Surrogate-powered Causal Inference with Censored Outcomes
Abstract
Clinical trials with survival endpoints lose information when participants are censored before death is observed. We develop target-preserving estimators that use posttreatment disease history, such as recurrence or progression, to recover information lost to censoring for marginal survival and restricted mean survival effects. The difficulty is that the intermediate event is downstream of treatment: naive adjustment can change the causal estimand, and the useful information enters only through the observed coarsening. We derive observed-data influence functions with and without recurrence history and obtain an exact gain identity. The identity shows that efficiency improvement is driven by the censoring hazard, the split of the alive risk set into recurrence states, and the residual-survival separation between those states. In the no-covariate illness-death model, the Aalen-Johansen estimator realizes the recurrence-augmented efficient score after standardization to the marginal target. With covariates, correctly specified Cox-Breslow transition hazards provide a root-N plug-in benchmark, while a hazard-induced one-step estimator gives rate robustness and, under primitive transition and censoring learner rates, canonical inference with a second-order product remainder. A semi-synthetic metastatic breast cancer study calibrated from digitized progression-free survival and overall survival curves illustrates the gain identity. The framework applies broadly to censored time-to-event studies with informative intermediate histories.
Related
- Flexible Doubly Robust Proximal Causal Inference for Censored Outcomes
- Efficient Causal Survival Inference under Limited Surrogate Monitoring
- GenAI Powered Dynamic Causal Inference with Unstructured Data
- Causal Inference for Heterogeneous Extreme Quantiles with Heavy-Tailed Outcomes
- Is the treatment beneficial? Causal Inference with Ordinal Outcomes: A Bayesian Solution
BibTeX
@misc{mukhin2026surrogatepowered,
title = {{Surrogate-powered Causal Inference with Censored Outcomes}},
author = {Yaroslav Mukhin and Tereza Oprea and Arielle Anderer and Christina Lee Yu and Jelena Bradic},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.05486},
eprint = {2610.05486},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.05486},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Surrogate-powered Causal Inference with Censored Outcomes
AU - Yaroslav Mukhin
AU - Tereza Oprea
AU - Arielle Anderer
AU - Christina Lee Yu
AU - Jelena Bradic
PY - 2026
DA - 2026/10/04
PB - arXiv
DO - 10.48550/arxiv.2610.05486
UR - https://arxiv.org/abs/2610.05486
AB - Clinical trials with survival endpoints lose information when participants are censored before death is observed. We develop target-preserving estimators that use posttreatment disease history, such as recurrence or progression, to recover information lost to censoring for marginal survival and restricted mean survival effects. The difficulty is that the intermediate event is downstream of treatment: naive adjustment can change the causal estimand, and the useful information enters only through the observed coarsening. We derive observed-data influence functions with and without recurrence history and obtain an exact gain identity. The identity shows that efficiency improvement is driven by the censoring hazard, the split of the alive risk set into recurrence states, and the residual-survival separation between those states. In the no-covariate illness-death model, the Aalen-Johansen estimator realizes the recurrence-augmented efficient score after standardization to the marginal target. With covariates, correctly specified Cox-Breslow transition hazards provide a root-N plug-in benchmark, while a hazard-induced one-step estimator gives rate robustness and, under primitive transition and censoring learner rates, canonical inference with a second-order product remainder. A semi-synthetic metastatic breast cancer study calibrated from digitized progression-free survival and overall survival curves illustrates the gain identity. The framework applies broadly to censored time-to-event studies with informative intermediate histories.
ER -