Efficient transport and generalization of survival treatment effects
Abstract
Randomized controlled trials provide internally valid estimates of treatment effects, but their results may not directly apply to broader target populations due to differences in baseline covariate distributions, adherence to treatment or variations in outcome mechanisms. Under standard transport and time-to-event identifiability assumptions, we develop nonparametric, debiased machine learning estimators for transporting and generalizing causal survival treatment effect differences from a source population to a target population in discrete time. We derive the efficient influence functions for the transport and generalization survival difference estimands and propose cross-fitted one-step estimators that are doubly robust and achieve semiparametric efficiency bounds under weak regularity conditions. We further introduce estimators that exploit known effect modifier subsets through an additive parameterization of the survival function, reducing the dimensionality of the reweighting and yielding smaller or equal asymptotic variance. We establish asymptotic normality, double robustness, and rates of convergence for all proposed estimators. Finite-sample properties are illustrated through Monte Carlo simulations under flexible and misspecified nuisance estimation scenarios. We apply the methods to data from the Women's Health Initiative to estimate the effect of hormone therapy on coronary heart disease across trial and observational populations.
Related
- Post-Progression Treatment Regimes and the Limits of Cross-Trial Survival Transport
- Efficient Causal Survival Inference under Limited Surrogate Monitoring
- Max-Stable Survival Copulas under Dependent Censoring: Sharp Identification and Efficient Inference
- Exact Survival Recurrence beyond Rate Homogenization in Periodic Transport
- Sparse Dictionary Learning for Measure-Valued Data via Linearized Optimal Transport: Efficient Algorithms and Sample Complexity
BibTeX
@misc{martin2026efficient,
title = {{Efficient transport and generalization of survival treatment effects}},
author = {Axel Martin and Iván Díaz and Michele Santacatterina},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.18764},
eprint = {2609.18764},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.18764},
}
RIS
TY - UNPB
TI - Efficient transport and generalization of survival treatment effects
AU - Axel Martin
AU - Iván Díaz
AU - Michele Santacatterina
PY - 2026
DA - 2026/09/16
PB - arXiv
DO - 10.48550/arxiv.2609.18764
UR - https://arxiv.org/abs/2609.18764
AB - Randomized controlled trials provide internally valid estimates of treatment effects, but their results may not directly apply to broader target populations due to differences in baseline covariate distributions, adherence to treatment or variations in outcome mechanisms. Under standard transport and time-to-event identifiability assumptions, we develop nonparametric, debiased machine learning estimators for transporting and generalizing causal survival treatment effect differences from a source population to a target population in discrete time. We derive the efficient influence functions for the transport and generalization survival difference estimands and propose cross-fitted one-step estimators that are doubly robust and achieve semiparametric efficiency bounds under weak regularity conditions. We further introduce estimators that exploit known effect modifier subsets through an additive parameterization of the survival function, reducing the dimensionality of the reweighting and yielding smaller or equal asymptotic variance. We establish asymptotic normality, double robustness, and rates of convergence for all proposed estimators. Finite-sample properties are illustrated through Monte Carlo simulations under flexible and misspecified nuisance estimation scenarios. We apply the methods to data from the Women's Health Initiative to estimate the effect of hormone therapy on coronary heart disease across trial and observational populations.
ER -