Type I error control for cluster randomized trials under varying small sample structures.

Type I error control for cluster randomized trials under varying small sample structures.
复制标题

不同小样本结构下整群随机试验的I型误差控制。

DOI:
10.1186/s12874-021-01236-7
复制
发表时间:
2021-04-03
影响因子:
4
通讯作者:
Kleinman KP
Kleinman KP
中科院分区:
医学3区
文献类型:
--
作者:
Nugent JR;Kleinman KP

文献摘要

参考文献

相似文献

线性混合模型(LMM)是分析整群随机试验(CRT)数据的常用方法。可以通过Wald检验或似然比检验(LRT)进行参数推断,但这两种方法在常见的有限样本设置中可能会给出不正确的I类错误率。聚类大小、聚类数、组内相关系数(ICC)和分析方法的不同组合对I类错误率的影响尚未得到很好的研究。对已发表的CRT的评论发现,小样本量并不罕见,因此在这些设置中不同推理方法的性能可以指导数据分析师做出最佳选择。使用随机截距LMM结构,我们使用模拟来研究I型错误率与LRT和Wald测试,不同的自由度(DF)的选择,在不同的组合的集群大小,集群数量,和ICC。我们的模拟表明,LRT可以反保守时,ICC是大的,簇的数量是小的,与效果最pronouced当簇的大小是相对较大的。Wald测试与内DF方法或Satterthwaite DF近似保持在规定的水平的I型错误控制,虽然它们是保守的,当集群的数量,集群的大小,和ICC是小的。根据CRT的结构,分析师应该选择一种假设检验方法,以保持其数据的适当I类错误率。Wald检验与Satterthwaite DF近似在许多情况下工作良好,但在其他情况下,LRT可能具有更接近标称水平的I型错误率。
Linear mixed models (LMM) are a common approach to analyzing data from cluster randomized trials (CRTs). Inference on parameters can be performed via Wald tests or likelihood ratio tests (LRT), but both approaches may give incorrect Type I error rates in common finite sample settings. The impact of different combinations of cluster size, number of clusters, intraclass correlation coefficient (ICC), and analysis approach on Type I error rates has not been well studied. Reviews of published CRTs find that small sample sizes are not uncommon, so the performance of different inferential approaches in these settings can guide data analysts to the best choices. Using a random-intercept LMM stucture, we use simulations to study Type I error rates with the LRT and Wald test with different degrees of freedom (DF) choices across different combinations of cluster size, number of clusters, and ICC. Our simulations show that the LRT can be anti-conservative when the ICC is large and the number of clusters is small, with the effect most pronouced when the cluster size is relatively large. Wald tests with the between-within DF method or the Satterthwaite DF approximation maintain Type I error control at the stated level, though they are conservative when the number of clusters, the cluster size, and the ICC are small. Depending on the structure of the CRT, analysts should choose a hypothesis testing approach that will maintain the appropriate Type I error rate for their data. Wald tests with the Satterthwaite DF approximation work well in many circumstances, but in other cases the LRT may have Type I error rates closer to the nominal level.
DOI: 10.1002/sim.6565
发表时间: 2015-11-30
影响因子: 2
作者:
Johnson, Jacqueline L.;Kreidler, Sarah M.;Catellier, Diane J.;Murray, David M.;Muller, Keith E.;Glueck, Deborah H.
通讯作者: Glueck, Deborah H.
DOI: 10.1186/s12874-015-0026-x
发表时间: 2015-04-23
影响因子: 4
作者:
Li P;Redden DT
通讯作者: Redden DT
DOI: 10.1056/nejmoa1207290
发表时间: 2013-06-13
影响因子: 158.5
作者:
Huang, Susan S.;Septimus, Edward;Platt, Richard
通讯作者: Platt, Richard
DOI: 10.1093/ije/dyx169
发表时间: 2018-02-01
影响因子: 7.7
作者:
Leyrat, Clemence;Morgan, Katy E.;Kahan, Brennan C.
通讯作者: Kahan, Brennan C.
DOI: 10.1098/rspa.1937.0109
发表时间: 1937-05-01
影响因子: --
作者:
Bartlett, MS
通讯作者: Bartlett, MS