期刊
STATISTICS IN MEDICINE
卷 28, 期 5, 页码 814-827出版社
WILEY
DOI: 10.1002/sim.3518
关键词
cluster-randomized trials; correlated binary data; generalized estimating equations; group-based interventions; intraclass correlation
类别
资金
- Jonsson Cancer Center Foundation
- NIH [R01GM072876, R01CA102486, P01CA109091, P30CA16042]
- California Breast Cancer Research Program of the University of California [4BB-1400]
In cluster-randomized trials, it is commonly assumed that the magnitude of the correlation among subjects within a cluster is constant across clusters. However, the correlation may in fact be heterogeneous and depend on cluster characteristics. Accurate modeling of the correlation has the potential to improve inference. We use second-order generalized estimating equations to model heterogeneous correlation in cluster-randomized trials. Using, simulation studies we show that accurate modeling of heterogeneous correlation can improve inference when the correlation is high or varies by cluster size. We apply the methods to a cluster-randomized trial of an intervention to promote breast cancer screening. Copyright (C) 2008 John Wiley & Sons, Ltd.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据