J Med Libr Assoc. 2022 Jan 1;110(1):47-55. doi: 10.5195/jmla.2022.1280.
Systematic reviews and other evidence syntheses, the pinnacle of the evidence pyramid, embody comprehensiveness and rigor; however, retracted data are being incorporated into these publications. This study examines the use of retracted publications in the field of pharmacy, describes characteristics of retracted publications cited by systematic reviews, and discusses factors associated with citation likelihood.
Using data from Retraction Watch, we identified retracted publications in the pharmacy field. We identified all articles citing these retracted publications in Web of Science and Scopus and limited results to systematic reviews. We classified the retraction reason, determined whether the citation occurred before or after retraction, and analyzed factors associated with the likelihood of systematic reviews citing a retracted publication.
Of 1,396 retracted publications, 283 were cited 1,096 times in systematic reviews. Most (65.0%) (712/1096) citations occurred before retraction. Citations were most often to items retracted due to data falsification or manipulation (39.2%), followed by items retracted due to ethical misconduct including plagiarism (30.4%), or concerns about or errors in data or methods (26.2%). Compared to those not cited in systematic reviews, cited items were significantly more likely to be retracted due to data falsification and manipulation, were published in high impact factor journals, and had longer delays between publication and retraction.
Further analysis of systematic reviews citing retracted publications is needed to determine the impact of flawed data. Librarians understand the nuances involved and can advocate for greater transparency around the retraction process and increase awareness of challenges posed by retractions.
系统评价和其他证据综合，是证据金字塔的顶峰，体现了全面性和严谨性；然而，已撤回的数据正被纳入这些出版物中。本研究考察了撤回出版物在药学领域的使用情况，描述了系统评价引用的撤回出版物的特征，并讨论了与引用可能性相关的因素。
使用来自 Retraction Watch 的数据，我们确定了药学领域的撤回出版物。我们在 Web of Science 和 Scopus 中识别了引用这些撤回出版物的所有文章，并将结果限制在系统评价中。我们对撤回原因进行分类，确定引用是在撤回之前还是之后发生，并分析与系统评价引用撤回出版物的可能性相关的因素。
在 1396 篇撤回出版物中，有 283 篇被 1096 次引用到系统评价中。大多数（65.0%）（712/1096）引用发生在撤回之前。引用最多的是因数据伪造或操纵而撤回的项目（39.2%），其次是因伦理不当行为包括剽窃而撤回的项目（30.4%），或因数据或方法存在问题或错误而撤回的项目（26.2%）。与未被系统评价引用的项目相比，被引用的项目更有可能因数据伪造和操纵而被撤回，发表在高影响因子期刊上，且发表后撤回的时间间隔更长。
需要进一步分析引用撤回出版物的系统评价，以确定有缺陷数据的影响。图书馆员了解其中的细微差别，可以倡导在撤回过程中提高透明度，并提高对撤回带来的挑战的认识。
