Everything You Need to Know About Conflicts of Interest
It's not just about the money.
Posted February 9, 2017
As we discussed in Part I of this three-part series on conflicts of interest, it's not exactly a surprise that money can be a powerful influence on scientists' and physicians' behavior. Scientists and physicians who take money from drug companies are prone to design studies in ways that favor the company’s products, to minimize adverse side effects the company’s medications cause, and to prescribe the company’s medications to their patients. Organizations that formulate guidelines for the treatment of various illnesses often have financial relationships with companies that make products recommended in the guidelines.
These facts have led to many policies that attempt to make financial relationships between the pharmaceutical industry and the biomedical world more transparent. Journals require disclosure statements by authors of potential conflicts of interest and laws like the federal Physicians Payment Sunshine Act of 2010 mandate that payments by medical product manufacturers to physicians and teaching hospitals be reported to the Centers for Medicare and Medicaid (CMS) and made public on a website.
Other forms of financial incentives to scientists also frequently make the news. Recently, for example, the media revealed that scientists who concluded that fat rather than sugar is the most harmful part of our diet received money to support their research from the sugar industry. Similar alarms have been raised about the relationships between agricultural scientists who study the effects of pesticides and the safety of GMOs and the companies that make those products.
But is money the at the root of all conflicts of interest? No one can doubt that these financial relationships are troubling and should be examined carefully before we accept the science that emanates from them. But emphasis on financial relationships may mislead us into thinking that these pose the only potential conflicts of interest. In fact, non-financial conflicts exist as well.
Recently, I was asked by a company to which I consult that does not manufacture any products to research the current evidence for the treatment of depression (Note: I have changed the details here to protect the identity of the principals involved, but the story is based on what really happened). Although antidepressant medications are clearly effective and safe, there is also a robust literature on the efficacy of psychosocial treatments and some indication that these may be superior to medications under many circumstances.
I wrote this in my report and was later surprised to see that another consultant who reviewed my work for the same company criticized one of the studies I cited. He called it unreliable because of bias. That study found that a form of cognitive behavioral therapy worked better than medication for a group of adults with depression. The consultant stated that the author of the paper was known to be biased against medication. On the other hand, I know that the consultant has been a strong proponent of medication for depression for many years.
Money does not seem to be involved in this story. The author of the paper showing that CBT worked better than medication did not receive payments from a drug company. The consultant who criticized that paper may have received such money in the past but not for reviewing my work this time. But both the paper’s author and the critical consultant are strongly identified with particular points of view. In fact, their professional identities are in part based on their respective positions of what works best to treat depression. Here, the potential conflicts of interest involve scientific partisanship, not money.
Emotional conflicts of interest are also common. A great example of this comes from a recent review of Paul Bloom's controversial book Against Empathy by Simon Baron-Cohen. In a parenthetical note in the review Baron-Cohen writes, “As someone who has represented the pro-empathy perspective, I have at times been a foil for his arguments — including in this book.” Indeed, Bloom makes several quite disparaging comments about Baron-Cohen’s work.
Unsurprisingly, the review was quite negative. In this case, the reviewer had a clear emotional conflict of interest. Is the simple acknowledgment of his relationship to the author under review sufficient to deal with that conflict? My guess is that if we challenged the reviewer he would insist that despite his being quite specifically singled out for criticism by Bloom he could nevertheless still be objective in writing his review. Do we believe that is really possible? Should the New York Times Book Review editors even asked Baron-Cohen to review the book? (Incidentally, I agree with many of Baron-Cohen’s criticisms, but that does not obviate the conflict of interest issue).
In another personal example of an emotional conflict of interest surround professional identification, I was recently asked to review a paper submitted to a scientific journal by a research group I highly respect. It did not take me long to realize, however, that the authors interpreted the results of their study as directly challenging a hypothesis I published in 2000 in another journal, that panic attacks involve activation of the brain structure the amygdala.
My first thought was “why did the editors send me this paper given how obvious it is that my work is being critiqued?” Nevertheless, I proceeded to read the paper and recognized that the authors had conducted a methodologically sound and quite elegant piece of research. I even had to admit to myself that their reservations about my original hypothesis were reasonable, and so I recommended the paper be accepted with minor revisions.
Once again, no money is involved in this example. What is involved is reputation. I could easily have vented my emotions by writing a savagely negative review. One might say I successfully suppressed this impulse and dealt with my conflict of interest. But the fact is that no one but me will ever know whether I did that or not. No rules require that we disclose scientific partisanship, bruised egos, or long-held personal dogmas. Yet these sources of emotional bias seem every bit as serious as those that involve financial rewards.
We do not minimize the serious nature of monetary payments to scientists and physicians. These are clearly problematic and as we said in Part I of this series, we need better ways to deal with them than merely demanding disclosures. What we are raising is the notion that conflicts of interest that involve emotions, reputations, and status are equally important but entirely absent from our conversation. They need to be considered.
In Part III of this series we will offer some concrete suggestions about how readers of scientific reports can decide whether conflicts of interest are present and if they are, whether they compromise a paper’s value. In the meantime, there are two great series on these issues that specifically discuss non-financial conflicts of interest, available here and here.
 Seife, C: The pharmaceutical industry funnels money to prominent scientists who are doing research that affects its products–and nobody can stop it. Scientific American. December 1, 2012
 Yeh JS, Franklin JM, Avorn J, Landon J, Kesselheim AS: Association of industry payments to physicians with the prescribing of brand-name statins in Massachusetts. JAMA Internal Medicine 2016;176(6):763-768.
 Campsall P, Colizza K, Strauss Stelfox HT: Financial relationships between organizations that produce clinical practice guidelines and the biomedical industry: a cross-sectional study. PLOS Medicine, May 31, 2016 http://dx.doi.org/10.1371/journal.pmed.1002029