用于癌症研究培训项目毕业生纵向追踪的自动化文献检索

Automated Literature Searches for Longitudinal Tracking of Cancer Research Training Program Graduates.

作者信息

Padilla Luz A, Desmond Renee A, Brooks C Michael, Waterbor John W

机构信息

Department of Epidemiology, School of Public Health, University of Alabama at Birmingham, 1665 University Boulevard, Birmingham, AL, 35294-0022, USA.

Division of Preventive Medicine, Department of Medicine, University of Alabama at Birmingham, 1717 11th Avenue South, Birmingham, AL, 35233, USA.

出版信息

J Cancer Educ. 2018 Jun;33(3):564-568. doi: 10.1007/s13187-016-1120-4.

DOI:10.1007/s13187-016-1120-4

PMID:27734282

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC5389941/

Abstract

A key outcome measure of cancer research training programs is the number of cancer-related peer-reviewed publications after training. Because program graduates do not routinely report their publications, staff must periodically conduct electronic literature searches on each graduate. The purpose of this study is to compare findings of an innovative computer-based automated search program versus repeated manual literature searches to identify post-training peer-reviewed publications. In late 2014, manual searches for publications by former R25 students identified 232 cancer-related articles published by 112 of 543 program graduates. In 2016, a research assistant was instructed in performing Scopus literature searches for comparison with individual PubMed searches on our 543 program graduates. Through 2014, Scopus found 304 cancer publications, 220 of that had been retrieved manually plus an additional 84 papers. However, Scopus missed 12 publications found manually. Together, both methods found 316 publications. The automated method found 96.2 % of the 316 publications while individual searches found only 73.4 %. An automated search method such as using the Scopus database is a key tool for conducting comprehensive literature searches, but it must be supplemented with periodic manual searches to find the initial publications of program graduates. A time-saving feature of Scopus is the periodic automatic alerts of new publications. Although a training period is needed and initial costs can be high, an automated search method is worthwhile due to its high sensitivity and efficiency in the long term.

摘要

癌症研究培训项目的一项关键成果衡量指标是培训后与癌症相关的、经过同行评审的出版物数量。由于项目毕业生通常不会定期报告他们的出版物，工作人员必须定期对每位毕业生进行电子文献检索。本研究的目的是比较一种创新的基于计算机的自动检索程序与反复进行的手动文献检索的结果，以识别培训后经过同行评审的出版物。2014年末，对之前参与R25项目的学生的出版物进行手动检索，共识别出543名项目毕业生中的112人发表的232篇与癌症相关的文章。2016年，指导一名研究助理进行Scopus文献检索，以便与针对我们543名项目毕业生的个人PubMed检索进行比较。截至2014年，Scopus检索到304篇癌症相关出版物，其中220篇是手动检索到的，另外还有84篇。然而，Scopus遗漏了12篇手动检索到的出版物。两种方法总共发现了316篇出版物。自动检索方法找到了316篇出版物中的96.2%，而个人检索仅找到了73.4%。诸如使用Scopus数据库这样的自动检索方法是进行全面文献检索的关键工具，但必须辅以定期的手动检索，以找到项目毕业生的初始出版物。Scopus的一个节省时间的功能是定期自动提醒新出版物。虽然需要一个培训期且初始成本可能很高，但由于其长期的高灵敏度和高效率，自动检索方法是值得的。