Reply to Letter to the Editor: “Technical performance of shear wave elastography for measuring liver stiffness in pediatric and adolescent patients: a systematic review and meta-analysis”
by Dong Wook Kim, Chan Park, Hee Mang Yoon, Ah Young Jung, Jin Seong Lee, Seung Chai Jung, Young Ah Cho (espoirhm@gmail.com)
Technical performance of shear wave elastography for measuring liver stiffness in pediatric and adolescent patients: a systematic review and meta-analysisDear Editor,
We would like to thank Dr. Wei and colleagues for their interest in our recent study in European Radiology [1], and appreciate the opportunity to address their comments.
Duplicate publication is the publication of an article that overlaps with an article published elsewhere, either on purpose or accidentally [2]. It may share similar hypothesis, population or sample size, methodology, results, and authors [3]. As in our study, the technique of meta-analysis has become popular for summarizing published data. Since a meta-analysis usually uses only the reported data from the publications not the original data per se, it can easily fall into the risk of including overlapped data. Indeed, inclusion of duplicate publication or overlapped data in a meta-analysis would result in overestimation of effect and reliability of the outcomes [4]. Therefore, identifying duplicate publication (or its overlapped data) is essential in conducting a meta-analysis.
It is sometimes not easy to discern duplicate publication since it may take various forms, from identical manuscripts to totally different authors and outcomes [2]. At the same time, it is also essential to include all the eligible studies for a meta-analysis to avoid selection bias. Thus, the Cochrane Handbook for Systematic Review for Interventions recommends some ‘detective work’ using the following criteria: author name, location and setting, specific details of the intervention, number of participants and baseline data, and study period [5].
In our work, we scrutinized all the candidate studies to detect and exclude duplicate or any overlapped population and described it as exclusion criteria in our publication. We even tried to contact to the authors only if uncertainties remained. As the readers pointed out, we already noticed that the two studies of interest shared the authors and study period [6, 7]. But it turned out that one [6] included children who received liver transplantation, whereas the other [7] was among those who have native liver; they did not share the population and are completely different studies.