From 1 October 2001, all of the Nature research journals, including Nature Medicine, will introduce a new policy asking authors to declare any competing financial interests that relate to research papers. Authors of manuscripts received on or after that date will be encouraged, should their work ultimately be accepted for publication, to disclose any competing financial interests that might impact on the research described. A summary of the policy can be found in a recent editorial in Nature (vol. 412, p751) and more comprehensive details will soon be available on our web site.

The introduction of this policy is not a response to any specific charges that researchers' commercial interests might compromise their integrity, or in any way an insinuation that commercial research is inherently corrupting. Rather it is in response to a growing recognition that potential problems exist.

Competing financial interests can come in many forms, and include research funding, paid consulting, stockholding, advisory board membership, patent holding, and recent, current or anticipated employment. The rewards of research that translates to the marketplace can be enormous, and it would be naive to think that such rewards do not on occasion bias researchers in their selection and interpretation of data, or in emphasis of presentation, whether inadvertent or not.

Outright fraud is undoubtedly rare, but more subtle effects have been reported in meta-studies conducted largely on research published in clinical journals. One such study found that pharmaceutical company-sponsored studies were eight times less likely than nonprofit-sponsored studies to report unfavorable qualitative conclusions1.

Conflict-of-interest policies are already employed by a minority of scientific journals, mostly among those covering clinical research, but compliance with these policies has been very low; one study found that only 0.5% of articles contained disclosure statements2. Although clinical trials are not a primary focus for Nature Medicine, research in this area has proven particularly fraught with conflicts of interest. The issues are more complex and the stakes, especially financial ones, much higher. Recently, a group of journals, including the New England Journal of Medicine and the Journal of the American Medical Association, announced that they will no longer review or publish studies conducted under conditions that allow the sponsor to have sole control of the data3. They intend that this policy will lead to the negotiation of less restrictive research contracts and win back scholarly independence and academic freedom. Their intentions are laudable.

It is important to state that the Nature journals' new policy is not a conflict-of- interest policy per se, but a policy intended to foster transparency through disclosure of scientists' competing financial interests that may be affected by the publication of a body of work. Whether such competing interests constitute a conflict is a far more subjective judgment, one that cannot usually be made by journal editors or peer reviewers.

Statements of competing interests will not be required during the review process, and will therefore not be available either to editors at the internal review stage or to independent external reviewers. Peer review is an honor system, and to undermine that trust would likely be more harmful to the scientific process than the infrequent cases of bias we are seeking to guard against. Also, the role of both peer reviewers and editors is to evaluate the quality of the science at hand, not to act as arbiters of personal financial interests.

Instead, we will request that the competing financial interests be declared only when a paper has been accepted for publication. A statement will be included in the published article. Of course, some researchers will be bound by confidentiality agreements and others will simply disagree with the principle of financial disclosure. These authors can choose not to respond to the disclosure requests and this will in no way prejudice the publication of their research. However, a statement noting the non-response will be published.

Disclosure of the actual sums of money involved will not be requested. Since it is difficult to define at what level a financial interest becomes significant, the new policy will use the rule of thumb: “any undeclared competing financial interests that could embarrass you were they to become publicly known after your work was published.” A more specific threshold would be that used by many universities in the United States, which require faculty members to disclose interests exceeding $10,000 or 5% equity in a company.

We believe that such a self-policing approach is the most appropriate and embraces the spirit in which science should be—and overwhelmingly is—conducted. Our aim is to engender transparency that will allow our readers to evaluate for themselves the significance of any possible conflict. Candidness on the part of authors will further strengthen the credibility of their work. We also hope the policy will address more nebulous public concerns that scientists and their research are for sale, and safeguard the public's trust in science.

We welcome comments, suggestions or correspondence regarding this new policy. Informal communications can be sent to medicine@natureny.com marked “Competing interests policy”; more formal correspondence may be submitted as a Letter to the Editor.