Disclosure and disinterest.

Abstract

S tatements—whether about science, or the economy, or human behavior—now often invite questions about the author's credibility; queries aimed not only at claims made during discussion but also at articles printed in journals or newspapers. Perhaps this epidemic of doubt is just another aspect of the erosion of public trust in institutions and occupations that were once highly respected, such as clergy, physicians, Congress, media, chief executive officers, and, yes, us scientists, too. For whatever reasons, society is now concerned about possible sources of bias and seeks assurance through disclosure that the data or opinions presented are those of a disinterested party. That concern arises in widely different institutional contexts. The U.S. government has played its own part in the rising tide of skepticism. First came a late-night amendment attached by Senator Richard Shelby (R-AL) to the 1999 spending bill. That opened primary data from government-sponsored research projects to Freedom of Information Act requests, allowing critics to reanalyze the findings and reach their own conclusions. Then, in June of 2001, the Office of Management and Budget (OMB) issued draft “data quality guidelines” to ensure the “quality, objectivity, utility and integrity of information disseminated by federal agencies.” That move created new avenues for administrative challenges to scientific findings. And now OMB has issued “proposed peer review standards for regulatory science.” Calling their process peer review seems strange to us here, given the OMB criteria. They advocate transparency through disclosing the identities of reviewers; they encourage public participation; for science supporting certain regulatory actions, they would disqualify potential reviewers with previously expressed views on the matter at hand and those with any prior connection to the regulatory agency. The American Association for the Advancement of Science, Science's publisher, has expressed concern over some of these criteria in comments on OMB's proposed standards. At Science , as at most scientific journals, we get the very best scientists in a field to evaluate the significance, quality, and reliability of the work. In contrast to the OMB guidelines, we believe that candor is best served if the referees are anonymous to the author and if the process is kept private. Rather than disqualify reviewers on the basis of previous statements on the topic, we would select individuals whose past history recommends them as objective evaluators. And as long as they weren't from the same institution as the author, we wouldn't worry about their relationship to an agency. Our experience with the “conflict of interest” problem involves authors, not reviewers, and primarily concerns financial interests. Indeed, the questions we have been asked about our own disclosure policies have involved possible ties of authors to commercial interests through consulting arrangements, equity holdings, or patents or licenses. Some of our authors have not sensed that affiliations of that kind might raise perceptions of conflict. So we supply authors with an electronic check-off form that lists various relationships of this kind. If the responses appear to require disclosure, we publish them so that readers may make their own judgments. We don't make the disclosure available to the peer reviewers, because we ask them to evaluate the science, not to serve as ethicists. It is understandable that scientists now are asked such questions. As a university president in the 1980s, I watched as faculty scientists began to patent their work through the university and start companies licensed to develop it. With more and more scientists straddling the line between the commercial and academic worlds, it's no surprise that editors and the public are asking deeper questions about institutional commitments and financial relationships. The problem has surfaced most publicly in medicine, where the corporate ties of investigators involved in clinical trials have received major media attention. Several medical journals then made it a policy to decline participation by all scientists with such connections. That didn't last long, because it ruled too many able people out. They have since opted for disclosure, and we think that's the right choice. One can find a little bit of everything in the institutional responses to our contemporary loss of trust. OMB has produced an overengineered, confusing bulletin that bears the wrong label. Some medical journals tried excluding authors with commercial ties, then wisely dropped that approach. Here at Science , we have tried to give authors more guidance about disclosure, and we'll continue to help our readers make their own informed judgments.