SURVEY FINDS FEW RESEARCHERS DISCLOSE CONFLICTS

by Constance Holden

Despite heightened sensitivity to the subject, a new report finds that few journals publish information about their authors' ties to commerce. Explicit guidelines are rare, the survey found, and many authors may feel the rules don't apply to their situation.

The survey, reported in the April issue of Science and Engineering Ethics, found that a mere 327 (0.5 percent) of the 61,134 papers appearing in 181 peer-reviewed journals in 1997 contained statements about authors' financial ties. Two-thirds of the journals published no disclosures; only three did so in at least 10 percent of their articles. Those journals were the only ones, out of 1,396 "high-impact" journals surveyed-most of them covering biomedical research-that had any rules regarding disclosures of potential conflicts of interest. The policies ranged from detailed questionnaires to request for authors to declare any ties that might be construed as a conflict of interest.

"I would say 0.5 percent is incredibly small when you look at all the information about the rise of patenting and commercial ties. I would expect at least 20 percent," says co-author Sheldon Krimsky, a professor of urban and environmental policy at Tufts University, in Medford, MA. He says an earlier survey of Massachusetts biomedical scientists found that one-third of those who published in 1992 had financial interests related to their research-from patents to advisory positions in biotech companies.

Krimsky thinks a lot of scientists "are looking at these policies and saying, 'Sure, I have interests, but they're not conflicts of interest.'" Marcia Angell, former editor of the New England Journal of Medicine [which admitted to a failure to divulge potential conflicts among authors of several papers in recent years (Science, 3 March 2000, p. 1573)], says the survey demonstrates that journals need to tighten up their policies. Many have a qualifying clause, as in financial ties, "that may bias your work," that are "big enough to drive a truck through," she says.

At the same time, it's not clear how well the 1997 data reflect the current situation. The study is "probably already dated because this is such a fast-moving area," says John Parrish, head of dermatology at Massachusetts General Hospital, in Boston, who believes financial disclosure is "getting to be the cultural norm." A group of medical school deans, led by Harvard's Joseph Martin, have drafted new conflict-of-interest guidelines for biomedical researchers, and the Association of American Medical Colleges has just established a committee to look into clinical research.

Krimsky and others don't think things have changed that much. While clinical trial mishaps have spurred universities to reexamine conflict-of-interest policies, he says, "journals have not had the same impetus for change." Angell agrees, although she believes that the issue of journals extends beyond self-reporting. "Research institutions [as well] need to have far more stringent regulations," she says.

#

Constance Holden is a news writer with Science. News of the Week, Science, 4 May 2001.