Treatment Effect Estimation in Causal Survival Analysis: Practical Recommendations

Abstract

The restricted mean survival time (RMST) difference offers an interpretable causal contrast to estimate the treatment effect for time-to-event outcomes, yet a wide range of available estimators leaves limited guidance for practice. We provide a unified review of RMST estimators for randomized trials and observational studies, establish identification and asymptotic properties, and supply new derivations where needed. Our extensive simulation study compares simple nonparametric methods (such as unweighted Kaplan-Meier estimators) alongside parametric and nonparametric implementations of the G-formula, weighting approaches, Buckley-James transformations, and augmented estimators under diverse censoring mechanisms and model specifications. Across scenarios, classical Kaplan-Meier estimators (weighted when required by the censoring process) and G-formula methods perform well in randomized settings, while in observational data G-formula estimators remain competitive; however, augmented estimators such as AIPTW-AIPCW generally offer robustness to model misspecification and a favorable bias-variance trade-off. Parametric estimators perform best under correct specification, whereas nonparametric methods avoid functional assumptions but require large sample sizes to achieve reliable performance. We offer practical recommendations for estimator choice and provide open-source R code to support reproducibility and application.

Figures (16)

• Figure 1: Estimated survival curves on synthetic data. The $\theta_{\mathrm{RMST}}$ at $\tau=50$ corresponds to the yellow area between them. Curves are estimated with the Kaplan-Meier estimator, see Section \ref{['sec-theoryRCT_indc']}.
• Figure 2: Causal graph in RCT survival data with independent censoring.
• Figure 3: Causal graph in RCT survival data with dependent censoring.
• Figure 4: Illustration of Inverse-Probability-of-Censoring and Buckley-James transformations.
• Figure 5: Causal graph in observational survival data with independent censoring.

Theorems & Definitions (39)

• Definition 1.1
• Definition 2.1
• Proposition 2.1: Consistency of CUT as RMST estimator with known $\pi$
• Proposition 2.2: Consistency of CUT as RMST estimator with estimated $\hat{\pi}$
• Corollary 2.1
• Definition 2.2
• Corollary 2.2
• Corollary 2.3
• Theorem 2.1: BJ minimizes the mean squared error among CUT
• Definition 3.1