Conflicts and Interests

This year is shaping up as unusually rancorous in the scientific establishment’s eternal grappling with conflict of interest.

In some of the most august journals of research, the topic has aroused accusations, denials, sharp exchanges, and mea culpas, with the din overflowing into the popular press. It’s contended that scientific conflict of interest — let’s call it CoI — endangers the public and undermines support of research. In a new twist, CoI has also drawn praise as a healthy manifestation of the profit motive and a valuable force for speeding the benefits of science to society.

What’s clear is that CoI has a bad name, since it conjures up trifling with trust — like a restaurant reviewer who dines gratis but purports to be independent and objective. The four stars might be merited, but suspicion comes naturally. CoI in the sciences frequently involves academics who publish studies relevant to the products of companies that employ them as consultants. Their findings may reflect unalloyed honesty, but, given the influence of money, who knows?

Disclosure is the commonly prescribed antidote for conflicted situations. Publicly announce the conflict and, poof, outsiders are put on alert and the conflicted, under scrutiny, will behave themselves. But it’s not so simple, starting with the question of when a relationship constitutes a conflict of interest. The tango of academics and pharmaceutical firms is defended as a natural and beneficial pairing: Industry seeks outstanding scientists and they help speed the development of useful drugs, earning income along the way. Major journals require and publish disclosure of such dealings as a condition for publication. But some researchers, perhaps many, do not comply, either from indifference or because they believe that they’re not conflicted.

Which brings us to the recent embarrassments of the nation’s most widely circulated medical journal, the Journal of the American Medical Association (JAMA), which this year has publicly acknowledged three separate research papers in which authors failed to disclose financial ties to pharmaceutical manufacturers, as required by JAMA since 1985. In one of these papers, published February 1, nine of the 13 co-authors failed to report their financial dealings with pharmaceutical firms. In more trustful times, the omissions would have stayed in the family, but not today. The spotlighted research concluded that cessation of drug treatment for depression during pregnancy could lead to a relapse — a controversial finding of obvious interest to manufacturers of anti-depressants.

A  front-page article in the July 11, 2006, Wall Street Journal, headlined “Financial Ties to Industry Cloud Major Depression Study,” poured it on, reporting that “The lead author — Lee S. Cohen, a Harvard Medical School professor and director of the perinatal and reproductive psychiatry research program at Massachusetts General Hospital — is a longtime consultant to three antidepressant makers, a paid speaker for seven of them, and has his research funded by four drug makers. None of his financial ties were reported in the study. In total, the authors failed to disclose more that 60 financial relationships with drug companies.” The authors explained that the research for their JAMA paper was financed by a federal agency, and, in their view, no conflict existed. Even so, JAMA responded, their dealings should have been disclosed when the paper was delivered. In a letter co-signed by several of his non-disclosing colleagues, lead author Cohen agreed, stating that “disclosures would have provided utmost transparency with respect to potential conflict of interest and we wholeheartedly support such a practice.”

The New York Times joined in with an editorial referring to JAMA as a “conflicted medical journal.”  And writing in the Boston Globe, Jerome Kassirer, a former editor of the New England Journal of Medicine, JAMA’s leading rival, warned that “journal editors fail to ask the questions necessary to identify the conflicts…. and still recruit financially conflicted physicians to write articles and editorials that involve selective interpretation of data.”  Kassirer added, “The public trust is at stake, and despite widespread publicity, editors fail to weed out writers with conflicts.” Kassirer also argued that disclosure is over-rated as an antidote to conflict. “The cure for public distrust is to employ people who have opted not to be compromised by money.”

On the opposite side of the fray is Thomas Stossel of Harvard Medical School, who frequently argues in public forums that close ties between academics and pharmaceutical companies are beneficial to health and that CoI concerns have evolved into a witch hunt. In a July 2, 2006, Washington Post article titled “What’s Wrong with Money in Science?” Stossel and a Harvard colleague, David Shaywitz, contended that many cures have evolved from “links that are now widely portrayed as dangerous, corrupting the pursuit of scientific truth and threatening the public.” And they took a swipe at public interest groups on the trail of CoI and “their boosters at JAMA.”

JAMA’s editor, Catherine D. DeAngelis, did not suffer in silence. “Mr. Stossel and Mr. Shaywitz’s position that there is nothing wrong with ‘money in science’ is untenable,” she declared in a letter to the Post. In a letter to the Times, she disputed the label “conflicted medical journal,” asserting that JAMA’s disclosure policies “are more stringent than those of any other leading medical journal in the world.”

However, in a long editorial in the August 7 issue of JAMA, she confessed that even a strengthened and clarified version of the journal’s rules, adopted in July, could not “guarantee that all financial relationships and arrangements of authors are disclosed. It is simply not feasible to independently investigate the financial relationships of every author….”

She also dismissed the idea of adopting severe penalties for non-disclosure, such as banishment, temporary or permanent, from the coveted pages of JAMA. Though a few journals have taken that step, DeAngelis rejected it, with an explanation from the competitive underside of big-league scientific publishing. Sanctions against transgressing authors would simply divert them to another journal, she said, a process that “cleans our house by messing others.” Collective action by journals, she said, risks anti-trust prosecution. The proper way to go, DeAngelis argued, is to refer offenders to their deans — a step that she reported taking on two occasions.

In this busy season of agitation over conflict of interest, a contender for the prize case showed up in the July issue of the journal Neuropsychopharmacology. Co-authored by eight academics, the paper favorably reported trials of an implantable electrical device for treating depression. Absent from the paper was any disclosure that the authors served as consultants to the company that makes the device, including the paper’s lead author, Charles B. Nemeroff, chairman of the Department of Psychiatry and Behavioral Sciences at the Emory University School of Medicine. In addition to that job, Nemeroff was also editor of Neuropsychopharmacology, a position he resigned following disclosure of his non-disclosure.

Will consensus or, perhaps, boredom end the conflicts over conflict of interest? Not likely. When I recently Googled “conflict of interest” in tandem with “research,” the cyber dredge produced 23 million entries. Deduct for duplication, scant relevancy, and outright error, and there still remain multi-mega quantities of news, discussion, conference proceedings, reports, speeches, rules, and other productions linking this pair. Lots of it was fresh stuff, which suggests that the topic is far from wrung dry. Not surprising, when academic science is hard up for money, and industry can buy its skills, good name, and university affiliations. The rancor over conflict of interest easily arises from these relationships.