A serious case of suspected data fabrication has emerged after the editor of a journal received correspondence from a third party regarding a published paper by four co-authors. The concern was raised due to the observation that all error bars in three separate figures were identical. Further investigation by the third party, involving the statistical appraisal of 32 publications from the same author group, revealed suspicions in 21 papers (66%), in stark contrast to a control sample from other institutions publishing in the same journals.
The Committee on Publication Ethics (COPE) recommended that, despite the involvement of multiple journals in this complex case, the journal in question should independently pursue the matter. To ensure due process, the editor was advised to reach out to the authors, requesting raw data without making direct allegations of data fabrication. The authors were given a one-month deadline to respond. In the absence of a satisfactory response, a second letter would be sent, indicating that the authors' institutions would be contacted.
Additionally, COPE suggested involving an independent statistical adviser to review the paper apart from the third party.
In the follow-up, the editor received a response from one of the authors, who acknowledged re-evaluating the results and identifying an error in one of the figures. The raw data and a covering note were shared with the third party who initially raised concerns, as well as an independent statistician for an unbiased analysis of the data and assessment of the author's comments. The investigation is ongoing to determine the validity of the reported findings.