Gorrell Lindsay M, Engel Roger M, Lystad Reidar P, Brown Benjamin T
Human Performance Laboratory, KNB 222, Faculty of Kinesiology, University of Calgary, Calgary, Alberta, T2N 1 N4, Canada.
Department of Chiropractic, Macquarie University, Building C5C West, Sydney, 2109, Australia.
BMC Med Res Methodol. 2017 Mar 14;17(1):41. doi: 10.1186/s12874-017-0320-x.
Reporting of adverse events in randomized clinical trials (RCTs) is encouraged by the authors of The Consolidated Standards of Reporting Trials (CONSORT) statement. With robust methodological design and adequate reporting, RCTs have the potential to provide useful evidence on the incidence of adverse events associated with spinal manipulative therapy (SMT). During a previous investigation, it became apparent that comprehensive search strategies combining text words with indexing terms was not sufficiently sensitive for retrieving records that were known to contain reports on adverse events. The aim of this analysis was to compare the proportion of articles containing data on adverse events associated with SMT that were indexed in MEDLINE and/or EMBASE and the proportion of those that included adverse event-related words in their title or abstract.
A sample of 140 RCT articles previously identified as containing data on adverse events associated with SMT was used. Articles were checked to determine if: (1) they had been indexed with relevant terms describing adverse events in the MEDLINE and EMBASE databases; and (2) they mentioned adverse events (or any related terms) in the title or abstract.
Of the 140 papers, 91% were MEDLINE records, 85% were EMBASE records, 81% were found in both MEDLINE and EMBASE records, and 4% were not in either database. Only 19% mentioned adverse event-related text words in the title or abstract. There was no significant difference between MEDLINE and EMBASE records in the proportion of available papers (p = 0.078). Of the 113 papers that were found in both MEDLINE and EMBASE records, only 3% had adverse event-related indexing terms assigned to them in both databases, while 81% were not assigned an adverse event-related indexing term in either database.
While there was effective indexing of RCTs involving SMT in the MEDLINE and EMBASE databases, there was a failure of allocation of adverse event indexing terms in both databases. We recommend the development of standardized definitions and reporting tools for adverse events associated with SMT. Adequate reporting of adverse events associated with SMT will facilitate accurate indexing of these types of manuscripts in the databases.
《试验报告的统一标准》（CONSORT）声明的作者鼓励在随机临床试验（RCT）中报告不良事件。凭借稳健的方法设计和充分的报告，随机对照试验有潜力提供关于与脊柱推拿疗法（SMT）相关的不良事件发生率的有用证据。在之前的一项调查中，很明显，将文本词与索引词相结合的综合检索策略在检索已知包含不良事件报告的记录时不够敏感。本分析的目的是比较在MEDLINE和/或EMBASE中索引的包含与SMT相关的不良事件数据的文章比例，以及在标题或摘要中包含不良事件相关词汇的文章比例。
使用了先前确定的140篇包含与SMT相关的不良事件数据的随机对照试验文章样本。检查文章以确定：（1）它们是否已在MEDLINE和EMBASE数据库中使用描述不良事件的相关术语进行索引；（2）它们在标题或摘要中是否提及不良事件（或任何相关术语）。
在140篇论文中，91%是MEDLINE记录，85%是EMBASE记录，81%同时出现在MEDLINE和EMBASE记录中，4%不在这两个数据库中。只有19%在标题或摘要中提到了不良事件相关的文本词。MEDLINE和EMBASE记录中可用论文的比例没有显著差异（p = 0.078）。在MEDLINE和EMBASE记录中都找到的113篇论文中，只有3%在两个数据库中都被赋予了不良事件相关的索引词，而81%在两个数据库中都没有被赋予不良事件相关的索引词。
虽然MEDLINE和EMBASE数据库中对涉及SMT的随机对照试验进行了有效的索引，但两个数据库中都没有分配不良事件索引词。我们建议为与SMT相关的不良事件制定标准化定义和报告工具。充分报告与SMT相关的不良事件将有助于在数据库中对这类手稿进行准确索引。
