Permutation tests for stepped-wedge cluster-randomized trials
Permutation tests for stepped-wedge cluster-randomized trials
复制标题
DOI：
10.1177/1536867x19893624
复制
发表时间：
2019-12-01
期刊：
影响因子：
4.8
通讯作者：
Fielding, Katherine
中科院分区：
文献类型：
--
作者：
Thompson, Jennifer;Davey, Calum;Fielding, Katherine
排列检验在阶梯楔形试验中很有用，可以为干预效应估计提供可靠的统计检验。但是，在此设置中，置换命令不会生成有效检验，因为单个观测不可交换。我们引入swpermute命令，它将簇排列为序列以保持交换。该命令提供了执行阶梯楔试验分析的附加功能。特别是，我们包括期内选项，它在研究的每个时期分别执行指定的分析，并将产生的特定时期的干预效应估计值合并为加权平均值。我们还包括测试非零假设的功能，以帮助构建置信区间。使用来自一项试验的数据给出了swpermute应用的例子，该试验测试了一种新的结核病诊断测试对结核病诊断的细菌确认的影响。
Permutation tests are useful in stepped-wedge trials to provide robust statistical tests of intervention-effect estimates. However, the permute command does not produce valid tests in this setting because individual observations are not exchangeable. We introduce the swpermute command, which permutes clusters to sequences to maintain exchangeability. The command provides additional functionality for performing analyses of stepped-wedge trials. In particular, we include the withinperiod option, which performs the specified analysis separately in each period of the study with the resulting period-specific intervention-effect estimates combined as a weighted average. We also include functionality to test nonzero null hypotheses to aid in the construction of confidence intervals. Examples of the application of swpermute are given using data from a trial testing the impact of a new tuberculosis diagnostic test on bacterial confirmation of a tuberculosis diagnosis.