We have recently become aware of two cases of publication of closely related data in papers submitted concurrently to Thorax and to another journal, without disclosure of the existence of the related paper. One of these concerns papers by Girault et al^1 2 and relates to a study of two forms of assisted ventilation in which the data published in Thorax¹ represent part of a study also published in Chest.² In the review process of the Thorax paper, which dealt with characteristics of assist control ventilation in patients with COPD, the associate editor and external reviewers all commented on the fact that data comparing assist control ventilation with pressure support ventilation would enhance the paper, and this comment was forwarded with other feedback to the authors. The authors duly responded with a revised manuscript which did not include pressure support ventilation data, and did not disclose either in the manuscript or in the accompanying covering letter that a comparison of assist control and pressure support ventilation in these patients was, in fact, available and contained in a paper already under consideration (and subsequently published) by Chest. We consider this to represent duplicate and/or redundant publication, with failure by the authors to disclose the existence of related additional data from the same study to us.

The other case relates to papers on the presence and potential source of matrix metalloproteinases in bronchoalveolar lavage samples from patients with emphysema and healthy controls published by Finlay et al in Thorax³ and in the American Journal of Respiratory and Critical Care Medicine.⁴ These two papers, which present results of different analyses relating to closely related hypotheses carried out on biological samples from the same cases, were under consideration by the two journals concurrently without disclosure of the existence of either related publication to either journal editor. It is our opinion that the common origin of the samples used in these studies should have been acknowledged, and that the existence of another closely related manuscript with another journal should have been disclosed explicitly to both journal editors.

As editors we understand that multiple analyses or investigations of existing datasets or biological resources are commonplace, and would regard this to be perfectly acceptable so long as this is made clear in the manuscript. Disclosure is crucial in these circumstances, however, so that editors and readers know that samples or data used in different papers are not independent and can interpret findings accordingly. We ask all authors submitting papers to Thorax to inform us of any related publications and submissions to other journals, at any stage of the review process of papers being considered by Thorax.