Sensitivity analyses for unmeasured confounding assuming a marginal structural model for repeated measures
Sensitivity analyses for unmeasured confounding assuming a marginal structural model for repeated measures
复制标题
DOI：
10.1002/sim.1657
复制
发表时间：
2004-03-15
影响因子：
2
通讯作者：
Robins, JM
中科院分区：
文献类型：
--
作者：
Brumback, BA;Hernán, MA;Robins, JM
罗宾斯引入了边际结构模型（MSM）和处理加权逆概率（IPTW）估计器，用于评估时变处理对重复测量平均值的因果影响。我们研究了 IPTW 估计器对未测量的混杂因素的敏感性。我们研究了一种基于不可识别模型的新的敏感性分析框架，该模型根据敏感性参数和用户指定的函数来量化不可测量的混杂因素。我们提出了 MSM 参数的增强 IPTW 估计器，并证明了它们对于 MSM 因果效应的一致性，假设未测量的混杂有正确的混杂偏差函数。我们应用这些方法来评估 Hernan 等人分析的敏感性，他们在多中心艾滋病队列研究中使用 MSM 来估计齐多夫定治疗对 HIV 感染男性重复 CD4 计数的因果影响。假设没有未测量的混杂因素，治疗效果的 95% 置信区间包括零。我们表明，在存在适量未测量混杂因素的假设下，治疗效果的 95% 置信区间不再包括零。因此，埃尔南等人的分析。对不可测量的混杂因素有些敏感。我们希望我们的研究能够鼓励和促进对其他应用中无法测量的混杂因素的敏感性分析。版权所有 (C) 2004 John Wiley Sons, Ltd.
Robins introduced marginal structural models (MSMs) and inverse probability of treatment weighted (IPTW) estimators for the causal effect of a time-varying treatment on the mean of repeated measures. We investigate the sensitivity of IPTW estimators to unmeasured confounding. We examine a new framework for sensitivity analyses based on a nonidentifiable model that quantifies unmeasured confounding in terms of a sensitivity parameter and a user-specified function. We present augmented IPTW estimators of MSM parameters and prove their consistency for the causal effect of an MSM, assuming a correct confounding bias function for unmeasured confounding. We apply the methods to assess sensitivity of the analysis of Hernan et al., who used an MSM to estimate the causal effect of zidovudine therapy on repeated CD4 counts among HIV-infected men in the Multicenter AIDS Cohort Study. Under the assumption of no unmeasured confounders, a 95 per cent confidence interval for the treatment effect includes zero. We show that under the assumption of a moderate amount of unmeasured confounding, a 95 per cent confidence interval for the treatment effect no longer includes zero. Thus, the analysis of Hernan et al. is somewhat sensitive to unmeasured confounding. We hope that our research will encourage and facilitate analyses of sensitivity to unmeasured confounding in other applications. Copyright (C) 2004 John Wiley Sons, Ltd.