Reducing and managing conflicts of interest in clinical practice guideline development: do we need Pan-Canadian standards?

The Canadian Institutes of Health Research in collaboration with the Public Health Agency of Canada

Objectives


On January 23, 2019, CIHR hosted a Best Brains Exchange (BBE) on the topic of Reducing and managing conflicts of interest in clinical practice guideline development: do we need Pan-Canadian standards?in collaboration with the Public Health Agency of Canada.

The Best Brains Exchange (BBE) was designed to address the following policy questions:

  1. What sources and types of conflict of interest (COI) exist in clinical practice guideline (CPG) development?
  2. What are some of the barriers and facilitators to reducing and managing COI in CPG development?
  3. Which approaches could be used to reduce and manage COI by Canadian CPG developers, and how should these be implemented and disseminated to maximize impact?

Policy context

CPGs are important tools for helping practitioners deliver high quality and cost-effective care. When developed under transparent, rigorous, and evidence-based protocols, and faithfully implemented, guidelines can directly influence patient care, and therefore, the health of Canadians. CPGs are increasingly popular: the Canadian Medical Association has a database of over 1,200 CPGs that were developed or endorsed by Canadian medical or health organizations. Outside of Canada, guidelines are being produced by organizations around the world. The Guidelines International Network (G-I-N) supports a global network of 105 organizations representing 53 countries from all continents, who are all interested in CPG development.

To be maximally useful, CPG should be free of real or perceived COI. In this context, COI refers to situations in which the judgement of an individual involved in developing the CPG can be unduly influenced by an opportunity to derive personal benefit (financial or non-financial). Real COI can create guidelines that are biased, usually overestimating benefit and underestimating harm of health interventions. Perceived and real COI can undermine the credibility of guidelines thereby wasting resources and potentially affecting patient care. On the other hand, guideline developers face a dilemma as in many fields, most or all subject experts will have a potential financial or non-financial COI. There is therefore often tension between the goal of eliminating COI and developing guidelines informed by the most knowledgeable experts.

The standards, principles, and/or policies for mitigating the effects of COI in CPGs vary amongst jurisdictions. Federal, provincial, and territorial governments commission and/or support the development of various CPGs, but specific official policies for identifying and managing COI are not always present or explicit. Similarly, many non-government CPG developers have their own individual policies and procedures for managing COI with varying features. Management of COI in CPGs produced in Canada is thus an inconsistent patchwork of organization-specific policies and protocols.

Identified need for evidence

It is in Government’s interest to ensure CPGs developed in Canada are trustworthy and have high utility. Identifying principles by which COI should be disclosed, reported and managed would improve the acceptability, credibility, and usability of CPGs and thus increase the return on investment associated with these products.

As mentioned above, various policies/procedures exist among CPG-developing organizations. Some strategies that have been suggested for dealing with COI in CPGs have included zero tolerance, recusal at voting for specific recommendations, non-participation in discussion for specific recommendations, including the contribution of individuals on opposite sides of an issue (balancing conflicts), among others.

PHAC and its partners identified a need to hear evidence and experiences from a range of jurisdictions and organizations regarding the challenges and impacts of COI in CPG development. By bringing experts and key stakeholders together, policy makers were given greater access to a range of perspectives and practical solutions for consideration. PHAC hopes to use the evidence and connections derived from the BBE to facilitate the uptake of effective community driven solutions for identifying and managing COI in CPG development.

Invited participants included experts in CPG production, representatives of major Canadian CPG producers, and end-users such as federal and provincial policy makers, medical professionals and Canadian medical journals.

Anticipated outcomes

The Public Health Agency of Canada will continue to engage with participants and other organizations in the guideline community to develop concrete solutions to reduce conflict of interest in clinical practice guideline development.

Presentation summaries

The meeting was facilitated by Terrence Sullivan (Principal, Terrence Sullivan and Associates). Below is a summary of the evidence shared by each of the invited presenters:

What sources and types of conflict of interest (COI) exist in clinical practice guideline (CPG) development?

Elie Akl, MD, MPH, PhD, Professor, Department of Internal Medicine, American University of Beirut (LB)

Multiple definitions of conflict of interest (COI) exist. The following represents a synthesis of those definitions with a focus on operationalization: a specific and relevant relationship or attribute that creates the risk of biasing a judgment, decision, or action being made in the context of a specific duty/responsibility, with a possible personal gain. In the context of guidelines, the following groups need to declare their COIs: members of oversight committees, systematic reviewers, panelists, editors, and peer reviewers. A COI can be at either the individual or the institutional level, and of a financial, intellectual, personal or cultural nature. An organization needs to judge whether a relationship (or an attribute) is relevant to the expected task in order to determine whether it represents a COI. Ideally it would also assess the risk of bias associated with COIs to guide their management plan. Finally, an organization should consider whether to verify declared COIs given relatively high rates of declarations that are inaccurate.

Competing interests in guidelines: an editor’s perspective

Diane Kelsall, MD, MEd, Editor-in-Chief (interim), Canadian Medical Association Journal

Over the past several decades, journals have made major strides in understanding the impact and management of competing interests in submitted work. General guidance for editors on competing interests has been developed by groups such as the Committee on Publication Ethics, International Committee of Medical Journal Editors, World Association of Medical Editors and the Council of Science Editors. However, undisclosed or poorly managed competing interests continue to affect the published record, including that of clinical practice guidelines. Additionally, journals may develop their own policies, which may not be clear to other stakeholders including authors, reviewers and readers, or be consistent with established guidance for editors. Of particular interest to editors currently are the following issues: 1. Reliance on trust in disclosure of competing interests; 2. Disclosure versus minimization of competing interests and which components are essential to minimize competing interests; and 3. The move to handling purposeful failure to disclose competing interests as academic misconduct.

Conflict of Interest and the Quality of Clinical Practice Guidelines

Joel Lexchin, MS, MD, Associate Professor, Faculty of Medicine, University of Toronto

Financial conflict of interest (fCOI) is widespread among committee members who produce Canadian clinical practice guidelines. Limited evidence shows an association between fCOI and biases in guidelines. This finding of bias as a result of fCOI is backed up by more extensive research into the source of sponsorship of clinical trials and their results and conclusions, and the results of randomized controlled trials when authors have a fCOI. Although there are also non-financial COIs, they are qualitatively different from fCOI and do not lead to the same widespread biases. There are two approaches to dealing with fCOI, managing conflicts or avoiding them. American literature shows that most guidelines ignore one or more of the recommendations from the National Academy of Medicine on how fCOI should be managed and many authors fail to declare their conflicts in guideline documents. Taking a precautionary principle approach dictates that fCOI in the creation of guidelines should be avoided as much as possible and not just managed.

Reducing and managing conflicts of interest in guideline development: experience from NICE

Gillian Leng, MBChB, MD, CBE, Deputy Chief Executive and Director of Health and Social Care, National Institute for Health and Care Excellence (UK)

NICE’s policy on declaring and managing conflicts of interest came into effect in April 2018. This was an update of an earlier policy, designed to provide consistency with the approach to managing conflicts of interest in the NHS, and with the Guidelines International Network.  New components included more rigorous handling of private practice, publishing committee members’ interests on the NICE website, and establishing a reference panel to give an independent response to challenging areas. 
Implementation of the policy required support in a number of ways, including training of staff, website development, recruitment to the reference panel and development of a new form for recording interests with a dedicated question and answer sheet.  It was agreed that impact of the new policy would be reviewed after 12 months. 

Initial feedback from guideline developers suggests that the restrictions around private practice may be reducing the number of eligible candidates for committee membership, so this will be reviewed in more detail.  The other major challenge appears to be encouraging committee members to declare all relevant interests, and a particular issue has arisen in relation to financial interests of patient organisations.  All feedback will be considered by the NICE Board later in the year.

Conflict of Interest Prevention: Health Expertise and the  Haute Autorité de Santé (HAS) in France 

Daniel Ludet, DES Public Law, National School for the Judiciary, Compliance Officer, Haute Autorité de Santé (France)
To ensure that health expertise is both independent and impartial, French law does now allow the involvement of a person with an interest in the matter and requires a compliance officer to oversee each health agency to prevent conflicts of interest. The HAS, an independent public authority, consults 2,500 experts to assess medicinal products, medical devices and professional acts for reimbursement purposes and to develop recommendations for good clinical and public health practices. Preventing conflicts of interest relies on analyzing information, such as information included in public declarations of interest that potential experts are required to complete and information from declarations that health industries must complete regarding the remuneration and benefits agreed upon by health professionals. This information is made public on two government sites: DPI Santé and Transparence Santé. At HAS, this information is analyzed by a validation of declarations of interest committee (Comité de validation des declarations d’intérêts, CVDI), chaired by the compliance officer, who conducts an analysis of all potential experts of a commission or working group. The analysis of each potential expert’s ties and relationships takes into account the specific subject matter of the expertise, how long the ties and relationships have existed, whether they are isolated or regularly occurring, and the amounts they represent. The goal is to determine whether there is a reasonable doubt that the person may be impartial. Please note that the existence alone of one or more ties and relationships of a person does not necessarily indicate that there is a conflict of interest. A potential conflict is detected only when the person’s ties and relationships are compared to the task for which they are being considered. The CVDI may rule in favour or in favour with a relative withdrawal requirement, or it may rule against recruiting. Intellectual independence of expertise is also ensured by preventing executive members of learned societies or various associations from participating, and by verifying that the candidate does not have ties or relationships though previous work. Certain challenges remain concerning whether the management of conflicts of interest is effective in the actual delivery of expert services, as well as the high demand for HAS services. Late information about a conflict of interest might discredit a recommendation that has already been adopted and published.

Declaration of interests and conflict of interest management for guideline development at WHO

Susan Norris, MD, MSc, Guidelines Review Committee, World Health Organization

The World Health Organization (WHO) produces a large variety of guidelines that meet the highest international standards for trustworthy and impactful guidelines.  WHO has a robust conflict of interest (COI) policy, updated in 2014, which is applicable to all the Organization’s normative work, including guidelines. WHO defines COI as any interest that may affect or reasonably be perceived to affect the expert's objectivity and independence and/or create an unfair competitive advantage for the expert or persons or institutions with whom the expert has financial business interests.  The policy includes assessment of a standard declaration of interests (DOI) form. Relevant interests include: direct financial interests (personal business interests, research funding, intellectual property); interests of others (family members, other parties with substantially similar interests); potential biases from professional or intellectual interests (e.g. public statements made and positions held, involvement with tobacco industry); and unfair or competitive advantage (where confidential information could be an unfair advantage).

Collection and assessment of interests occurs whenever WHO is relying on the independent advice of an expert in order to make decisions or recommendations.  For guidelines this includes members of the Guideline Development Group, contracted methodologists and systematic reviewers, external technical experts, and peer reviewers. Further information on WHO’s COI policy and the DOI forms can be found online.

In addition to the DOI, WHO’s COI policy also mandates that WHO staff gather additional information in order to identify any obvious public controversies or interests that may lead to compromising situations for WHO or the expert concerned.  This generally occurs via internet searches and review of recent publications identified from bibliographic database searches.  Finally, WHO also publishes brief biographies of individuals considered for participation on WHO’s advisory bodies together with a description of the objectives of the meetings for public comment.  Senior WHO staff assess the DOI and the additional information gathered and make an assessment as to whether any of the interests and information may represent a COI based on the nature and relevance to the guideline, monetary amount of the interest, currency, and other factors. 

This information-gathering, assessment of any identified COI, and formulation of a management plan occur prior to the issuance of a formal invitation to the expert for participation in guideline development.  The management plan is one of: 1) full participation (no declared or identified relevant interests); 2) full participation with declaration of any interests deemed non-significant; 3) partial exclusion (interests related to one or more parts of the guideline are deemed to be significant necessitating recusal or various levels of non-participation for parts of the guideline process); or 4) exclusion from the guideline development process. 

Each WHO guideline must report who contributed and their DOI, assessment of those interests, and any management plan for COI that were identified. 

The Canadian Task Force on Preventive Health Care: Managing Conflicts of Interest

Brett Thombs, MA, PhD, Chair, Canadian Task Force on Preventive Health Care

The Canadian Task Force on Preventive Health Care is an independent organization, funded by the Public Health Agency of Canada, whose mandate is to develop clinical practice guidelines for primary and preventive health care, based on systematic analysis of the best available evidence. The Task Force strives to develop guidelines that are relevant, patient-centered, highly usable, and credible. Task Force credibility depends on its ability to demonstrate that it is independent from outside influences, including from government, medical specialty organizations, and industry. His presentation described steps that the Task Force takes to evaluate and manage conflict of interest among Task Force members, clinical and content experts who advise the Task Force, and peer reviewers who evaluate Task Force guidelines.

Recommended readings

  1. Schünemann et al. Guidelines International Network: Principles for Disclosure of Interests and Management of Conflicts in Guidelines. Annals of Internal Medicine. 2015; 164(7): 548-553.
  2. Ngo-Metzger et al. Conflicts of Interest in Clinical Guidelines: Update of U.S. Preventive Services Task Force Policies and Procedures. American Journal of Preventive Medicine. 2018; 54(1): S70-S80. Abstract available online.
  3. International Committee of Medical Journal Editors (2018). Recommendations on the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals. See also attached annotated version of the recommendations.
  4. National Institute for Health and Care Excellence (NICE). September 2014. Policy on Conflicts of Interest [ PDF (201 KB) - external link ].
  5. National Institute for Health and Care Excellence (NICE). April 2018. Policy on declaring and managing interests for NICE advisory committees [ PDF (668 KB) - external link ].
  6. Shnier, A., Lexchin, J., Romero, M., & Brown, K. Reporting of financial conflicts of interest in clinical practice guidelines: a case study analysis of guidelines from the Canadian Medical Association Infobase. BMC Health Services Research. 2016; 16(1): 383.
  7. World Health Organization (2014). WHO handbook for guideline development: Chapter 6 (2nd edition). World Health Organization.
Date modified: