Re: Plagiarism and data falsification are the most common reasons for retracted publications in obstetrics and gynaecology

Integrity of randomized controlled trials: challenges and solutions

“…The first task involves a review of the statistical methods described and used in any given clinical trial. Concerns should be raised when such a review reveals [1] the description of the statistical methods is absent or too brief for the methods to be understood and therefore reproduced; [2] the choice of statistical method is clearly inappropriate given the research question, study design, and resulting data structure; [3] the summary statistics (e.g., confusing standard deviation and standard error), between-group comparisons, regression model output, and any associated test statistics and P values are incorrectly interpreted; and [4] the procedures specified in the methods section and what is actually presented in the results section and the corresponding tables or figures are inconsistent. The last point includes failing to adhere to a registered statistical analysis protocol or the use of statistical methods that were not specified in advance of data collection.…”

“…However, a systematic review by Chambers et al (3) discovered 18 retracted RCTs by June 2018, of which only eight were labeled as plagiarism or data fabrication/falsification. When we continued their research, however, we found the retraction notices and information on the publisher's website frequently used euphemistic language that hid the real reasons-usually data fabrication or falsification-for the retractions (4). Intentionally or involuntarily, the status quo suggests that research misconduct is being treated leniently.…”

Violation of research integrity principles occurs more often than we think

“…From indexation until June 2018, a total of 176 articles were retracted, of which 40 (22.7%) and 37 (21.0%) were retracted because of plagiarism and data manipulation, respectively. These numbers, however, are likely to be underestimated because retraction notices and information on the publisher's website frequently used euphemistic language and concealed the real reasons, usually data fabrication, for retractions (Li and Mol, 2019). Also, fraud only comes to light if it is investigated.…”

Authors’ reply re: Plagiarism and data falsification are the most common reasons for retracted publications in obstetrics and gynaecology

“…We want to thank Dr Li and Dr Mol for their interest in our work investigating article retractions in the obstetrics and gynaecology literature and appreciate the opportunity to respond …”

“…During our data collection, we observed that retraction notices are not uniform and frequently contain insufficient information to determine the cause of retraction. Our study was limited to the investigation of retracted articles and notices in the National Library of Medicine PubMed database . However, using ‘Retraction Watch’, Drs Li and Mol were able to find an additional six RCTs retracted owing to scientific misconduct .…”