Contrary to popular belief, the scientific literature is distorted and biased, says a new editorial in this week's PLoS Medicine. Publication bias (where positive studies are more likely to be published than negative ones), outcomes reporting bias (where certain favorable outcomes are chosen to be included in the scientific paper), and the influence of funders or researchers' conflicts of interests are just some of the biases that can threaten the validity of the scientific record, says the editorial.
The PLoS Medicine editors say that without resistance from all those involved in the research enterprise, "...journals may increasingly become closer to works of fiction telling the stories dictated by various lobbies rather than works of science."
The editors propose five ways in which everyone involved in the publication process can help to reestablish the integrity of the published record. First, editors themselves should recognize and declare their own competing interests. Second, noncommercial interests held by editors, reviewers, and authors (such as personal relationships, rivalries, lobbyist activities) must be acknowledged and declared, an issue discussed in more detail in a recent editorial (http://dx.doi.org/10.1371/journal.pmed.0050199). Third, the issue of ghostwriting of scientific papers must be tackled, particularly when it involves studies on drugs and devices—the editors propose that journals develop detailed authorship policies in order to combat ghostwriting (doi:10.1371/journal.pmed.1000023). Fourth, mechanisms to control outcome reporting bias should be developed, including requiring copies of study protocols and comparing them against submitted papers. Finally, editors should "...base their decisions on the importance of the underlying research question, and not on the headline potential of actual results," in order to prevent reporting bias.
Citation: The PLoS Medicine Editors (2009) An unbiased scientific record should be everyone's agenda. PLoS Med 6(2): e1000038. doi:10.1371/journal.pmed.1000038
IN YOUR COVERAGE PLEASE USE THIS URL TO PROVIDE ACCESS TO THE FREELY AVAILABLE PAPER: http://medicine.plosjournals.org/perlserv/?request=get-document&doi=10.1371/journal.pmed.1000038
PRESS-ONLY PREVIEW OF THE ARTICLE: http://www.plos.org/press/plme-06-02-editorial.pdf
CONTACT:
The PLoS Medicine Editors
Public Library of Science
185 Berry Street, Suite 3100
San Francisco
CA 94107 USA
PLoSMedicine@plos.org
Journal
PLoS Medicine