Conflicts of Interest and the Trustworthiness of Clinical Practice Guidelines

Clear separation between the interests of industry and recommendations that affect people’s health is essential for public and professional confidence in the process. Clinical practice guidelines are a valuable tool in the practice of evidence-based medicine, objectively synthesizing evidence on clinical issues to provide recommendations that optimize patient care. Conflict of interest is an important source of bias in clinical practice guideline development and stems from undue financial or nonfinancial interests, which sway a clinical practice guideline author’s professional judgment. The International Committee of Medical Journal Editors (ICMJE) states that the appearance of a potential conflict of interest is as important as a true conflict of interest. Financial conflict of interest commonly arises from author relationships with or payment by industry, whereas nonfinancial conflict of interest occurs when “academic activities … create the potential for an attachment to a specific point of view that could unduly affect an individual’s judgement about a specific recommendation” (1). The importance of conflict of interest in clinical practice guideline development has been highlighted (2), and mitigation strategies have been developed. For example, the World Health Organization and Australia’s National Health and Medical Research Council have implemented policies on disclosure and management of financial conflict of interest for clinical practice guideline members. Several countries have databases that track physicians’ financial payment by industry rather than relying on self-report (3). Despite these efforts, a recent systematic review of 37 studies assessing financial conflict of interest in clinical practice guideline development reported that 45% of nearly 15,000 authors had at least one financial conflict of interest (2). In this issue of CJASN, Murayama et al. (4) report the outcome of a cross-sectional study examining the financial and nonfinancial conflicts of interest of the 142 authors of the Japanese Society of Nephrology’s 2018 evidence-based clinical practice guidelines for CKD. Previous studies of conflict of interest in clinical practice guideline development have relied on authors’ self-report. This is the first study to use a payment database reported by pharmaceutical companies. Nonfinancial conflict of interest was assessed by self-citation. First, guideline author information regarding demographics, self-reported conflict of interest statements, and publications was gathered by author survey, review of official institutional web pages, and online database searches (PubMed, Scopus, Google Scholar, Medical Online, and Ichu-shiWeb). Next, payments to each author between 2016 and 2019 from all 92 pharmaceutical companies affiliated with the Japan Pharmaceutical Manufacturers Association were identified from personal payment databases disclosed by each company; reporting to these databases is voluntary. Finally, to contrast the conflict of interest management strategies with those of other developed countries, the conflict of interest policies for guideline development were collated from professional nephrology organizations in Europe and North America. Regarding financial conflict of interest between 2016 and 2019, 125 (88%) of the 142 guideline authors studied received personal payment totaling over US$6.5 million from 56 pharmaceutical companies. The 4-year median payment per author was approximately $8000. Those authors in leadership positions received substantially higher payments; for example, the guideline chair and vice chair received over US$200,000 each. A formal Gini index calculation, a measure of inequality across a population, suggested that large payments were concentrated among certain authors. Guideline authors who were women received approximately one third the median payment received by men. Regarding nonfinancial conflict of interest, 83 (58%) of the 142 guideline authors were authors of work cited in the 2018 Japanese CKD guideline. Of 861 total references in the examined guideline, 69 (8%) were self-citations. On review of the available self-reported conflict of interest statements, there was no disclosure of nonfinancial conflict of interest by any author. Finally, when comparing Japanese conflict of interest policies with those of other developed countries, the authors found varying degrees of rigor by organization. For example, the European Dialysis and Transplant Association (EDTA) had the most transparent approach, requiring all past, present, and future financial conflicts of interest to be disclosed, but EDTA did not limit the proportion of guideline authors with conflicts of interest. The 2018 Japanese guideline under examination in this study did not report any financial conflict of interest for guideline authors; the Japanese Society of Nephrology stated that the conflict of interest policy was in transition in 2018, and thus, it was out of scope and not applied. Financial conflict of interest for individuals is situated in the wider problem of finance bias in the generation and dissemination of scientific knowledge in medicine. At a systems level, trials of new and existing drugs are commonly financed by pharmaceutical companies. One could imagine drug development and trials funded through the taxation or through taxation of pharmaceutical manufacturers (5) and entirely in the public interest; however, this is far from the current model, which relies on the vast resources of pharmaceutical companies in the risky and expensive development of new treatments, most of which will fail. In the world as it is, industry continues to exert more control than ideal on which trials are done. Academic involvement, at its best, attempts to ensure that the studies performed are unbiased reflections of the truth, but this creates conflict of interest for the academics involved. Finance bias is also present at the level of funding of guidelines, which was beyond the scope of the investigation of Murayama et al. (4). In the same way that the better conduct, regulation, and reporting of trials has improved the way that evidence is produced but rendered its production more expensive (6), the rigorous processes and evidence assessment teams used by some clinical practice guidelines increase the expense of this work and, with it, the potential for finance bias whenever industry contributes to the costs of clinical practice guideline development. Absolute separation between funders and the products of the clinical practice guideline is essential to preserving the trustworthiness of the process. Finally, there is the possibility of finance bias in the individual people serving on clinical practice guidelines. It is challenging for academics to decide between working with industry to improve or disseminate studies or working on guideline committees without overlap between the two. One might hope that academics engaged in the creation of guidelines are able to put these relationships and payments out of their minds in making their judgments. However, the magnitude of the monetary payments identified by Murayama et al. (4) reflects important and likely enduring connections, and the knowledgeable and informed individuals useful to a clinical practice guideline committee are identical to industry’s perception of key opinion leaders (5). A systematic review of clinical practice guideline recommendations identified an association (relative risk, 1.3; 95% confidence interval, 0.9 to 1.7) between financial conflicts of interest and decisions favorable to those interests (7). In the work by Murayama et al. (4), a large number of guidelines were on the basis of low-level evidence, there were many strong recommendations, and the relationship between the strength of evidence and the strength of recommendations was weak; judgment is clearly being exercised by members and chairs, which allows for the possibility of influence by conscious and unconscious bias arising from relationships with industry. Murayama et al. (4) also identify heterogeneity and inconsistency between different guideline committees’ policies, in keeping with a previous analysis (8). This suggests the need for an internationally recognized standard in this area—one that operationalizes the principles specified by the Institute of Medicine and the Guidelines International Network, adopts a high bar in each category, includes all of the categories, includes reporting on family members, and incorporates longer look-back periods and updating (9). A formal ethics framework may be feasible and useful (although it adds further to the cost). Divestment of shares and a commitment to avoid conflicts of interest for a specified and substantial future period are essential. Before the selection of an individual for any role on a guideline panel, the Institute of Medicine recommends disclosure of all current or planned commercial activity; independent assessment of this information for high-risk conflict of interest should include the magnitude of the monetary value of the payment and whether the company specifically has a commercial interest in the output of the clinical practice guideline. In 2019, the American College of Physicians created a framework in which “any active relationship (financial or otherwise) with a high-risk entity, which … defines as an entity that has a direct financial stake in the clinical conclusions of a guideline or guidance statement” is a high-level conflict of interest that precludes participation (10). Alternatively, methodologic experts rather than clinical content experts can hold leadership roles and make final decisions (1). Current assessment methods for members’ conflicts of interest involve forms and processes that differ between guideline committees, covering differing look-back periods and differing domains. A single publicly-visible method of registering all conflicts of interest in the form of an enduring, comprehensive, parallel curriculum vitae would have obvious advantages; the Open Researcher and Contributor ID (ORCID) Database already captures any of these categories and could be used for this purpose. In addition to documenting the disinterestedness of clinical practice guideline members, this would be an asset in publishing; at present, journals must choose whether to require the completion of a detailed 3-year look-back ICMJE form from each author, which adds to the already substantial work of submission, or to rely on author declarations during the submission or in the manuscript itself. Intellectual conflicts of interest have received less attention (11) but are no less important. One overt and measurable indicator is self-citation, identified as prevalent by Murayama et al. (4). Self-citation can be managed through independent review of the relevance of the citation—citations to review articles and opinion pieces, for example, could be disallowed. However, an individual’s desire to influence outcomes and for that individual’s perspective to predominate goes beyond the desire to inflate the citations of a researcher’s papers, and its other manifestations may be completely unmeasurable. Disclosures C.M. Clase has received consultation, advisory board membership, honoraria, or research funding from Amgen, Astellas, Baxter, Boehringer-Ingelheim, Janssen, Leo Pharma, the Ontario Ministry of Health, Pfizer, Sanofi, and the University of Alberta and through Liv Academy, AstraZeneca. She works on the ONTARGET and TRANSCEND databases; these trials were funded by Boerhinger Ingelheim. She is coinvestigator on the Report study, funded by Astellas, and on the FLUID study, funded by Baxter. In 2018, she cochaired a Kidney Disease Improving Global Outcomes potassium controversies conference sponsored at arm’s length by AstraZeneca, Bayer HealthCare, Boehringer Ingelheim, Fresenius Medical Care, Relypsa, and Vifor Fresenius Medical Care, companies with financial interests in the outcome of the conference. In 2019, she cochaired the Canadian commentary on Kidney Disease Improving Global Outcomes metabolic bone disease guidelines, which was unfunded. She cochairs the cloth mask knowledge exchange, a stakeholder group that includes cloth mask manufacturers and fabric distributors. She is a member of the American Association of Textile Chemists and Colorists, the American Society of Nephrology, ASTM International, the Canadian Society of Nephrology (CSN), and the Green Party of Ontario. She is an associate editor of the American College of Physicians journal club and Editor-in-Chief of the Canadian Journal of Kidney Health and Disease, the official journal of the CSN. A. Mathew is the Chair of the Canadian Clinical Practice Guidelines committee. She is a member of the American Society of Nephrology, CSN, and Hypertension Canada.