Causal Inference for Epidemic Models

2024-10-15

AI Summary

This paper addresses the longstanding conflation of statistical descriptive interpretations with causal counterfactual interpretations in epidemiological modeling. It is the first to systematically distinguish the data-generating mechanism (DGM) from the structural causal model (SCM), thereby rigorously defining the true causal effects of interventions—such as vaccination or social distancing—on disease transmission dynamics. Methodologically, it proposes a parameter identifiability framework grounded in do-calculus and SCM, integrating causal graph modeling, maximum likelihood estimation, and sensitivity analysis into a verifiable causal inference pipeline. Theoretically, it establishes consistency conditions for intervention effect estimation. Empirically, the framework demonstrates significantly improved unbiasedness and robustness in estimating causal effects on both synthetic and real-world epidemic datasets. This work advances epidemiological causal modeling by providing a principled, transparent, and empirically validated paradigm for causal inference in infectious disease dynamics.

Technology Category

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

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Epidemic models describe the evolution of a communicable disease over time. These models are often modified to include the effects of interventions (control measures) such as vaccination, social distancing, school closings etc. Many such models were proposed during the COVID-19 epidemic. Inevitably these models are used to answer the question: What is the effect of the intervention on the epidemic? These models can either be interpreted as data generating models describing observed random variables or as causal models for counterfactual random variables. These two interpretations are often conflated in the literature. We discuss the difference between these two types of models, and then we discuss how to estimate the parameters of the model.

Problem

• Distinguish causal vs data-generating epidemic models
• Estimate intervention effects on disease spread
• Clarify model interpretations for COVID-19 analysis

Innovation

• Causal models for counterfactual epidemic variables
• Differentiating data generating vs causal interpretations
• Parameter estimation methods for epidemic models