From the Department of Orthopaedic Surgery and Rehabilitation Medicine, University of Chicago, Chicago, IL (Scott, Dirschl), and the Department of Orthopaedic Surgery, University of Kentucky, Lexington, KY (Landy).
J Am Acad Orthop Surg. 2021 Dec 1;29(23):e1274-e1281. doi: 10.5435/JAAOS-D-20-00733.
Level of evidence grading has become widely used in orthopaedics. This study reviewed clinical research articles published in leading orthopaedic journals to describe the association between level of evidence and number of future citations, which is one measure of an article's impact in the field.
The first 100 clinical research articles published in 2014 by each of the Journal of Bone and Joint Surgery, Clinical Orthopaedics and Related Research, and the American Journal of Sports Medicine were reviewed for level of evidence and article characteristics. Web of Science was used to identify the number of citations of each article over the following 5 years. Univariable analyses and multivariable linear regression were used to describe the associations.
Three hundred articles were evaluated. Univariable analysis revealed no association between level of evidence and number of citations, with a median number of citations for level 1 articles of 23 (interquartile range [IQR], 14-49), level 2 articles 24 (IQR, 13-47), level 3 articles 22 (IQR, 13-40), and level 4 or 5 articles 20 (IQR, 10-36). Univariable analyses showed weak associations between other article characteristics and citations. Even after adjusting for other variables, the standardized regression coefficient for level 1 versus level 4 or 5 was only 0.14 and the overall model had a poor fit with an R2 of 0.18.
Among clinical research articles published in leading orthopaedic journals, no notable association was found between level of evidence and future citations.
Readers of the orthopaedic literature should understand that no association was found between level of evidence and future citations. Additional work is needed to better understand the effect level of evidence has on clinicians and researchers.
证据分级在骨科领域得到了广泛应用。本研究回顾了发表在领先骨科期刊上的临床研究文章，描述了证据水平与未来引文数量之间的关系，这是衡量文章在该领域影响力的一个指标。
对《骨科与相关研究杂志》《临床骨科》和《美国运动医学杂志》2014 年发表的前 100 篇临床研究文章的证据水平和文章特征进行了回顾。利用 Web of Science 确定了每篇文章在随后 5 年内的引用次数。采用单变量分析和多变量线性回归来描述相关性。
共评估了 300 篇文章。单变量分析显示，证据水平与引文数量之间无显著相关性，1 级文章的中位数引文数为 23（四分位距 [IQR]，14-49），2 级文章为 24（IQR，13-47），3 级文章为 22（IQR，13-40），4 级或 5 级文章为 20（IQR，10-36）。单变量分析显示，其他文章特征与引文之间存在较弱的相关性。即使在调整了其他变量后，1 级与 4 级或 5 级的标准化回归系数仅为 0.14，且整体模型的拟合度较差，R2 为 0.18。
在发表于领先骨科期刊的临床研究文章中，证据水平与未来引文之间无显著相关性。需要进一步研究以更好地理解证据水平对临床医生和研究人员的影响。
骨科文献的读者应了解到，证据水平与未来引文之间无显著相关性。需要进一步研究以更好地理解证据水平对临床医生和研究人员的影响。
