News

Spoof research paper is accepted by 157 journals

BMJ 2013; 347 doi: http://dx.doi.org/10.1136/bmj.f5975 (Published 04 October 2013) Cite this as: BMJ 2013;347:f5975
  1. Nigel Hawkes
  1. 1London

A spoof medical paper full of easily detectable flaws was submitted in a sting operation to 304 open access journals and accepted by more than half of them.

The results, reported in Science1 by journalist John Bohannon, reveal “an emerging Wild West in academic publishing,” he says, with mushrooming numbers of new journals profiting from the open access model, in which authors rather than readers pay the cost of publication.

Bohannon invented the names of authors and affiliations—such as Ocorrafoo Cobange of Wassee Institute of Medicine—and submitted in their names variations of the same basic paper claiming to have found anticancer properties in a molecule extracted from a lichen.

The paper contained obvious flaws, such as claiming to have demonstrated a dose-response relationship when the data showed nothing of the sort. To simulate the poor English of some developing world authors, the text was translated into French by Google Translate, then back into English.

The paper written by “Cobange” was accepted by the Journal of Natural Pharmaceuticals, with only superficial changes sought, while other versions were accepted by journals hosted by well known publishers such as Sage and Elsevier. “Acceptance was the norm, not the exception,” writes Bohannon. It was even accepted by journals in completely alien fields, such as the Journal of Experimental and Clinical Assisted Reproduction.

The Sage journal that accepted the paper was the Journal of international Medical Research, which sent a letter of acceptance that asked for no changes but included an invoice for $3100 (£1914; €2293).

The journal’s editor in chief, Malcolm Lader, emeritus professor at the Institute of Psychiatry in London, told Bohannon, “I take full responsibility for the fact this spoof paper slipped through the editing process.” Acceptance would not have guaranteed publication, he added, because it would have been subject to technical editing that is detailed and expensive. The $3100 fee was to cover the cost of this process.

Many of the journals conceal their geographic location, despite having names such as the American Journal of Medical and Dental Sciences (actually published in Pakistan). About a third of those targeted are in India, of which 64 accepted the paper and only 15 rejected it. Drug Invention Today, which appears on the Elsevier platform, is edited by an Indian professor. Elsevier told Bohannon that it reviewed journals before they were hosted on the Elsevier platform and as a result of the sting would be conducting another review.

As soon as papers were accepted, they were withdrawn by Bohannon, so none was actually published. Among the journals that rejected the paper were PLoS One and two journals published by Hindawi, an open access publisher in Cairo that employs 1000 staff and publishes 559 journals. Paul Peters, of Hindawi, told Bohannon, “It is a relief to know our system is working.”

Of the 304 papers submitted, 157 had been accepted and 98 rejected by the time Science went to press. Of the remaining 49 journals to which articles were sent, 29 appear to be defunct and 20 had yet to reply. The majority of decisions were taken without peer review, with only 36 of the 304 submissions generating comments that recognised the flaws. And 16 of those were accepted despite the referees’ comments.

Notes

Cite this as: BMJ 2013;347:f5975

References