Journal editors get tough on dubious research

The web is often blamed for making it easier to commit plagiarism, yet the reverse seems to be true in the realm of radiology research. Internet access and web-based tools are making it easier to identify possible duplicate papers, according to the editors of three leading journals.

The American Journal of Roentgenology moved to a new web-based online submission and peer review system in February. This program provides reviewers with direct links to related topics, and allows them to search for work produced by the same author. Sometimes this reveals some interesting duplication of effort, said Dr. Robert Stanley, editor-in-chief of AJR.

"We do worry about that, but we are getting on top of duplication a lot more than was possible in the past, when the Internet didn't exist," he said at Saturday's Ask the Editors session.

Reviewers assessing the quality of a paper can now use online tools to check the relevance of citations. This process may inadvertently uncover almost identical publications that are being considered elsewhere, said Dr. Albert L. Baert, editor-in-chief of European Radiology.

"Some authors are so eager to have another citation with their name on the manuscript that they reveal that they have submitted the same or similar manuscript to another journal," he said.

The author's view of what constitutes original research will often vary from that of the journal. Many radiologists believe that taking data they have reported on previously and performing a different kind of analysis does not count as duplication. Baert begs to differ. He will write to an author personally if any doubts are raised about research submitted to European Radiology.

Sometimes discovering dud papers is still pure serendipity. In one memorable AJR case, a reviewer found himself reading a paragraph that seemed oddly familiar. He then realized that the section had been lifted word for word from one of his own papers.

"The editors can do so much and so can the reviewers, but sometimes you just have to be lucky," Stanley said.

Radiology's guidelines for authors include a section explaining what the journal considers to be a duplicate publication, and what possible sanctions may be imposed on the discovery of such a paper. Most identical or overly similar publications are flagged by reviewers following online citation searches, said Dr. Anthony Proto, editor of Radiology. But again, the discovery of dubious authorship may be down to luck.

"On three separate occasions, another journal sent the same paper to the same reviewer at the same time. What a coincidence that is," he said. "We have also been informed of a duplicate publication by a librarian who picked up the same article published twice."

He will always ask authors of suspected duplicates to justify why their paper is not redundant. If he is unconvinced by the explanation, the paper will then be passed to selected associate editors or deputy editors. If they agree unanimously that the paper is unoriginal, and the author still disputes this, then Radiology will issue a notice to university officials.

"We have published probably four or five notices of duplicate publications since I have been editor," Proto said. "So it is not commonplace, at least in terms of our recognition. How prevalent it is, how often it occurs, I really don't know. But I think the online scenario is going to make it more and more difficult for authors to duplicate."