Dependent stochastic block models for age-indexed sequences of directed causes-of-death networks

2025-10-02

AI Summary

This study addresses the lack of stochastic block models tailored to age-indexed directed network sequences in mortality modeling. We propose a novel Bayesian dependent random partition prior model to jointly infer smooth, age-evolving group structures underlying underlying and contributing causes of death. Our method integrates directed network sequence modeling, Bayesian nonparametric inference, and MCMC sampling—enabling, for the first time, a unified characterization of modular co-occurrence patterns between cause types and their age-varying interactions. On synthetic data, our approach outperforms existing methods in accuracy and robustness. Applied to U.S. mortality records, it successfully identifies age-specific cause group compositions and their temporal evolution, revealing the population-level interaction mechanisms underpinning “deaths of despair” and uncovering previously undetected structural features in cause-of-death networks.

Technology Category

• Reasoning under Uncertainty: Relational Probabilistic Models
• Machine Learning: Probabilistic Circuits and Graphical Models
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• User Modeling, Personalization and Recommendation: Social recommender systems and personalization for the social Web

Abstract

Death events commonly arise from complex interactions among interrelated causes, formally classified in reporting practices as underlying and contributing. Leveraging information from death certificates, these interactions can be naturally represented through a sequence of directed networks encoding co-occurrence strengths between pairs of underlying and contributing causes across ages. Although this perspective opens the avenues to learn informative age-specific block interactions among endogenous groups of underlying and contributing causes displaying similar co-occurrence patterns, there has been limited research along this direction in mortality modeling. This is mainly due to the lack of suitable stochastic block models for age-indexed sequences of directed networks. We cover this gap through a novel Bayesian formulation which crucially learns two separate group structures for underlying and contributing causes, while allowing both structures to change smoothly across ages via dependent random partition priors. As illustrated in simulations, this formulation outperforms state-of-the-art solutions that could be adapted to our motivating application. Moreover, when applied to USA mortality data, it unveils structures in the composition, evolution, and modular interactions among causes-of-death groups that were hidden to previous studies. Such findings could have relevant policy implications and contribute to an improved understanding of the recent "death of despair" phenomena in USA.

Problem

• Modeling age-varying interactions between underlying and contributing death causes
• Learning separate group structures for causes with dependent partitions
• Analyzing evolving co-occurrence patterns in cause-of-death networks

Innovation

• Bayesian model with separate group structures for causes
• Dependent random priors for smooth age-based transitions
• Learns evolving modular interactions in mortality networks