Keep data for 10 years, say research integrity guidelines

Funders should stipulate that academics keep their data for at least 10 years so that research misconduct investigations have access to original findings, according to a proposed new set of guidelines on cooperation between universities and journals.

It is hoped that the code, which has been drawn up by scholars and journal editors from around the world, will address the problems that can occur when the need arises to probe whether the findings of research are genuine or not; and, in particular, lead to more transparency about who periodicals should contact with their concerns, and when.

Journals often do not know who to contact at an institution about research integrity, for example, or the original data held in laboratory notebooks or equipment logs may not be readily available. In some countries, such as the US, the researcher in question may already be under investigation but institutions are bound by strict confidentiality measures and so cannot inform journals until the proceedings have concluded.

The Cooperation and Liaison between Universities and Editors (Clue) guidelines recommend the creation of national registries of research integrity officers at universities and the better sharing of investigation reports on trustworthiness between institutions and all potentially affected journals.

The protocols, which have been published on preprint server bioRxiv, call for journals to develop specific criteria to determine whether an institution should be contacted instead of a researcher where there are strong suspicions or clear evidence of wrongdoing, in case they destroy vital evidence.

They also call for more care to be taken with the identity of whistleblowers, and for the creation of a mechanism to assess whether findings are genuine that is independent of a university’s investigation into whether misconduct has occurred.

Sabine Kleinert, one of the authors and a senior executive editor of medical journal The Lancet, told Times Higher Education that there are  “often difficulties on both sides” when journals have to work with institutions to decide whether research is valid and trustworthy and what the appropriate action should be.

“It would be really good if they lead to more transparency about who in an institution needs to be contacted and at what stage,” she said.

The Clue guidelines build on the 2012 Committee on Publication Ethics guidelines about what journal editors should do if questions are raised about the integrity of research. Soon after the Cope guidelines were published, a discussion at the World Conference on Research Integrity in Montreal flagged up that further guidance on the practical detail of how to go about investigating research integrity was needed.

At a follow-up meeting last year in Heidelberg, journal editors, publishers, research integrity experts, universities and a lawyer looked at the issues in more detail. The event led to the creation of the draft Clue guidelines, which are now available for comment and are set to be revised further based on this feedback.

holly.else@timeshighereducation.com