Journals, authors cited for conflicts of interest
By Robert Davis, USA TODAY
Some leading scientific and medical journals do not always enforce their conflict of interest policies with the authors of published studies, according to a new report.
The Center for Science in the Public Interest found that in some cases the journals did not disclose contributing authors' financial conflicts of interest even though the journals' own rules require such disclosures.
"There is a consistent pattern here," says Merrill Goozner, CSPI project director. "This is an unacceptable level, and the journals need to take action."
The findings come amid growing concern over the influence that private industry has on scientific research. For example, journal editors, including those responsible for some of the content that was studied by CSPI, are considering requiring drug companies to register all clinical trials in a database for more accountability.
The study examined 163 articles in The New England Journal of Medicine, TheJournal of the American Medical Association (JAMA), Environmental Health Perspectives, and Toxicology and Applied Pharmacology and found that the authors of 13 articles had relevant conflicts of interest that were not reported to readers.
The most obvious conflicts were reported, such as when the research was funded by a company that employs the author. But the CSPI researchers found hidden conflicts in what Goozner calls "the margins," in which there was no direct link, but the researcher stood to benefit from the same industry.
One author of a study on heart disease, for instance, failed to reveal relationships with 20 companies that made cardiovascular drugs or devices.
"This is important for the general public and the scientific community because full disclosure gives you another piece of information for evaluating these studies," Goozner says. "If you hide the fact that there is a conflict of interest with the researcher, then you are deceiving people."
CSPI is a consumer advocacy organization. It says that journal editors should:
• Require all authors to disclose all financial arrangements made with private firms within the past three years, whether or not those arrangements are directly related to the subject of the article. Patents, patent applications and an intention to apply for a patent also should be revealed.
• Impose sanctions against authors who fail to disclose conflicts of interests, perhaps banning the author from publishing in the journal for three years.
The study found the highest incidence of unreported conflicts of interests at JAMA, where six of 57 articles, or 11%, failed to disclose a financial conflict of interest. In most cases, the authors of the JAMA studies in question told CSPI that they had not disclosed the information to the journal editors.
Phil Fontanarosa, JAMA's executive deputy editor, said Monday that he had not yet seen the study and could not comment on it. But he said the study "will help keep the issue on the front burner."
"We all want the same thing, which is transparency of reporting of conflicts of interests," he said.
"These journals were picked because they have the best policies," Goozner says. "Imagine what is happening at the lesser journals."