[G methods for handling time-varying confounding in the longitudinal study].

The conventional analytical methods cannot effectively adjust for time-varying confounding that occur in a longitudinal study and thus cannot correctly estimate the causal effects. This study explains the necessity of precisely controlling time-varying confounding and outlines G methods, including parametric g-formula, inverse probability of weighting, and G-estimation. We also compare the methods above to provide a reference for correctly estimating causal effects in the longitudinal study.传统分析方法不能有效地控制纵向研究中的时依混杂以得到无偏因果效应估计值。本研究解释了纵向研究中正确控制时依混杂的必要性，概述了现有控制时依混杂的G方法——参数g-formula、逆概率加权和G估计，并通过比较它们的优缺点和适用情况，为研究者在纵向研究中估计因果效应提供参考。.