A Bayesian Nonparametric Approach for Semi-Competing Risks with Application to Cardiovascular Health

2025-06-25

AI Summary

This paper addresses causal inference under semi-competing risks in cardiovascular research—where a fatal event (e.g., death) precludes observation of a non-fatal event (e.g., heart failure hospitalization). We propose a Bayesian nonparametric causal estimation framework grounded in principal stratification. Our method innovatively integrates vine copulas with an Enriched Dirichlet Process Mixture (EDPM) to flexibly model the joint distribution of event times and complex dependence structures, avoiding restrictive parametric assumptions. A sensitivity parameter explicitly encodes unverifiable identification assumptions required for causal identification. Posterior inference is performed via MCMC, yielding causal effect estimates—specifically, the treatment effect on the hazard of non-fatal events among survivors—along with rigorous uncertainty quantification. The approach demonstrates robustness and interpretability in real-world cardiovascular data and supports comprehensive sensitivity analyses across multiple scenarios.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Knowledge Representation and Reasoning: Nonmonotonic Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Security and Privacy: Large-scale security measurements
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

We address causal estimation in semi-competing risks settings, where a non-terminal event may be precluded by one or more terminal events. We define a principal-stratification causal estimand for treatment effects on the non-terminal event, conditional on surviving past a specified landmark time. To estimate joint event-time distributions, we employ both vine-copula constructions and Bayesian nonparametric Enriched Dirichlet-process mixtures (EDPM), enabling inference under minimal parametric assumptions. We index our causal assumptions with sensitivity parameters. Posterior summaries via MCMC yield interpretable estimates with credible intervals. We illustrate the proposed method using data from a cardiovascular health study.

Problem

• Causal estimation in semi-competing risks scenarios
• Modeling joint event-time distributions with minimal assumptions
• Assessing treatment effects on non-terminal cardiovascular events

Innovation

• Bayesian nonparametric Enriched Dirichlet-process mixtures
• Vine-copula constructions for joint distributions
• Principal-stratification causal estimand with sensitivity parameters