Policy changes: article retraction and removal

In 2002, a paper published in Society journal Human Immunology was withdrawn at the request of the Society. Its immediate removal from ScienceDirect sparked a debate about Elsevier’s removal procedures, which resulted in significant policy changes.

In exceptional circumstances, articles need to be removed or retracted from journals for reasons such as plagiarism, fraud or legal issues. Because material cannot be changed once printed, an explanation of why the article has been removed from the publisher, editor or author is simply printed in the following issue. The relative ease of changing data stored electronically means that articles can theoretically be ‘deleted’ from ScienceDirect as and when necessary.

Big debate
The Human Immunology incident prompted Elsevier to rethink its Article Removal Policy. “The academic community pointed out that if printed papers cannot be removed, why should we be able to remove exactly the same material published online?” says Lindi Belfield, senior product manager ScienceDirect. "The debate also brought to our attention the fact that, by physically removing an article from ScienceDirect, we are changing the historical record - something the scholarly community feel that publishers should not have the right to do."

Clear message
After much discussion, it became clear that the article should still be available on ScienceDirect in PDF format, together with a retraction notice explaining the reason. The PDF would be watermarked with “RETRACTED” on every page to make certain that no one could be mistaken about the paper’s status.

Editors’ decision
“Academics felt strongly that it was not the place of the publisher to judge authors and editors and that it is up to the scholarly community to moderate peers,” continues Belfield. "Elsevier concluded they were right, so it was decided that chief editors, in conjunction with authors where relevant, should be the ones to request retractions and removals. Elsevier would facilitate these requests."

Full guidelines
Policies were then revised to give Elsevier publishing editors comprehensive guidelines about how, why and when articles should be retracted, removed and/or replaced. Details of what they need to request from the chief editor to effect a removal, as well as the procedure within Elsevier to ensure correct implementation are also included.

Removal process
If a chief editor believes an article should be removed, s/he needs to inform their publishing editor who submits a retraction/ removal form. The chief editor is asked to write the reason for the retraction or removal which will subsequently be published. The request is reviewed by The Retraction/Removal Panel to ensure that Elsevier policy is being correctly adhered to. In all cases of removal, the original version is now retained in Elsevier’s official archive at the National Library of the Netherlands, ensuring that history is not tampered with.

Legal reasons
Most problems with articles can be dealt with via the retraction route, however, there are a few rare circumstances when the article is physically removed and replaced by removal notice. The chief editor also supplies or approves the text explaining the reason. Such circumstances include serious legal infringement or where Elsevier – together with all others involved with the paper - decides that leaving the article in could result in harm to life, for example, when an incorrect dosage is published in a medical journal.

Reinstating articles
Since 1995, only around thirty articles have been removed. Belfield and her team are now working hard to reinstate these articles according to the new procedures in order to complete the historical archives.