Causal mediation analysis for longitudinal and survival data in continuous time using Bayesian non-parametric joint models

2025-06-24

AI Summary

This study addresses the challenge of estimating causal treatment effects on survival and their underlying mediation mechanisms in observational cohort studies—complicated by irregular observation times, longitudinal confounding, and time-varying mediators. We propose a joint causal mediation model where exposure, confounders, mediators, and event times are all modeled as smooth functions of age. By integrating Bayesian nonparametrics with an enhanced Dirichlet process mixture (EDPM) model, our framework enables continuous-time inference—even at unobserved age points—thereby avoiding biases inherent in discrete-time approaches and enhancing flexibility and robustness in mediation effect estimation. Applied to the Atherosclerosis Risk in Communities (ARIC) cohort, the method successfully quantified both direct and indirect effects of cardiovascular medications on time to cardiovascular death, mediated through dynamic risk factors such as blood pressure and lipid levels. This work provides a generalizable methodological framework for real-world longitudinal causal inference.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Observational cohort data is an important source of information for understanding the causal effects of treatments on survival and the degree to which these effects are mediated through changes in disease-related risk factors. However, these analyses are often complicated by irregular data collection intervals and the presence of longitudinal confounders and mediators. We propose a causal mediation framework that jointly models longitudinal exposures, confounders, mediators, and time-to-event outcomes as continuous functions of age. This framework for longitudinal covariate trajectories enables statistical inference even at ages where the subject's covariate measurements are unavailable. The observed data distribution in our framework is modeled using an enriched Dirichlet process mixture (EDPM) model. Using data from the Atherosclerosis Risk in Communities cohort study, we apply our methods to assess how medication -- prescribed to target cardiovascular disease (CVD) risk factors -- affects the time-to-CVD death.

Problem

• Analyzing causal treatment effects on survival via mediators
• Handling irregular data intervals and longitudinal confounders
• Modeling longitudinal and time-to-event data continuously

Innovation

• Bayesian non-parametric joint models
• Continuous-time causal mediation framework
• Enriched Dirichlet process mixture model