School of Medicine, University of Missouri - Kansas City School of Medicine, Kansas City, Missouri, USA.
Department of Neurosurgery, Emory University, Atlanta, Georgia, USA.
World Neurosurg. 2021 Jul;151:e988-e994. doi: 10.1016/j.wneu.2021.05.021. Epub 2021 May 18.
Academic misrepresentation is not an unknown phenomenon, with recent reports in neurosurgery detecting a 45% misrepresentation rate in prospective neurosurgical residents. The purpose of this study was to determine current rates of academic misrepresentation by prospective neurosurgical residents at a single institution across 2 distinct application cycles.
We retrospectively reviewed all Electronic Residency Application Service applications to 1 institution's neurosurgical residency program in the 2015 (n = 320) and 2020 (n = 355) application cycles. Reported academic works were verified through an extensive Web search of PubMed, Google Scholar, and the individual journal Web sites. Misrepresentation was defined in our study as listing work that does not exist, self-promotion to primary authorship, self-promotion (excluding primary authorship), incorrectly listing online-only publications, and listing non-peer-reviewed work as peer-reviewed.
In 2015, 253 (79.1%) applicants reported a total of 2097 citations and 305 (85.9%) applicants reported a total of 3018 citations in 2020 (P < 0.05). Median peer-reviewed articles per applicant rose significantly in 2020 (3.0 vs. 4.0, P < 0.001). Misrepresentation rates decreased dramatically in 2020 to 18.4% from a previously reported misrepresentation rate of 45% in 2012 (P < 0.0001). Increased United States Medical Licensing Exam Step 2 scores were associated with a decreased likelihood of misrepresentation (odds ratio = 0.97, P < 0.001).
Misrepresentation rates within neurosurgical residency candidates have significantly decreased despite an increase in reported citations. A variety of steps including education, modifying reporting methods, and increased screening may help even further decrease misrepresentation.
学术造假并非未知现象，最近有神经外科的报道称，有 45%的拟神经外科住院医师存在学术造假。本研究的目的是确定单一机构的神经外科住院医师在两个不同申请周期内的学术造假率。
我们回顾性地审查了 2015 年（n=320）和 2020 年（n=355）申请周期中，一家机构神经外科住院医师项目的所有电子住院医师申请服务（Electronic Residency Application Service，ERAS）申请。通过对 PubMed、Google Scholar 和各个期刊网站的广泛网络搜索，对报告的学术成果进行了验证。在我们的研究中，学术造假被定义为列出不存在的作品、将自己列为第一作者、自我宣传（不包括第一作者）、错误地列出仅在线出版物以及将非同行评议的作品列为同行评议。
2015 年，253 名（79.1%）申请人共报告了 2097 项引用，2020 年 305 名（85.9%）申请人共报告了 3018 项引用（P<0.05）。2020 年，每位申请人的同行评议文章中位数显著增加（3.0 篇比 4.0 篇，P<0.001）。与 2012 年报告的 45%的学术造假率相比，2020 年的学术造假率显著下降至 18.4%（P<0.0001）。美国医师执照考试（United States Medical Licensing Exam，USMLE）第二阶段成绩的提高与学术造假的可能性降低相关（比值比=0.97，P<0.001）。
尽管报告的引用量有所增加，但神经外科住院医师候选人中的学术造假率已显著下降。包括教育、修改报告方法和增加筛选在内的各种措施可能有助于进一步降低学术造假率。
