It is no secret that many physicians maintain financial relationships with pharmaceutical companies and device manufacturers. These relationships are often necessary for drug and device development, testing, outcomes research and innovation and typically take the form of payments made from industry to physicians. These payments may lead to financial conflicts of interest (FCOI) when physicians serve in other capacities on the same subjects for which they are receiving payments.
The impact of FCOI on various aspects of the healthcare system, including patient outcomes, remains unclear. We present an objective overview of the current state of the evidence linking FCOI to clinical decisions. Specifically, we focus on three areas: (1) FCOI of physicians responsible for drug regulation and approval; (2) FCOI of physicians who author documents intended to guide clinical practice; and (3) FCOI of frontline physicians and their impact on prescribing patterns.
FCOI in drug approval committees
Health Canada and the U.S. Food and Drug Administration (FDA) routinely convene expert advisory committees to inform regulatory decisions on certain drugs and devices. Committee members must disclose all affiliations, financial interests and/or intellectual interests. A study of Health Canada committees found that FCOI were infrequent among panel members; however, the majority declared indirect or intellectual interests. Similarly, an analysis of FDA oncologic drug advisory committees found a significant reduction in voting members with FCOI between 2000 and 2014. Together, these studies suggest that FCOI among gatekeepers is relatively minor.
FCOI among authors of clinical practice guidelines
Physicians of influence are often called upon to develop clinical practice guidelines (CPGs) that are intended to be objective, evidence-based documents that outline the gold standard for clinical practice. Yet an abundance of studies examining U.S. guidelines related to gastroenterology, dermatology, oncology, urology and high-grossing medications have found that the majority of authors declare financial conflicts of interest. In one study of Canadian guidelines, 75 per cent of guidelines had at least one author disclosing payments, 29 per cent had a majority of authors with FCOI and 21.4 per cent of guidelines had all authors with FCOI. This suggests that FCOI is common among physician authors of CPGs and have the potential to affect the integrity of guideline recommendations, though no studies have linked FCOI to voting patterns.
FCOI among physicians in clinical practice
In 2018, 627,000 physicians in the U.S. received more than $2.18B in general payments (e.g., food and beverage, travel, consulting fees, honoraria) from industry and the average physician received a median of $160.68. One study focused on pharmaceutical sponsored meals and found that meals valued as little as $20 were associated with a change in prescribing patterns in favour of the sponsoring company. Moreover, another study found a linear relationship between payments and prescribing, where every additional $1,000 in payments resulted in a 0.1 per cent increase in prescribing brand-name statins. This was mirrored in a study of inflammatory bowel disease where payments were associated with higher Medicare spending on two drugs of interest. These findings are of particular concern given that more than 50 per cent of U.S. physicians receive some kind of payments each year; reporting by Canadian news agencies suggest a similar situation in Canada.
Strategies to manage FCOI
In response to growing concerns surrounding FCOI, the U.S. introduced the Physician Payments Sunshine Act in 2010. This act mandated public reporting of payments from pharmaceutical and medical device companies to academic hospitals and physicians to the Centers for Medicare and Medicaid Services. It represented a bold move toward increased transparency in physician payments; since then, many other countries, including the United Kingdom and Australia, have implemented similar policies.
Despite these advances, Canada is lagging behind. In 2017, a voluntary disclosure policy led to 10 of Canada’s largest pharmaceutical companies releasing non-specific information about $75 million in payments to physicians. While this is a step in the right direction, publication of the full details of these payments will require government legislation. The Ontario government was poised to be the first in Canada to do so when it passed the Health Sector Payment Transparency Act in December 2017. Unfortunately, the regulations were not finalized before the 2018 provincial elections and have since been put on hold.
While the government has been relatively inactive on this topic, others have not. Drs. Ahmed Bayoumi and Andrew Boozary launched Open Pharma to build awareness and prompt federal action on disclosure of payments. It is their belief that efforts by individual organizations are commendable but detailed and accessible data will only be possible through governmental legislation.
Transparency is a good first step but is not enough to mitigate the potential influence of FCOI on clinical decision-making. Disclosure of payments is necessary but merely shifts “closed” bias to “open” bias without necessarily obviating the impacts.
To maintain integrity in our processes, a number of additional steps should be taken. Health Canada, the FDA, and guideline development specialty societies should consider creating panels where members are only chosen if they do not have any FCOI relevant to the subject matter being discussed. Specialty societies should also check their panels for FCOI themselves since disclosures may be incomplete and exact details may be difficult to recall. Finally, it is important to note that some conflicts – such as payments directly from a relevant society or intellectual investment in a certain outcome – are not manageable via disclosure and should be eliminated.
Ultimately, there is no current research that links FCOI to patient outcomes; however, it is clear that FCOI are influencing clinical decision-making at multiple levels. To maintain integrity in our institutions and commitment to delivering the best patient care, it is vital that the issue of conflicts of interest be properly addressed.
The comments section is closed.
The issue is not as black and white as the authors of this article believe. The physicians with declared conflicts are also the ones who are the most knowledgeable about the conditions the drug is being reviewed for. I would think that the health outcomes of their patients would come before any financial incentives from the pharmaceutical industry. I guess I have more faith in my doctors than the writers of this article!
Thanks for your comment.
Whether payments to physicians influence clinical decision making is not black and white. Physicians have an essential role in driving innovation in pharma and devices, and part-and-parcel of these relationships will be a financial payment to physicians.
The National Academy of Medicine provides a very reasonable framework to balance FCOIs among clinical practice guidelines, where expertise is required, but where undue influence may potentially affect content. Notable in these guidelines are that (1) the chair be free of conflict; and that (2) <50% of the membership be conflicted. This we believe is the current benchmark standard that should be used by specialty societies to mitigate potential risk, applied a priori when designing CPG committees.
Whether clinical decisions of physicians are affected by payments is also uncertain. However, as we mention, there are many examples of associations where payments to physicians associate with certain drug choices over others when both are indicated. As an example, in the study by DeJong and colleagues that we reference in the article, even small payments such as those of meals are associated with a change in practice to use of branded over non-branded medications, indicated for the same purposes. While causation is hard to ascertain, it is hard to imagine <$20 payments being reflective of payments made because of expertise.
Finally, you are correct that it is quite possible that patient outcomes are improved by decisions made by conflicted physicians – but this has yet to be investigated. Transparency in identifying these payments in Canada will allow this to be evaluated.
Thanks for the article and for advancing the discussion on this topic. In some cases I think that you downplay the significance of conflict of interest. In the study that I did on Health Canada committees and panels, six of the 11 had a majority of members who declared a direct or indirect financial interest. Whether these financial interests led to biased voting is not known since the summaries of the meetings that Health Canada publishes don’t give individual voting patterns. It’s true that we don’t know how COI affects voting patterns on clinical guideline committees but the guideline from the American Psychiatric Association on the outpatient treatment of major depression shows a very strong association between COI and flaws in the guideline. All the panel members had conflicts with an average of 22 conflicts per person. In the guideline only 44% of the cited studies supporting the recommendations met criteria for high quality, 34% of cited studies did not study outpatients with major depressive disorder, 17% of cited studies did not measure clinically relevant results
and 20% of references were not congruent with recommendations.
Excellent start on a complex issue.