribing trends for the MME dispensed per person living with cancer (−0.74%, 95% CI −0.98% to −0.50%) and receiving palliative care (−0.26%, 95% CI −0.49% to 0.03%; Table 1 and Figure 2). For patients living with cancer, the MME dispensed increased (2.58%, 95% CI 0.14% to 5.02%), while we observed no effect on opioids dispensed to patients receiving palliative care.
In absolute terms, 24 months after the 2016 practice standard’s implementation, MME dispensed per person were 8.2%, 15.0%, and 6.1% lower than expected for patients treated for CNCP, living with cancer, and receiving palliative care, respectively. For a 30-day month, this can be interpreted as approximately 2.5, 4.5, and 1.8 days less supply than before the implementation of the practice standard.
Interpretation
Our analyses revealed that the 2016 practice standard influenced physicians’ opioid prescribing in BC. Among people with CNCP receiving LTOT, pre-existing declines in total opioid volume dispensed, high-dose prescribing, coprescribing with benzodiazepines or other hypnotics, supplies of more than 90 days or 250 pills, or daily doses above 90 MME accelerated. We also observed an immediate drop in MME dispensed per person and in the proportion of patients coprescribed benzodiazepines or other hypnotics, patients receiving supplies of more than 90 days or 250 pills, and patients whose daily dose exceeded 90 MME, along with an increase in the proportion of patients who were aggressively tapered. When the 2016 practice standard was replaced in 2018, the downward trends slowed. For MME dispensed per person and the proportion of patients with a supply of more than 90 days or 250 pills, declines reversed. We observed the same prescribing trends among patients with cancer and those receiving palliative care.
Overall, our study findings align with those of descriptive studies noting declines in defined daily doses dispensed following the release of Canadian guidelines in 2017.32,33 Our findings are also consistent with those of US studies that observed declines in opioid prescribing,15 increases in aggressive tapering, 34,35 and declines in the number of people treated with LTOT36 after the implementation of the US Centers for Disease Control and Prevention’s 2016 prescribing guidelines. Although our results appear to conflict with Crabtree and colleagues’9 observation that the 2016 BC practice standard had no effect on opioid prescribing trends, 2 methodological differences between our work and theirs may explain this discrepancy. First, their observation period may have been too short to detect the small but accumulating changes in opioid prescribing trends.9 Second, although we observed declines in MME dispensed per person for all 3 pain types, when aggregated across all patients, the total MME dispensed did not change dramatically. This is because, for much of the observation period, the number of people living with cancer or receiving palliative care increased.
Together, these findings demonstrate that prescribing guidelines and practice standards can have immediate and long-lasting effects on physician prescribing. Although most of the changes may be positive (e.g., fewer opioids in the community, a reduction in coprescribed benzodiazepine), incorrect interpretation can increase harms for some patients. Aggressive tapering can have downstream consequences, including people resorting to unregulated opioids for pain relief despite their risks.37–39 Further, although practice standards may be specific to a pain type, they may have a concerning effect on access to opioids for non-target patient groups as well. Our work overlaps with pre-existing changes in opioid prescribing behaviours among physicians, with work led by Narayan and colleagues revealing opioid-phobic prescribing attitudes irrespective of patients’ circumstances and need.40
Limitations
Since we used secondary administrative data and coding algorithms to define our cohort of people living with CNCP, we may have excluded people treated with opioids for intermittent or acute episodes of breakthrough pain (e.g., sickle cell crises). Moreover, some methadone formulations are used in specialized pain centres for refractory neuropathic pain. Researchers in BC distinguish prescriptions for opioid agonist treatment from analgesics by using a combination of product information numbers (PIN) and Health Canada’s drug information numbers (DIN; Appendix 1). To ensure our criteria were generalizable to other provinces, we favoured DINs, thereby potentially misclassifying some patients receiving methadone dispensations as recipients of opioid agonist treatment. When we compared analyses using DIN instead of PIN identifiers from our exclusion criteria, there was no appreciable difference in trends or level effects (difference of 148 167 [1.5%] prescriptions across 78 months). By prioritizing a conservative algorithm and using complete records of opioid dispensations in the community (e.g., irrespective of patients’ insurance coverage), we were able to examine the effects of the practice standard on the target population, namely primary care physicians responsible for most of the CNCP management in the community. Furthermore, we did not account for the plausible effects of the 2017 Canadian Guideline for Opioids Therapy and Chronic Non-Cancer Pain in our main analysis.18 Although we may have misattributed the effects of both the 2016 provincial practice standard and 2017 federal guideline to June 2016 using a single intervention break, a sensitivity analysis where we included a break for May 2017 (not shown here) did not reveal measurable effects. Since the 2017 guideline was not legally enforceable and was more open to interpretation than the 2016 practice standard, the effects of this guideline on physician prescribing in BC appear to be small, if present at all. Finally, we captured only opioids dispensed from community pharmacies (i.e., excluded patients in long-term care). Although increases in institutional dispensations may have offset observed trends, this is beyond this paper’s scope.41
Conclusion
We found that pre-existing declines in physicians’ opioid prescribing accelerated among patients with CNCP, those with cancer and those receiving palliative care after the 2016 practice standard was implemented. This demonstrates the ability of practice standards to modify physician behaviour but also highlights how misinterpretation can harm patients. Patient groups and physicians affected by standards or guidelines should be consulted before their release to reduce unintended consequences. Meanwhile, studies should measure the effects on exempt populations (e.g., changes in opioid access for patients with cancer or those receiving palliative care), and those in long-term care facilities. Colleges and government agencies should carefully assess the potential impacts of adopting practice standards.
Supplementary Information
Footnotes
Competing interests: Tara Gomes reports funding from the Ontario Ministry of Health, the Public Health Agency of Canada, and the Ontario College of Pharmacists; payment for expert testimony from the Office of the Chief Coroner of Ontario; and travel support from Indigenous Services Canada. Dr. Gomes is a scientific advisor for the British Columbia Auditor General. Philippa Hawley reports funding from the BC Cancer Foundation; honoraria from Pallium, the University of BC, and various hospice societies; and travel support from the Canadian Cannabis Research Network. Rita McCracken reports consulting fees from Simon Fraser University. No other competing interests were declared.
This article has been peer reviewed.
Contributors: Dimitra Panagiotoglou conceived and designed the work. All of the authors contributed to data acquisition, analysis, and interpretation. Dimitra Panagiotoglou drafted the manuscript. All of the authors revised it critically for important intellectual content, gave final approval of the version to be published, and agreed to be accountable for all aspects of the work.
Funding: This project was supported by a Canadian Institutes of Health Research Project Grant (no. 461668). Dimitra Panagiotoglou holds a Tier 2 Canada Research Chair in the Economics of Harm Reduction. Tara Gomes holds a Tier 2 Canada Research Chair in Drug Policy and Substance Use. M. Ruth Lavergne holds a Tier 2 Canada Research Chair in Primary Care.
Data sharing: The individual-level, deidentified data were provided by Population Data BC. Access to the data are subject to approval, but can be requested for research projects through the Data Steward(s) or their designated providers. All inferences, opinions, and conclusions drawn in this publication are those of the authors, and do not reflect the opinions or policies of the Data Steward(s). Further information regarding these data sets can be found at the Population Data BC project webpage (https://my.popdata.bc.ca/project_listings/20-165/).
References
- 1.Unregulated drug deaths - summary. Burnaby (BC): British Columbia Coroners Service; 2024. Available: https://www2.gov.bc.ca/assets/gov/birth-adoption-death-marriage-and-divorce/deaths/coroners-service/statistical/augsept2024_2_unregulated_drug_pdf_of_dashboard.pdf (accessed 2025 Apr. 17). [Google Scholar]
- 2.Safe prescribing of drugs with potential for misuse/diversion. Vancouver: College of Physicians and Surgeons of British Columbia; 2016. [Google Scholar]
- 3.Dowell D, Haegerich TM, Chou R. CDC guideline for prescribing opioids for chronic pain — United States, 2016. JAMA 2016;315:1624–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.FAQ safe prescribing of opioids and sedatives. Vancouver: College of Physicians and Surgeons of British Columbia; 2018. Available: https://www.cpsbc.ca/files/pdf/PSG-Safe-Prescribing-FAQ.pdf (accessed 2025 Apr. 17). [Google Scholar]
- 5.Safe and effective use of opioids for chronic non-cancer pain. Hamilton: National Opioid Use Guideline Group; 2010. Available: https://www.cpd.utoronto.ca/opioidprescribing/files/2016/11/opioid_guideline_part_b_v5_6.pdf (accessed 2025 Apr. 17). [Google Scholar]
- 6.Chou R, Deyo R, Devine B, et al. The effectiveness and risks of long-term opioid treatment of chronic pain. Evid Rep Technol Assess (Full Rep) 2014;(218):1–219. [DOI] [PubMed] [Google Scholar]
- 7.Busse JW, Wang L, Kamaleldin M, et al. Opioids for chronic noncancer pain: a systematic review and meta-analysis. JAMA 2018;320:2448–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Oetter HM. Clarification from the College of Physicians and Surgeons of BC on commentary about limitations of the CDC guideline for prescribing opioids [letter]. CMAJ 2017;189:E508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Crabtree A, Rose C, Chong M, et al. Effects of the new prescribing standards in British Columbia on consumption of opioids and benzodiazepines and z drugs. Can Fam Physician 2019;65:e231–7. [PMC free article] [PubMed] [Google Scholar]
- 10.Schieber LZ, Guy GP, Jr, Seth P, et al. Variation in adult outpatient opioid prescription dispensing by age and sex — United States, 2008–2018. MMWR Morb Mortal Wkly Rep 2020;69:298–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Fulton-Kehoe D, Sullivan MD, Turner JA, et al. Opioid poisonings in Washington State Medicaid: trends, dosing, and guidelines. Med Care 2015;53:679–85. [DOI] [PubMed] [Google Scholar]
- 12.Pritchard KT, Baillargeon J, Lee WC, et al. Trends in the use of opioids vs non-pharmacologic treatments in adults with pain, 2011–2019. JAMA Netw Open 2022;5:e2240612. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Townsend T, Cerdá M, Bohnert A, et al. CDC guideline for opioid prescribing associated with reduced dispensing to certain patients with chronic pain. Health Aff (Millwood) 2021;40:1766–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Tucker J, Salas J, Zhang Z, et al. Provider specialty and odds of a new codeine, hydrocodone, oxycodone and tramadol prescription before and after the CDC opioid prescribing guideline publication. Prev Med 2021;146:106466. [DOI] [PubMed] [Google Scholar]
- 15.Bandara S, Bicket MC, McGinty EE. Trends in opioid and non-opioid treatment for chronic non-cancer pain and cancer pain among privately insured adults in the United States, 2012–2019. PLoS One 2022;17:e0272142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Salas J, Li X, Xian H, et al. Opioid dosing among patients with 3 or more years of continuous prescription opioid use before and after the CDC opioid prescribing guideline. Int J Drug Policy 2021;97:103308. [DOI] [PubMed] [Google Scholar]
- 17.Kroenke K, Alford DP, Argoff C, et al. Challenges with implementing the Centers for Disease Control and Prevention opioid guideline: a consensus panel report. Pain Med 2019;20:724–35. [DOI] [PubMed] [Google Scholar]
- 18.Voon P, Callon C, Nguyen P, et al. Denial of prescription analgesia among people who inject drugs in a Canadian setting. Drug Alcohol Rev 2015;34:221–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Busse JW, Craigie S, Juurlink DN, et al. Guideline for opioid therapy and chronic noncancer pain. CMAJ 2017;189:E659–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Antoniou T, Ala-Leppilampi K, Shearer D, et al. “Like being put on an ice floe and shoved away”: a qualitative study of the impacts of opioid-related policy changes on people who take opioids”. Int J Drug Policy 2019;66:15–22. [DOI] [PubMed] [Google Scholar]
- 21.Morin KA, Eibl JK, Franklyn AM, et al. The opioid crisis: past, present and future policy climate in Ontario, Canada. Subst Abuse Treat Prev Policy 2017;12:45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Dassieu L, Kabore JL, Choiniere M, et al. Chronic pain management among people who use drugs: a health policy challenge in the context of the opioid crisis. Int J Drug Policy 2019:71:150–6. [DOI] [PubMed] [Google Scholar]
- 23.Darnall BD, Juurlink D, Kerns RD, et al. International stakeholder community of pain experts and leaders call for an urgent action on forced opioid tapering. Pain Med 2019;20:429–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Safe prescribing of opioids and sedatives. Vancouver: College of Physicians and Surgeons of British Columbia; 2018. (accessed 2018 Oct. 4). [Google Scholar]
- 25.Cook TD, Campbell DT. Quasi-experimentation: design analysis issues for field settings. Boston: Houghton Mifflin Company; 1979. [Google Scholar]
- 26.PharmaNet. V2. Victoria (BC): British Columbia Ministry of Health; 2022. [Google Scholar]
- 27.Discharge Abstract Database (hospital separations). Ottawa: Canadian Institute for Health Information; 2022. [Google Scholar]
- 28.Medical Services Plan (MSP) payment information file. Victoria (BC): British Columbia Ministry of Health; 2022. [Google Scholar]
- 29.National Ambulatory Care Reporting System. Ottawa: Canadian Institute for Health Information; 2022. [Google Scholar]
- 30.BC Cancer Registry. Vancouver: BC Cancer; 2022. [Google Scholar]
- 31.Vital events — deaths. British Columbia Ministry of Health; 2022. [Google Scholar]
- 32.Jones W, Kaoser R, Rudoler D, et al. Trends in dispensing of individual prescription opioid formulations, Canada 2005–2020. J Pharm Policy Pract 2022;15:27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Jones W, Kaoser R, Fischer B. Patterns, trends and determinants of medical opioid utilization in Canada 2005–2020: characterizing an era of intensive rise and fall. Subst Abuse Treat Prev Policy 2021;16:65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Westra J, Raji M, Kuo YF. National patterns of cessation of prescription opioids among Medicare beneficiaries, 2013–2018. Medicine (Baltimore) 2022;101:e29944. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Beaugard CA, Chui KKH, Larochelle MR, et al. Abrupt discontinuation from long-term opioid therapy in Massachusetts, 2015–2018. Am J Prev Med 2022;62:404–13. [DOI] [PubMed] [Google Scholar]
- 36.Zhu W, Chernew ME, Sherry TB, et al. Initial opioid prescriptions among U.S. commercially insured patients, 2012–2017. N Engl J Med 2019;380:1043–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Kennedy MC, Crabtree A, Nolan S, et al. Discontinuation and tapering of prescribed opioids and risk of overdose among people on long-term opioid therapy for pain with and without opioid use disorder in British Columbia, Canada: a retrospective cohort study. PLoS Med 2022;19:e1004123. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Covington EC, Argoff CE, Ballantyne JC, et al. Ensuring patient protections when tapering opioids: consensus panel recommendations. Mayo Clin Proc 2020;95:2155–71. [DOI] [PubMed] [Google Scholar]
- 39.Maharjan S, Ramachandran S, Bhattacharya K, et al. Opioid tapering and mental health crisis in older adults. Pharmacoepidemiol Drug Saf 2024;33:e5698. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Narayan S, Brath H, Di Marco D, et al.“I’m almost opioid-a-phobic”: family medicine residents’ perceptions of enhancing opioid analgesic and agonist treatment training in a Canadian setting. Educ Prim Care 2013;34:161–7. [DOI] [PubMed] [Google Scholar]
- 41.Access to palliative care in Canada, 2023. Ottawa: Canadian Institute for Health Information; 2023. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.