Journals should set a new standard in transparency

SIR — We applaud your commitment, as
expressed in the Editorial “Peer review
and fraud” (Nature 444, 971–972; 2006),
to raising peer-reviewer awareness about
detecting fraud. For studies involving
humans, independent research ethics
committees (in the United States, institutional
review boards) provide
the first independent critical scrutiny of
research protocols. We recently examined
the instructions to authors of 103 medical
journals and found that none requires authors
to provide to readers (as online
supplementary information accompanying
the publication) the protocols approved by
these committees.
As concern increases about the integrity of
published scientific research, we believe that
biomedical journals should establish a new
standard in human-research transparency.
They should require authors to state at
submission — and, where judged necessary,
in their published articles — that the research
has been approved by the relevant ethical
committees. All journals publishing
research on non-human animals (“Animal
experiments under fire for poor design”
Nature 444, 981; 2006) should do the same for
non-human animal protocols.
Journals should also require authors
to provide the full protocols approved by
these committees for the editors and peer
reviewers, and to allow the journal, if it
wishes, to publish these protocols as online
supplementary information accompanying
publication of the main paper.

Robert P. Dellavalle*†, Kristy Lundahl†,
Scott R. Freeman†, Lisa M. Schilling†
*Department of Veterans Affairs Medical Center,
1055 Clermont St, Denver, Colorado 80220, USA
†University of Colorado at Denver and Health
Sciences Center, Aurora, Colorado, USA