The conference heard from Doug Altman, statistical adviser to the BMJ, how medical journals continue to be full of serious methodological errors, meaning that many studies reach false conclusions. The problem was identified many years ago, and yet there have been few improvements. Medical research is too often done by untrained people for the wrong reasons, including career advancement.
Four years ago Rennie argued the need for journals to adopt open peer review systems, whereby authors and eventually readers know the identity of reviewers. “The ethical arguments against open peer review are disgraceful,” he said, “and yet hardly any journals have opened up their peer review process.” The BMJ and the British Journal of Psychiatry are almost alone among established journals in having done so.
Rennie has also argued for the need to move from a system of authorship to contributorship—where contributors to a study describe exactly what they did. The International Committee of Medical Journal Editors now supports such descriptions, and many journals have adopted the system.
Journals should also be moving, Rennie argues, towards prepublication review by readers and encouraging authors to update their studies. Almost no progress has been made with either issue.
He then lambasted editors for “giving no time, energy, and thought to their craft.” It was “pretty disgraceful” that so few editors had turned up to the only conference that looked at the evidence base for their craft. There may be 15000 journal editors, and yet fewer than 400 had booked for the conference.
