Methods for including information from multi-arm trials in pairwise meta-analysis
Methods for including information from multi-arm trials in pairwise meta-analysis
复制标题
DOI：
10.1002/jrsm.1259
复制
发表时间：
2017-12-01
影响因子：
9.8
通讯作者：
Schwarzer, Guido
中科院分区：
文献类型：
--
作者：
Ruecker, Gerta;Cates, Christopher J.;Schwarzer, Guido
进行成对荟萃分析的系统评价者有时会遇到多组研究。为了纳入这些研究，并避免分析单位错误，通常将两个或多个组合并或将对照组分开。在本教程中，我们提出了5种不同的方法，可以使用。特别是，我们提出了一种新的方法（方法4），据我们所知，还没有提出过。我们展示了它们在3个选定的数据集上的应用，讨论了它们的应用范围以及它们的优点和局限性，并给出了建议。
Systematic reviewers conducting pairwise meta-analyses sometimes encounter multi-arm studies. To include these studies, and to avoid a unit-of-analysis error, often two or more arms are combined or the control arm is split. In this tutorial, we present 5 different approaches that can be used. Particularly, we present a novel approach (method 4) that to the best of our knowledge has not been presented before. We demonstrate their application on 3 selected data sets, discuss their scope of application and their advantages and limitations, and give recommendations.