CIHR Assessment of the Canadian Guideline for Opioids for Chronic Non-Cancer Pain

Table of contents

  1. Context
  2. Overview of the Assessment
  3. Review and Observations
    1. Review of the Guideline Development Process
    2. Internal Review of the CMAJ pre-publication process
    3. External assessment of the Scientific Merit of the 2017 Canadian Guideline
  4. Conclusion
  5. Appendices

1. Context

Canada is experiencing a dramatic rise in opioid-related deaths. In 2016 alone, almost 2,500 opioid-related deaths occurred across CanadaFootnote 1. Likewise, the number of opioid prescriptions filled in Canada each year is increasingFootnote 2. In Ontario alone, the number of opioid prescriptions filled increased by five per cent over three years (2013-14 to 2015-16) while the number of people who filled at least one of these prescriptions remained approximately the sameFootnote 3. Inappropriate opioid prescribing practices is one factor contributing to this growing crisis as cases arise of patients becoming addicted while under the care of a physician.

In 2010, recognizing the need to improve safe opioid prescribing practices, a national group of physicians and researchers developed the Canadian Guideline for Safe and Effective Use of Opioids for Chronic Non-Cancer Pain. This guideline provided recommendations to medical professionals on prescribing opioids safely and effectively. Once the 2010 guideline was published, McMaster University’s Michael G. DeGroote National Pain Centre assumed responsibility for keeping it current by highlighting new evidenceFootnote 4. In the years following, the Government of Canada provided funding, through the Canadian Institutes of Health Research and Health Canada, to McMaster University researchers to update the 2010 guideline with new evidenceFootnote *.

The revised 2017 Canadian Guideline for Opioids for Chronic Non-Cancer Pain (the 2017 Canadian Guideline) was published in the Canadian Medical Association Journal (CMAJ) on May 8, 2017. Following its release, The Globe and Mail reported that McMaster University did not exclude medical experts who had received income from opioid manufacturers from the 2017 Canadian Guideline’s voting panelFootnote 5. In particular, the article noted the potential conflict of interest from one panel member, Dr. Sol SternFootnote 6.

2. Overview of the Assessment

In response to public criticism, the Minister of Health directed CIHR to assess what impact, if any, the potential conflict of interest had on the scientific rigour of the 2017 Canadian Guideline. This review aimed to determine whether the 2017 Canadian Guideline provides unbiased, evidence-based guidance to clinicians on opioid prescribing practices.

The CIHR assessment of the 2017 Canadian Guideline consisted of the following three components:

  1. a review of the process used by McMaster University to develop the 2017 Canadian Guideline
  2. a review of the process used by CMAJ to make publication decisions; and
  3. a comparison between the Guideline and an international equivalent.

The assessment was led by the Scientific Directors of the CIHR Institute of Health Services and Policy Research (Dr. Robyn Tamblyn) and the CIHR Institute of Musculoskeletal Health and Arthritis (Dr. Hani El-Gabalawy). The third task involved two Directors from the United States’ National Institutes of Health (NIH): the Director of the National Institute on Drug Abuse (Dr. Nora Volkow) and the Director of the National Institute of Neurological Disorders and Stroke (Dr. Walter Koroshetz).

Dr. Robyn Tamblyn

Scientific Director, CIHR Institute of Health Services and Policy Research

Dr. Robyn Tamblyn of McGill University was selected because of her extensive experience in health policy research. Dr. Tamblyn's ground-breaking research on educational outcomes have clarified the relationships between health professional training, licensure, and practice and subsequently guided credentialing policies. In addition, her work on prescription drug use, its determinants, and ways to improve drug safety has been internationally recognized.

Dr. Hani El-Gabalawy

Former Scientific Director, CIHR Institute of Musculoskeletal Health and Arthritis

Dr. Hani El-Gabalawy of the University of Manitoba was selected because of his extensive expertise in pain management. Dr. El-Gabalawy, an internationally recognized rheumatologist, is a Professor of Medicine and Immunology and a senior clinician-scientist at the University of Manitoba where he holds the Endowed Rheumatology Research Chair.

Dr. Nora Volkow

Director, NIH National Institute on Drug Abuse

Dr. Nora Volkow was selected because of her role as the head of the United States’ National Institute on Drug Abuse and her extensive research experience. Her work was instrumental in demonstrating that drug addiction is a disease of the human brain. She also documented changes in the dopamine system affecting, among others, the functions of frontal brain regions involved with motivation, drive, and pleasure in addiction.

Dr. Walter Koroshetz

Director, NIH National Institute of Neurological Disorders and Stroke

Dr. Walter Koroshetz was selected because of his role as the head of the United States’ National Institute of Neurological Disorders and Stroke and his role as Chair of the National Institutes of Health Pain Consortium. A major focus of Dr. Koroshetz’ clinical research career was to develop measures in patients that reflect the underlying biology of their conditions. With the Massachusetts General Hospital team, he discovered increased brain lactate in Huntington’s Disease patients using MR spectroscopy. He helped the team to pioneer the use of diffusion/perfusion-weighted MR imaging and CT angiography/perfusion imaging in acute stroke.

Their full biographies are included in Appendix A.

3. Review and Observations

i) Review of the Guideline Development Process

The first element of the CIHR assessment was to review the process McMaster University used to develop the 2017 Canadian Guideline. This review examined what impact, if any, Dr. Stern’s perceived conflict of interest had on the scientific rigour of the 2017 Canadian Guideline.
Dr. Tamblyn and Dr. Michelle Peel, Acting Director General of CIHR’s Science, Knowledge Translation and Ethics branch, held a key informant interview with Dr. Jason Busse of McMaster University and primary editor of the 2017 Canadian Guideline. The objectives of this discussion were to review the overall development process and discuss whether this was sufficient to ensure that the 2017 Canadian Guideline is both evidence-based and free from bias.

The development process

The process to develop the 2017 Canadian Guideline began by synthesizing the recommendations from the 2010 version of the guideline and six other recent guidelinesFootnote 7Footnote 8Footnote 9Footnote 10Footnote 11Footnote 12. The team presented this synthesis at a national stakeholder meeting with representatives from patient advocacy groups, governments, medical regulators, the research community, the medical community, and law enforcement. Meeting attendees identified 24 areas where clinical practice recommendations would be helpful.

Governance structure

The research team established four guideline development groups (Figure 1), including one voting panel:

Figure 1: Process for Generating the 2017 Canadian Guideline for Opioids for Chronic Non-Cancer Pain

Figure 1 long description

The process for generating the 2017 Canadian Guideline for Opioids for Chronic Non-Cancer Pain was led by the four-person Guideline Steering Committee. This committee was supported by three other groups: 1) a 16-member patient advisory committee that provided advice on patient values and preferences, 2) a 13-member clinical advisory committee that provided advice on opioid use for chronic pain, and 3) an evidence synthese/grading group. Finally, there was a 15-member voting panel comprised of clinicians and patients that voted on the final recommendations that made up the 2017 Canadian Guideline.

To inform the design of the 2017 Canadian Guideline development process, the research team recruited an experienced guideline methodologist, Dr. Gordon Guyatt of McMaster University. Dr. Guyatt, who became the Chair of the Guideline Steering Committee, had been significantly involved in many iterations of the American College of Chest Physicians Evidence-Based Clinical Practice Guidelines on Antithrombotic Therapy and Prevention of Thrombosis, most recently as chair of the ninth edition. The design of the development process was also informed by Clinical Practice Guidelines We Can Trust which was published by the Institute of Medicine, a division of the National Academies of Sciences, Engineering and Medicine of the United States.

Making recommendations

In the Guideline Steering Committee’s view, to be useful, clinical practice recommendations need to be informed not only by evidence but also by context, including patient values and preferences. The Guideline Steering Committee therefore developed a values and preferences statement to guide all recommendations.Footnote 13 This statement was based on a systematic review of the literature on patient values and preferences regarding opioids for chronic non-cancer pain and on focus group interviews with the Patient Advisory Committee.

The Guideline Steering Committee also determined that the panel would only make clinical practice recommendations if there was sufficient evidence to inform the recommendations. Systematic reviews of the evidence were conducted and the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system was applied to move evidence to recommendations. This system is an approach used internationally to grade the quality (or certainty) of evidence and strength of recommendations. As a result, the final 2017 Canadian Guideline included three categories of guidance: “recommendations (supported by evidence from randomized controlled trials or observational studies), good practice statements (supported by indirect evidence), and expert guidance (supported by little or no published evidence).”Footnote 14

After the systematic reviews of the literature were complete, the Guideline Panel and the Clinical Expert Committee met to discuss the evidence and clarify issues for which there was no, or limited, research evidence on which the panel could make its recommendations. In a second meeting of only the Guideline Panel, members voted on 10 recommendations on which there was sufficient evidence. Recommendations required 80% consensus amongst panel members in order to avoid the possibility that a few members could have undue influence on the process. If any panel member disagreed they would record a formal dissenting statement; however, no requests for dissenting statements occurred. The Guideline Panel came to an agreement on all 10 recommendations.

These recommendations were posted online for a one-month public review process. The Guideline Steering Committee reviewed over 500 responses and, in response, revised the language of some recommendations. There were, however, no cases where the public feedback changed the direction (for or against) or strength (weak or strong) of any recommendation.

The final recommendations were presented in a summary paper to the CMAJ. The authors replied to reviewer comments and then the manuscript and guidelines were accepted for publication.

Conflict of Interest

The Institute of Medicine recommends that in the process of guideline development, fewer than 50% of voting members should have financial conflicts of interest (COI).Footnote 15 The Guideline Steering Committee set a higher threshold for managing COI for this guideline. The intent was that the voting Guideline Panel would be composed entirely of individuals without any financial or intellectual conflicts. This was largely successful; no members had any declared intellectual COI and only one of fifteen had a material financial COI.

The 2017 Canadian Guideline required input from both patients and clinical experts, but it was recognized that bias would be prevalent among these groups. As such, the Guideline Steering Committee strove to ensure an equal balance in views among members of both the Clinical Expert Committee and the Patient Advisory Committee, but did not allow these groups to vote on the recommendations or to be present during the voting process.

Due to an administrative oversight, there was one individual on the voting Guideline Panel, Dr. Stern, with a material financial COI; he had received payment for providing medical education talks for industry. Though this was declared prior to voting, it was not brought to the attention of the Guideline Steering Committee through the administrative COI process. After the 2017 Canadian Guideline was developed, the Guideline Steering Committee identified the COI when reviewing the COI forms it collected for submission to the CMAJ. According to Dr. Busse, had this been identified earlier, Dr. Stern would have been proposed for the non-voting Clinical Expert Committee and excluded from the voting Guideline Panel.  

Once the COI was identified, the Guideline Steering Committee met with the project coordinator who collected and administered the COI forms and reviewed Dr. Stern’s participation in the development process. The Guideline Steering Committee concluded that there was no evidence of bias during Dr. Stern’s participation. It must be noted that the 2017 Canadian Guideline provides recommendations to avoid prescribing opioids and to prescribe at lower thresholds. In addition, 7 of 10 recommendations focus on harm reduction. These recommendations would not generally be seen to favour industry.

CIHR Observations

In reviewing the process McMaster University used to develop the 2017 Canadian Guideline, CIHR did identify a flaw in how conflict of interest was handled. The COI process was weakened by depending on a single individual (i.e., the project coordinator) to identify potential conflicts. Given the process used to develop the 2017 Canadian Guideline and the entirety of the evidence its committees examined, there is no evidence that a COI did or could have had any real influence on the final recommendations. CIHR found that the 2017 Canadian Guideline development process was rigorous, and concluded that Dr. Stern’s conflict did not ultimately impact the 2017 Canadian Guideline’s recommendations.

ii) Internal Review of the CMAJ pre-publication process

The second element of the CIHR assessment was to review the CMAJ process (Figure 2) that led to the publication of 2017 Canadian Guideline in May 2017.

Figure 2: CMAJ Editorial Process

Figure 2 long description

The editorial process for submitting research and guideline articles to the CMAJ is as follows:

  1. A researcher submits a manuscript
  2. Editors at the CMAJ do an internal review, which results in either a rejection, revision, or advancing to the peer-review stage.
  3. At the peer review stage, reviewers assess the article, leading to either rejection or a manuscript meeting.
  4. Following the manuscript meeting, the article is either rejected, accepted, revised, or undergoes further peer review.
  5. The last stage is internal review or a manuscript meeting where, after further scrutiny, the article is either rejected, accepted, or revised for the final time.

As part of its review, a senior CIHR official interviewed Dr. Diane Kelsall, Interim Editor-in-Chief of CMAJ, on the journal’s review of conflicts of interest. Prior to accepting the 2017 Canadian Guideline for publication, CMAJ requested authors to complete an International Committee of Medical Journal Editors (ICMJE) form for Disclosure of Potential Conflicts of Interest to disclose any perceived or real conflicts of interest.

The CMAJ reviewed the completed ICMJE forms and acknowledged that perceived or real conflicts existed. Nevertheless, CMAJ felt it was an important guideline to publish. It proceeded to publish the 2017 Canadian Guideline and included the disclosures in the publication.

CIHR Observations

While the conflict of interest processes used by CMAJ and McMaster University were similar, the CMAJ process did use a longer time period to capture self-declared potential conflicts. However, since Dr. Stern declared the potential conflict on both the CMAJ and McMaster University forms, CIHR does not feel the decision to use one form over another had any bearing on the scientific rigour of the 2017 Canadian Guideline.

Following its review, CIHR is confident in the CMAJ process that led to the publication of the 2017 Canadian Guideline. The process identified potential conflicts of interest, considered their weight in deciding to publish, and disclosed them in the publication. This process met international standards for assessing the scientific merit of a publication in a peer-reviewed journal.

iii) External assessment of the Scientific Merit of the 2017 Canadian Guideline

The third element of the CIHR assessment was an independent comparative analysis between the 2017 Canadian Guideline and an internationally-recognized guideline on the same topic. CIHR invited Dr. Nora Volkow, Director of the U.S. NIH National Institute on Drug Abuse, to provide an expert opinion on how the 2017 Canadian Guideline recommendations compare to those developed by the Centre for Disease Control (CDC) in the United States. The CDC Guideline for Prescribing Opioids for Chronic PainFootnote 16 was developed in 2016 to provide recommendations for prescribing opioid pain medication in primary care settings. The CDC Guideline has been endorsed by the U.S. Food and Drug Administration and a number of Canadian colleges of physicians and surgeons (e.g., AlbertaFootnote 17, British ColumbiaFootnote 18, Nova ScotiaFootnote 19). Like the 2017 Canadian Guideline, the CDC Guideline provides recommendations on using opioids to manage non-cancer pain in adults (18 years of age or older) that has persisted for three months or more.  

Under Dr. Volkow’s leadership, experts from the National Institute on Drug Abuse (NIDA) and the National Institute on Neurological Disorders and Stroke (NINDS) reviewed the 2017 Canadian Guideline, compared it to the CDC Guideline, and provided some feedback on the 2017 Canadian Guideline for consideration (included in Appendix B). The experts found that the methodology and extent of evidence reviewed to develop both guidelines were similar. In each case, development teams conducted systematic reviews and used standardized categorization of the strength of recommendations. The experts also noted that while some differences exist, the similarities indicate a rigorous development process. The 2017 Canadian Guideline is consistent with available evidence and can be used as a resource in clinical practice.

In addition to the similarities in the development process, the experts identified several similarities in the recommendations of the two guidelines. Both guidelines recommend that opioids should not be the first-line therapy. They also recommend the use of the lowest effective dosage to initiate opioids and careful titration. In addition, the guidelines both further suggest limits on daily morphine of 90 milligram equivalents with special consideration for using larger amounts. Finally, both guidelines recommend co-prescribing naloxone with at-risk patients and that opioids should be tapered and discontinued if they are not effective.

The experts noted only two specific differences. First, that the 2017 Canadian Guideline suggests that controlled release opioids for continuous pain while the CDC Guideline recommends avoiding their use overall and is against their use to initiate opioid treatment. Secondly, the 2017 Canadian Guideline strongly recommends against the use of opioids in patients with active substance use disorder (SUD) and suggests using a non-opioid therapy rather than a trial of opioids. The CDC Guideline, however, does not recommend against the use of opioids in patients with active or history of SUD. Instead, it outlines risk assessment for SUD and recommends providing medication-assisted treatment for those identified with SUD (e.g., co-prescribing naloxone).

The NIDA and the NINDS are of the view that the 2017 Canadian Guideline is based on a rigorous developmental process similar to that used by the Centers for Disease Control and Prevention’s (CDC). Notwithstanding certain differences noted and commentary provided for future consideration, the 2017 Canadian Guideline is consistent with available evidence for clinical practice.

CIHR Observations

In reviewing the comparative analysis submitted by Dr. Volkow, CIHR noted the similarities in the development processes of the two guidelines. It further observed the parallels in the broad strokes of the two guidelines, particularly as related to a preference for non-pharmacologic therapy and non-opioid pharmacologic therapy, daily dosage caps, tapering and co-prescribing naloxone.

Despite the differences mentioned above, CIHR does not feel these reduce the merit of the 2017 Canadian Guideline as a whole. For example, the 2017 Canadian Guideline’s comment on controlled release opioids is not a GRADE recommendation, but one of several guidance statements that provide direction in areas where little published evidence is available. The 2017 Canadian Guideline authors openly note that “the benefit and safety of controlled release or sustained release over immediate release preparation is not clearly established.”  

In addition to the comparative analysis, Dr. Volkow provided additional suggestions to inform future iterations of the 2017 Canadian Guideline (refer to Appendix B). CIHR has taken the liberty of providing these to officials at Health Canada under separate cover.

4. Conclusion

Following the three components of its assessment, CIHR finds that the process used to develop the 2017 Canadian Guideline for Opioids for Chronic Non-Cancer Pain was scientifically rigorous despite the flaw in promptly identifying the COI in a single voting member of the fifteen member panel.

Prior to publication, the 2017 Canadian Guideline was subject to many checks and balances in order to minimize undue influence on the final recommendation. These checks included the multi-step process for drafting recommendations, the consensus-based voting process, and the rigorous process to apply for the journal. Coupled with the comparison with the CDC Guideline, it seems clears that the perceived COI of one individual on the voting Guideline Panel did not have any impact on the final recommendations. CIHR concludes that the 2017 Canadian Guideline does provide unbiased, evidence-based guidance to clinicians on opioid prescribing practice that is aligned with international comparators.

Appendices

Appendix A – Biographies

The four experts declared not having any competing interests in their participation in this review.

Appendix B – National Institutes of Health (US) Review of Canadian Guideline for Opioids for Non-Cancer Pain (August 10, 2017)

Summary

Experts from the National Institute on Drug Abuse (NIDA) and the National Institute on Neurological Disorders and Stroke (NINDS) reviewed the draft [2017] Canadian Guideline for Opioids for Non-Cancer Pain, conducted a comparison with those developed by the Centers for Disease Control and Prevention (CDC), and provided recommendations for consideration. Overall, the draft guideline is based on a rigorous developmental process utilizing a systematic review and standardized categorization of recommendations similar to that used by the CDC. Notwithstanding certain differences noted and recommendations provided for future consideration, the [2017] Canadian Guideline is consistent with available evidence for clinical practice.

Comparison with CDC guidelines

Both sets of guidelines are focused on chronic non-cancer pain and non-end of life pain in adults. Both recommend non-opioid, non-pharmacological approaches prior to initiating opioids for chronic pain in opioid naïve patients, and in combination with patients who are prescribed opioids. The methodology and expanse of evidence reviews were similar with systematic reviews conducted and standardized categorization of the “strength” of recommendations.

Specific differences
Specific similarities

Feedback for Consideration

Format and Organization
Date modified: