On the simultaneous inference of susceptibility distributions and intervention effects from epidemic curves

2025-10-26

AI Summary

This paper addresses the challenge of simultaneously inferring individual susceptibility heterogeneity and intervention effects from epidemic curves. We propose an identifiable joint inversion framework based on an extended SEIR model that incorporates individual-level variation in susceptibility and exposure. Using maximum likelihood estimation, we innovatively fit multiple epidemic datasets jointly to estimate shared parameters—thereby resolving the fundamental identifiability limitations inherent in conventional variable-susceptibility models. To our knowledge, this is the first rigorous validation of parameter inferability for such models. Our results demonstrate that, under realistic data quality conditions, the framework accurately recovers the true susceptibility distribution and key intervention-effect parameters. The method substantially improves the reliability of epidemic forecasting and the precision of public health policy evaluation. By establishing both theoretical guarantees and empirical evidence, it provides a foundation for integrating population heterogeneity into data-driven public health decision-making.

Technology Category

• Reasoning under Uncertainty: Relational Probabilistic Models
• Search and Optimization: Mixed Discrete/Continuous Search
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Susceptible-Exposed-Infectious-Recovered (SEIR) models with inter-individual variation in susceptibility or exposure to infection were proposed early in the COVID-19 pandemic as a potential element of the mathematical/statistical toolset available to policy development. In comparison with other models employed at the time, those designed to fully estimate the effects of such variation tended to predict small epidemic waves and hence require less containment to achieve the same outcomes. However, these models never made it to mainstream COVID-19 policy making due to lack of prior validation of their inference capabilities. Here we report the results of the first systematic investigation of this matter. We simulate datasets using the model with strategically chosen parameter values, and then conduct maximum likelihood estimation to assess how well we can retrieve the assumed parameter values. We identify some identifiability issues which can be overcome by creatively fitting multiple epidemics with shared parameters.

Problem

• Estimating individual susceptibility variation from epidemic curve data
• Assessing intervention effects using SEIR models with heterogeneity
• Overcoming identifiability issues in epidemic model parameter inference

Innovation

• Simulates datasets using strategically chosen parameter values
• Conducts maximum likelihood estimation for parameter retrieval
• Overcomes identifiability by fitting multiple epidemics with shared parameters