纵向修改治疗政策的中介效应识别与估计
摘要
我们演示了一种综合的半参数方法，用于因果中介分析，解决纵向连续处理、混杂因素和中介因素的复杂性。我们的 метод论利用非参数结构方程模型和基于 doubly robust 伪结果的交叉拟合顺序回归技术，yielding an efficient, asymptotically normal estimator without relying on restrictive parametric modeling assumptions. We are motivated by a recent scientific controversy regarding the effects of invasive mechanical ventilation (IMV) on the survival of COVID-19 patients, considering acute kidney injury (AKI) as a mediating factor. We highlight the possibility of "inconsistent mediation," in which the direct and indirect effects of the exposure operate in opposite directions. We discuss the significance of mediation analysis for scientific understanding and its potential utility in treatment decisions.
引用
@article{arxiv.2403.09928,
title = {Identification and estimation of mediational effects of longitudinal modified treatment policies},
author = {Brian Gilbert and Katherine L. Hoffman and Nicholas Williams and Kara E. Rudolph and Edward J. Schenck and Iván Díaz},
journal= {arXiv preprint arXiv:2403.09928},
year = {2025}
}
备注
revised interpretation of estimand in application, additional simulation, minor changes throughout