Recent high profile cases of research misconduct have relied upon the sharing of relevant information among the Editors-in-Chief of the journals concerned during the months and years leading up to the final settlement of the cases (see http://onlinelibrary.wiley.com/journal/10.1111/%28ISSN%291365-2044/homep...)
Discussions with publishers suggest that such sharing of information risks accusations and/or legal claims of defamation, since submissions should be handled confidentially, even though the journals concerned might carry a statement on their website saying that manuscripts and related documents may be shared if that is considered necessary to investigate possible misconduct.
However, without the (apparently risky) sharing of such information by emails, the above cases would not have been brought to light in such an effective way—and possibly would never have been revealed at all. Sharing the information by telephone is not practical given the potentially large number of journals and Editors-in-Chief involved, especially across different time zones.
The type of information shared might include general enquiries about whether a particular author has submitted certain types of cases to other journals within a particular field, or specifics about manuscripts, including data or even reviews. Such sharing might allow comparisons of submitted data in different versions of the same manuscript for example, or of potentially plagiarised text, or other inconsistencies.
Questions for discussion:
- Are publishers right to be cautious about sharing of such information?
- Would a disclaimer on a journal’s website prevent such claims?
- Would COPE’s endorsement or publication of a Code of Conduct that allows sharing of information under certain circumstances prevent such claims?
- Such a Code might consist of the following:
- Editors-in-Chief and journals have a duty towards authors to treat their work in confidence, except where sharing it is a necessary part of the review/publication process;
- However, Editors-in-Chief should be able to inform other Editors-in-Chief of current enquiries relating to possible data fabrication, lack of ethical approval, serious plagiarism or duplicate publication affecting multiple papers;
- This should not occur at the time of the first enquiry/letter to the author but only if the response is inadequate or there is no response at all within a reasonable time (eg, a month);
- The format of the summary should be purely factual, eg “I have just sent a letter to Dr X asking about the publication of six papers arising from a single study”;
- Circulating this information in no way indicates a judgement of wrongdoing on behalf of the named author(s), but is merely to help the alerting Editor-in-Chief (in case any of the others has information that might be useful—perhaps even to exonerate Dr X) as well as the other Editors-in-Chief, who may be currently appraising manuscripts from the same author or who may also be considering similar letters (since it would be much better to so jointly);
- It is appropriate to use email for such communications, all of which should be marked as, and treated as, confidential;
- Participating journals would inform authors via their webpages that information may be shared under these circumstances.
- Even were legal advice to suggest that such a Code would not prevent a legal claim arising, should COPE endorse or publish such a Code anyway, at least to advise Editors-in-Chief of best practice?
As well as those at the Webinar, we would like to extend the discussion to those unable to join the Webinar, by commenting here, so please do leave your comments.
Comments
Is there any difference between sharing information (ie suspicions) about a submitted (but as yet unpublished) manuscript and a published document? It might be helpful to consider both situations. For redundant publication or possible plagiarism, it might be worth talking to Hal (Skip) Garner about the Deja Vu database, as I understand they had some legal concerns but managed to overcome these. Use of text matching software makes it easier to make an objective statement (eg we found 40% text overlap between X and Y) but suspicions of fabrication or falsification are harder to communicate without being defamatory.
to post comments