Managing conflicts of interest in research
There have been a number of recent reports of conflicts of interest in medical research.
There are at least two types of conflicts of interest – financial conflicts, where researchers stand to gain financially from their work; and intellectual conflicts, where researchers stand to gain professionally.
Ensuring that all conflicts of interest are declared and managed is important for public confidence in science and clinical practice guidelines.
What is a conflict of interest? How serious can they be?
A number of high profile cases of research misconduct have recently been reported where researchers have published false information that inappropriately influenced the quality of patient care. Recent examples include Don Poldermans, a high profile cardiovascular researcher in the Netherlands who used fictitious data in his studies as well as Andrew Wakefield, a surgeon and researcher in the United Kingdom whose published studies on the link between autism and vaccines were later found to be fraudulent . As best we know, these cases are relatively rare. Conflicts of interest, however, are not.
A recent study found that 52% of the experts involved in developing clinical practice guidelines for the management of diabetes in the United States and Canada had a financial conflict of interest. Managing conflicts of interest during guideline development is important for at least two reasons. Guidelines can be influential – they are used by doctors to make treatment recommendations to their patients, by regulators and payers to hold organizations accountable for the quality of care being provided, and by lawyers to sue doctors that did not “follow guidelines”. Guideline development is also often open to considerable interpretation because the research evidence upon which they are based is often not definitive. This leaves considerable room for judgment, which can be affected by unmanaged conflicts of interest.
A Canadian ‘Framework for Responsible Conduct in Research’, developed by national research funding agencies and released in December 2011, defined conflicts of interests as arising “when activities or situations place an individual in a real, potential or perceived conflict between the duties or responsibilities related to research, and personal, institutional or other interests. These interests include, but are not limited to, business, commercial or financial interests pertaining to the individual, their family members, friends, or their former, current or prospective professional associates.”
Approaches to managing financial conflicts of interest
Typically, attention has been focused on easily identifiable financial conflicts of interest – most commonly when researchers receive funds from industry, such as pharmaceutical companies that have a financial interest in the results of their research. Industry pays researchers and doctors in a number of ways – for services such as sitting on company advisory boards, for speaking at company-sponsored events, by providing stock options, or as support for their research. Conflicts can arise when the company’s primary interest (selling their product) is different from the best interests of evidence-based patient care.
Many organizations involved in research like universities, funding bodies, clinical guidance bodies, health technology organizations and peer-reviewed journals have policies about declaring conflicts of interest which rely on experts to disclose financial and personal relationships that may influence their objectivity. A recent study found that over half of the members of guideline committees declared a conflict of interest, while a clear conflict of interest was not declared by 11%.
Jovita Sundaramoorthy, Vice President, Research and Education at the Canadian Diabetes Association (CDA) says that “while the CDA is very sensitive to the fact that authors disclose conflicts of interest, we can only police to a certain extent” and suggests that “when people fail to disclose they may not be hiding it, since conflicts change over time –just the same way that a stock portfolio changes over time”.
Once a conflict is declared, it needs to be managed. There is a lack of consistency about how this is done. The Institute of Medicine in the United States proposed recommendations in 2009 that go beyond current practices. They suggest that organizations need to develop more detailed conflict of interest policies to help them assess the severity of conflicts, and also that the government mandate industry to report how much it is paying researchers and doctors. Some pharmaceutical companies are now posting this information on their web sites, but this is not a universal practice. The actual amount of financial support received by those developing guidelines is rarely publicly reported, even though the amount that industry pays to doctors and researchers varies, and it is possible that most would consider honoraria of $2000 to be less likely to cause bias than honoraria of $50,000.
In Canada, some argue that disclosure does not appropriately minimize conflicts of interest. Gordon Guyatt, a Professor in the Faculty of Medicine at McMaster University says that “people have accepted that they must declare that they took money – and then it is ignored – making disclosing financial conflicts of interest extremely ineffective”.
It’s not just money – managing intellectual conflicts of interest
Guyatt also suggests that “intellectual conflicts of interest are completely ubiquitous” and have generally been ignored..
Intellectual conflicts occur when clinicians or researchers may be too deeply embedded in their own area of expertise to objectively look at a research question “with an open mind”. Guyatt argues that “even when money is not involved …. we [scientists] get very attached to our ideas”. In universities, researchers are rewarded if their work is highly referenced by others and is perceived to be influential. This environment may create an incentive for those participating in guideline development to highlight their research in clinical practice guidelines.
The American College of Chest Physicians recently adopted a new method of guideline development in an attempt to better mitigate financial and intellectual conflicts of interest in developing their guidelines about the management of thrombotic (clotting) disorders. The researchers who led this process were motivated by the argument that “clinical experts who write guidelines are influenced by (usually) declared financial conflicts and by equally important undeclared intellectual conflicts of interest”. As such, experts with significant declared conflicts of interest were excluded from the guideline writing process, although they provided expert input into the guidelines. The guidelines were written by experts in clinical research and methods, the vast majority of whom were doctors, and had no known conflicts of interest in this area.
The American College of Chest Physicians is the first group to attempt this approach. Some argue against it, saying that there needs to be greater involvement of experts even if they have conflicts of interest because only they can understand the nuances of the research findings and the diseases being treated.
Time will tell if the ACCP approach is adopted on a wider scale, and whether the guidelines developed by this group are perceived as both free of conflicts of interest and clinically credible.