A Bayesian Approach to Causal Cure Models

2026-07-21

AI Summary

This study addresses causal inference for time-to-event data with a cured fraction by proposing a Bayesian framework that jointly models treatment effects on both the cure proportion and the survival function of the uncured subgroup. Building upon principal stratification, the method establishes a Bayesian inference system directly linked to classical mixture cure models and employs MCMC algorithms for flexible and efficient posterior computation. It innovatively defines causal estimands for survival effects tailored to subpopulations that are not always uncured. Simulation studies demonstrate the approach’s robustness and estimation efficiency, and its practical utility is confirmed through successful application to a randomized controlled trial involving patients with postoperative hypoxemic respiratory failure following abdominal surgery.

Abstract

Time-to-event data often include individuals who will never experience the failure event, and are therefore considered cured. In such settings, frequently encountered in clinical research, a substantial proportion of patients may remain event-free throughout the observation period, leading to the appearance of a survival plateau that is commonly interpreted as evidence of a cured fraction. Analysing such data requires inference on the cure fraction and on the survival function for the uncured subpopulation, tasks which are traditionally achieved with mixture cure models. However, assessing the causal effect of a treatment on these quantities is non-trivial. We consider principal stratification causal estimands, which have been proposed to evaluate effects on the cure fraction and on the survival for an always-uncured stratum. We additionally introduce a novel estimand, which considers the causal effect on the survival for a non-always-cured union of strata. We frame the problem from a Bayesian model-based perspective, which provides a flexible and unified estimation strategy while maintaining a direct link with classical mixture cure model quantities. The reliability of the proposed approach is validated through simulations, demonstrating competitive and robust performance relative to existing methods. Finally, we illustrate its practical usefulness through an application to a randomized trial comparing non-invasive ventilation with standard oxygen therapy in patients with hypoxemic respiratory failure following abdominal surgery.

Problem

• cure models
• causal inference
• time-to-event data
• principal stratification
• Bayesian approach

Innovation

• Bayesian causal inference
• mixture cure models
• principal stratification
• survival plateau
• non-always-cured strata