Estimating Survival Treatment Effects with Covariate Adjustment Using Propensity Score

Propensity score is widely used to estimate treatment effects in observational studies. The covariate adjustment using propensity score is the most straightforward method in the literature of causal inference. In this article, we estimate the survival treatment effect with covariate adjustment using propensity score in the semiparametric accelerated failure time model. We establish the asymptotic properties of the proposed estimator by simultaneous estimating equations. We conduct simulation studies to evaluate the finite sample performance of the proposed method. A real data set from the German Breast Cancer Study Group is analyzed to illustrate the proposed method.