Cochrane Reviews as privileged sources to report misconduct behaviours: an informative case of duplicate publication

Article type
Year
Authors
Saiz LC1, Erviti J1, Garjón J1, Elizondo J2, Azparren A1, Gaminde I2, Áriz MJ2
1Drug Prescribing Service, Navarre Regional Health Service, Spain
2Navarre Regional Health Service, Spain
Abstract
Background: Unethical behaviours causing different types of bias have been extensively reported in biomedical literature. At the same time, the quality of systematic reviews can be affected by misconduct in primary publications.

Objective: To describe several cases of serious duplicate publication detected as a result of a Cochrane Review, raising potential synergies between publishers and Cochrane.

Methods: The results of a search in a Cochrane Review on hypertension were independently assessed by authors in pairs and a serious case of duplicate publication was identified. On that basis, we searched for all randomized controlled trials (RCT) published by the main author involved in PubMed (accessed on 20 September 2015) and double checked them to exclude additional cases of plagiarism.

Results: As a result of peer-review tasks in a Cochrane Review, we considered eight articles about 2 RCT authored by the same first author as potential duplicate publications. After contacting the publishers, six articles were retracted.Then we retrieved 129 RCT published by this researcher from the PubMed database, 86 as first author. A total of 47 RCT articles (54.7%), including the eight RCT previously identified, were considered as potential duplicate publications (Table 1). The 47 RCT included antidiabetics, antihypertensives and anti-obesity drugs, were published from 2004-2014, and came from 17 different studies. We informed all the 26 journals affected (linked to 10 publishers) in October 2015. As of 20 April 2016, only one additional retraction was confirmed.(http://retractionwatch.com/2016/02/12/investigation-leads-to-5th-retraction-for-drug-researcher). Three journals decided not to retract on the basis of first copyright and the other three journals considered the potential duplicate content as acceptable. The rest of investigations were still ongoing.

Conclusions: A comprehensive information retrieval and a peer-review methodology are strong points for Cochrane Reviews. Cochrane authors must play an active role in order to correct scientific fraud. Journals are expected to implement convincing measures against unethical attitudes in a timely manner.