Janega Jessica B, Murray David M, Varnell Sherri P, Blitstein Jonathan L, Birnbaum Amanda S, Lytle Leslie A
Department of Psychology, The University of Memphis, 202 Psychology Building, 3693 Norriswood, Memphis, TN 38152-3230, USA.
Addict Behav. 2004 May;29(3):595-606. doi: 10.1016/j.addbeh.2004.01.002.
This article compares four mixed-model analyses valid for group-randomized trials (GRTs) involving a nested cohort design with a single pretest and a single posttest, the most common design used in GRTs. This study makes estimates of intraclass correlations (ICCs) available to investigators planning GRTs with alcohol, tobacco, and other drug measures as the outcomes of interest. It also provides formulae demonstrating the potential benefits to the standard error of the intervention effect of both adjustments for fixed and time-varying covariates, as well as correlations over time. These estimates will allow other researchers using these variables to plan their studies by performing a priori power analyses for any of four common analytic options.
本文比较了四种适用于群组随机试验（GRTs）的混合模型分析方法，这些试验采用嵌套队列设计，有一次预测试和一次后测试，这是GRTs中最常用的设计。本研究为计划以酒精、烟草和其他药物措施作为感兴趣结果的GRTs的研究人员提供了组内相关系数（ICCs）的估计值。它还提供了公式，展示了对固定和随时间变化的协变量进行调整以及随时间的相关性对干预效应标准误差的潜在益处。这些估计值将使其他使用这些变量的研究人员能够通过对四种常见分析选项中的任何一种进行先验功效分析来规划他们的研究。
