限制数据库选择对快速综述检索引文与相关研究比例的影响
该meta-research研究基于三篇慢性原发性腰痛干预系统综述的原始检索文件,比较仅检索CENTRAL、MEDLINE和Embase与完整检索的差异。限制检索的引文相对产出为54.7%至86.0%,相对查全率为65.4%至100.0%,最多遗漏34.6%的相关研究。结论认为限制数据库可减少工作量但可能引入选择偏倚,是否纳入专题数据库应视具体主题而定。
要点
- 限制为CENTRAL、MEDLINE和Embase时相对查全率65.4%至100.0%,最多遗漏34.6%相关研究。
- Embase、CINAHL和ICTRP识别的所有研究均已被MEDLINE或CENTRAL覆盖。
- 针刺综述遗漏的研究均来自中文数据库,提示专题数据库需按主题决定是否纳入。
核心结论
- 限制数据库选择可减少筛查工作量并检索到多数研究,但可能遗漏相关研究。
- 在慢性腰痛综述中仅检索两个生物医学数据库足以识别限制检索中的所有相关研究。
- 当专题数据库是主要贡献来源时,排除它们可能引入选择偏倚,应逐案决定是否纳入。
研究设计与样本
- 基于三篇慢性原发性腰痛干预(针刺、教育、TENS)随机对照试验系统综述的原始检索文件进行meta-research研究。
- 完整检索包括CENTRAL、MEDLINE、Embase、CINAHL和ICTRP,针刺综述另检索CNKI和万方,教育和TENS综述另检索PEDro。
- 限制检索按Cochrane快速综述方法组建议仅纳入CENTRAL、MEDLINE和Embase。
主要发现
- 相对产出范围为54.7%至86.0%,相对查全率范围为65.4%至100.0%,最多遗漏34.6%的相关研究。
- 针刺综述遗漏的研究均在中文数据库中被识别,TENS综述遗漏的1项研究在PEDro中被识别。
- 增量查全率:CENTRAL为0.0%至20.0%,MEDLINE为7.7%至20.0%,Embase为0.0%。
- 需阅读数:CENTRAL为11.0至158.7,MEDLINE为31.3至172.8,Embase为14.2至68.5。
临床怎么用
- 快速综述可优先检索MEDLINE和CENTRAL以提高效率,但需评估主题是否依赖专题数据库。
- 若研究主题涉及中文文献或特定专业数据库,应逐案决定是否纳入相应数据库以避免遗漏。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
OBJECTIVE: To determine the proportion of citations retrieved and recall when restricting database selection in three systematic reviews of interventions for the management of chronic primary low back pain (LBP). We also describe where missed studies were identified, duplication across databases and incremental recall of each database. STUDY DESIGN AND SETTING: We used original search files from three systematic reviews of randomized controlled trials for interventions (acupuncture, education, and transcutaneous electrical nerve stimulation [TENS]) conducted to inform the WHO guideline on nonsurgical management of chronic primary LBP. Databases searched were CENTRAL (C), MEDLINE (M), Embase (E), CINAHL, and International Clinical Trials Registry Platform. The acupuncture review also included China National Knowledge Infrastructure and WanFang, and the education and TENS reviews included PEDro. Following Cochrane Rapid Review Methods Group recommendations, the restricted search included CENTRAL, MEDLINE, and Embase. We report the relative yield, proportion of citations retrieved compared to the full search, and relative recall, proportion of relevant studies in the restricted search compared to the full search. We calculated incremental recall of each database (unique studies from each database) and cumulative number needed to read (number of citations/number of relevant studies). RESULTS: The relative yield ranged from 54.7% to 86.0. Relative recall ranged from 65.4% to 100.0%, missing up to 34.6% of relevant studies. All studies missed in the acupuncture review were identified in Chinese databases and the study missed in the TENS review was identified in Physiotherapy Evidence Database. Incremental recall ranged from 0.0% to 20.0% for CENTRAL, 7.7% to 20.0% for MEDLINE and 0.0% for Embase. Number needed to read ranged from 11.0 to 158.7 for CENTRAL, 31.3 to 172.8 for MEDLINE, and 14.2 to 68.5 for Embase. CONCLUSIONS: Restricting database selection reduces the workload while retrieving most studies. In reviews of chronic LBP, the selection of two biomedical databases is sufficient to identify all relevant studies from a restricted search. However, there is a risk of missing relevant studies when topic-specific databases are key contributors. Excluding such databases may introduce selection bias; therefore, inclusion of topic-specific databases should be determined on a case-by-case basis. PLAIN LANGUAGE SUMMARY: Rapid reviews are a streamlined version of systematic reviews, designed to provide evidence for decision-makers more efficiently. This study looked at what happens when researchers search a smaller number of databases to identify studies for rapid reviews to determine whether this trade-off leads to missing studies. We examined three existing systematic reviews on treatments for chronic LBP (acupuncture, education, TENS). We compared the original full search, which included major medical databases, topic-specific databases, and trial registries to a more limited search that only included three large databases: MEDLINE, Embase, and CENTRAL. We identified which citations came from which database and calculated how many total citations and relevant studies were found using the restricted approach. When fewer databases were searched, 55% to 86% of citations from the full searches were retrieved. More importantly, the restricted search missed up to 35% of relevant studies suggesting that rapid reviews may omit studies included in full reviews. We also found that excluding EMBASE did not lead to bias because all studies identified in Embase, CINAHL, and International Clinical Trials Registry Platform were already identified in MEDLINE or CENTRAL. In the acupuncture review, including the CNKI database added screening work (419 records) but yielded 31% additional studies, while adding Physiotherapy Evidence Database to the education and TENS reviews did not add much screening work and added 1 relevant study in the TENS review. In summary, our findings suggest that restricting database selection to MEDLINE and CENTRAL can improve efficiency but may risk selection bias with certain review topics. For example, Chinese databases contributed many citations to the acupuncture review, so excluding them could lead to missed studies. Therefore, the decision to exclude databases should be informed by the topic under review.
引用
- Corso M, Brunton G, Negrini S, et al. Evaluating the proportion of retrieved citations and relevant studies when database selection is restricted to MEDLINE, CENTRAL, and Embase: part 4 of a meta-research study of rapid review methodological trade-offs. J Clin Epidemiol. 2026. PMID: 42785648. doi:10.1016/j.jclinepi.2026.112535
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。