Mathes Tim, Pieper Dawid
Institute for Research in Operative Medicine, Chair of Surgical Research, Faculty of Health, School of Medicine, Witten/Herdecke University, Ostmerheimer Str. 200, 51109, Cologne, Germany.
BMC Med Res Methodol. 2017 Jul 17;17(1):107. doi: 10.1186/s12874-017-0391-8.
Distinguishing cohort studies from case series is difficult.We propose a conceptualization of cohort studies in systematic reviews of comparative studies. The main aim of this conceptualization is to clarify the distinction between cohort studies and case series. We discuss the potential impact of the proposed conceptualization on the body of evidence and workload.All studies with exposure-based sampling gather multiple exposures (with at least two different exposures or levels of exposure) and enable calculation of relative risks that should be considered cohort studies in systematic reviews, including non-randomized studies. The term "enables/can" means that a predefined analytic comparison is not a prerequisite (i.e., the absolute risks per group and/or a risk ratio are provided). Instead, all studies for which sufficient data are available for reanalysis to compare different exposures (e.g., sufficient data in the publication) are classified as cohort studies.There are possibly large numbers of studies without a comparison for the exposure of interest but that do provide the necessary data to calculate effect measures for a comparison. Consequently, more studies could be included in a systematic review. Therefore, on the one hand, the outlined approach can increase the confidence in effect estimates and the strengths of conclusions. On the other hand, the workload would increase (e.g., additional data extraction and risk of bias assessment, as well as reanalyses).
区分队列研究和病例系列很困难。我们提出了一种在比较研究的系统评价中对队列研究的概念化方法。这种概念化的主要目的是厘清队列研究和病例系列之间的区别。我们讨论了所提出的概念化对证据主体和工作量的潜在影响。所有基于暴露抽样的研究收集多种暴露情况（至少有两种不同的暴露或暴露水平），并能够计算相对风险，在系统评价中应将其视为队列研究，包括非随机研究。术语“能够/可以”意味着预先定义的分析比较不是前提条件（即每组的绝对风险和/或风险比已给出）。相反，所有有足够数据可用于重新分析以比较不同暴露情况的研究（例如，出版物中有足够的数据）都归类为队列研究。可能有大量关于感兴趣暴露情况没有对照的研究，但确实提供了计算比较效应量所需的数据。因此，更多的研究可以纳入系统评价。所以，一方面，所概述的方法可以增加对效应估计的信心和结论的力度。另一方面，工作量会增加（例如，额外的数据提取、偏倚风险评估以及重新分析）。
