Integrating Modeling Methods for Quantifying Uncertainty in Infectious Disease Models

Mathematical and computational models play an important role in improving our understanding and our ability to predict the spread of infectious diseases at both the within host and population levels. These models inform policy decisions, help evaluate intervention strategies, and can guide public health preparedness. Unfortunately, model predictions are inherently uncertain due to various factors, including data limitations, assumptions about model structure, uncertainty in parameter estimates, and stochastic effects, so it is critical to develop methods for quantifying uncertainty.

Current practices for assessing uncertainty in model predictions vary widely across disciplines and modeling approaches. As a result, uncertainty is often under-characterized or inconsistently reported and interpreted, reducing confidence in model outputs and potentially leading to impaired decision-making. There is thus a need to develop methods that not only quantify uncertainty more rigorously but also facilitate comparisons across model types and effectively communicate uncertainty to stakeholders. This Research Topic seeks to address these challenges by promoting collaboration across modeling communities in advancing the development of unified approaches to quantifying uncertainty in models of infectious diseases.

The central goal of this Research Topic is to foster the integration of modeling methodologies to improve the quantification and interpretation of uncertainty in infectious disease models. While progress has been made within different modeling paradigms, such as statistical inference, agent-based simulations, and compartmental models, a lack of standardized practices for evaluating and comparing uncertainty across different model structures remains. This compartmentalization hinders reproducibility, limits the generalizability of insights, and reduces public trust in model predictions. This Research Topic will highlight research that addresses current limitations by combining or comparing methods, developing robust uncertainty quantification frameworks, and designing tools for uncertainty visualization and interpretation. This includes not only new methodologies but also practical applications to real-world infectious disease scenarios, both within-host and at the population level. We encourage submissions from a variety of disciplines and modeling frameworks, in the hopes of bridging the gaps, ultimately contributing to more reliable and transparent model predictions.

We welcome submissions that explore methodological, theoretical, and applied aspects of uncertainty quantification in infectious disease modeling. Topics of interest include, but are not limited to:

- Bayesian and probabilistic approaches to parameter inference for infectious disease modeling
- Sensitivity, identifiability analysis, and uncertainty propagation
- Ensemble modeling techniques
- Stochastic versus deterministic model uncertainty
- Integration of mechanistic and data-driven models
- Real-time uncertainty quantification in forecasting
- Uncertainty due to data quality, reporting delays, and missing information
- Visualization and communication of uncertainty to non-technical audiences
- Case studies involving COVID-19, influenza, vector-borne disease, etc

We encourage original research articles, methodological papers, and reviews that demonstrate the implementation or comparison of uncertainty quantification techniques across different modeling frameworks. This Research Topic will address a timely and critical need to rigorously integrate uncertainty quantification methods into infectious disease modeling. It will serve as a resource and reference point for both theoretical advances and practical implementation in public health modeling.