A paper was accepted and published in journal A which dealt with a cohort of patients with an unusual respiratory pathogen. A similar paper had been published in a US journal B a few months before. It dealt with more or less the same patients (a few more had been added) and provided some extra secondary outcome data but with the same conclusions.
The editor of journal A considered this to be duplication but the authors deny this on the grounds that there are further data.
This is a difficult issue as the editor considers it a case of duplicate publication but the authors disagree. Some journals ask authors to send related papers when submitting their articles for publication. The Forum agreed that journals should have some form of declaration in the instructions to authors or the submission information must be very clear on the rules of duplicate publication. Some argued that a rule of thumb is that if the “extra” data do not stand alone, then it is probably duplicate publication.
In this difficult area, the decision really needs to be left to the judgement of the editor and a correction published in both journals if duplicate publication is believed to have occurred. Software can sometimes help as it can give (in percentages) the amount of overlap between two papers and then editors can judge what amount is acceptable. However, if undetected, all agreed that this is a serious problem as the data may be counted twice in meta-analyses.
The editor indicated to the author that this was a case of duplicate publication and the paper was withdrawn from the website. A notice of duplicate publication was published in the editor’s journal.