Journals scolded for slack disclosure rules
Even high-profile papers may violate conflict-of-interest policies.
The fallout from the scandal surrounding Korean stem-cell research has sparked a call for journals to toughen up their financial disclosure policies.
The call follows Science's retraction of two papers based on work from the lab of Woo Suk Hwang, of Seoul National University in South Korea. Science's formal retraction was issued on 12 January, two days after the university's investigative panel found that both papers were fabricated (see ' Verdict: Hwang's human stem cells were all fakes').
Before the papers were published, both Hwang and his former colleague, Gerald Schatten of the University of Pittsburgh in Pennsylvania, applied for patents based on their cloning and stem-cell technologies (see ' Schatten in the spotlight').
Should the patents be granted, the scientists could make a fortune from the techniques. But no disclosure of these potential conflicts of financial interest appeared in either of the Science papers1,2, nor in a 2005 Nature paper about a cloned dog3 that also bears both men's names.
Editors of the journals say they are investigating whether the authors complied with conflict-of-interest rules. But an independent critic says both publications should revise their policies and give them more clout, to prevent such omissions in future.
On 12 January, Merrill Goozner of the Washington-based Center for Science in the Public Interest sent letters to Science and Nature asking them to "tighten" their policies. He called Science's policy "weak and ineffectual" and said the journal should publish all conflicts of interest "that might bear a relationship to the subject matter of the contribution".
Goozner also advised Science and Nature to impose a three-year ban on authors who violate their policies, as the US journal Environmental Health Perspectives does.
Science requires authors submitting papers to the journal to check a box if they have "a planned, pending, or awarded patent on this work by you or your institution". There is then a space for authors to provide details.
Natasha Pinol, a spokeswoman for Science's publisher, the American Association for the Advancement of Science, says this box was not checked for Hwang's 2004 paper. But Hwang or someone on his team did check it for their 2005 paper. No details were provided about the potential conflict of interest, however, and no disclosure appeared in the published paper.
When asked why Science decided not to print any disclosure information, editor Donald Kennedy replied: "Normally we check the relationship between the topic of the patent and the substance of the paper. In this case the box was checked, but no description was given." The journal would not clarify whether it pressed the authors for further details or not.
Nature has a very similar policy, asking authors to declare whether they have "patents or patent applications whose value may be affected by publication"; disclosures are published with the paper. For the 2005 paper about a cloned dog, Hwang ticked a box for "no competing interests" on behalf of all the authors.
Philip Campbell, Nature's editor-in-chief, says that the apparent breach of the journal's policy is being taken up with authors. If it is confirmed, the journal will publish a correction.
Critics say the Hwang debacle has exposed general shortcomings in the journals' conflict-of-interest policies. In July 2004, the Center for Science in the Public Interest released a report on papers in four journals over three months. The report found that authors had failed to disclose conflicts of interest in 8% of the articles; a finding that spurred several journals, including Environmental Health Perspectives, to alter their policies.
"It occurs disturbingly often," Goozner says. "Now you are seeing people not disclosing in some very high-profile cases, and this is possible because the policies of Science and Nature do not have teeth."
Pinol says that Science is looking at its financial conflict-of-interest policy as part of an overall review after the retractions. In an e-mail, Kennedy says, "We are certainly looking at the problem, with an expectation that we can institute some changes."
The Hwang affair has led Nature to revisit its policies on the data needed to support cloning papers (see ' Standards for papers on cloning'), and it is also reviewing procedures that concern image manipulation. Campbell adds that the journal is considering Goozner's points about conflicts of interest.
- Hwang W. S., et al. Science, 303 . 1669 - 1674 (2004).
- Hwang W. S, et al. Science, 308 . 1777 - 1783 (2005).
- Lee B. C, et al. Nature, 436 . 641 (2005).