MS09 - MEPI-11

Advances in infectious disease modelling: towards a unifying framework to support the needs of small and large jurisdictions (Part 3)

Friday, July 18 at 4:00pm

SMB2025 Follow

Note: this minisymposia has multiple sessions. The other sessions are: Part-1, and Part-2.

Amy Hurford (Memorial University), Michael Li, Public Health Agency of Canada

Description:

Homogeneous mixing and the aggregation of diverse population groups into one group are frequent simplifying assumptions that may produce erroneous models and recommendations that exacerbate health inequities. Yet, models that make these simplifying assumptions have well-understood dynamics, and can be quickly solved, facilitating data fitting and uncertainty analysis that can support policy recommendations. Advancing the methodology around these model-building tensions is needed, and the best modelling approach may depend on the application. The motivation for developing these modelling approaches is from the COVID-19 pandemic in Canada. Atlantic Canada, the Canadian territories, and other small Canadian jurisdictions experienced different epidemiology, and needed different types of modelling support, than the larger Canadian provinces. There is a need to advance infectious disease modelling to support jurisdictions at all levels, and this session furthers this goal by including talks that describe: infectious disease spread in structured communities; importations and mobility networks; models that were developed for specific small jurisdictions, methods for calculating the reproduction number, estimating healthcare demand, and describing how the needs of small jurisdictions can be integrated into pandemic preparedness plans.

Wade McDonald

University of Saskatchewan

"Use of Synthetic Data to Improve Wastewater-based Epidemiological Models in a Small Jurisdiction"

Previous studies have shown that applying methods such as Particle Filtering and Particle Markov-Chain Monte Carlo (pMCMC) to stochastic mechanistic epidemiological models can enhance model accuracy compared to simple parameter calibration. Addition of data streams to the filter can improve model fit even if those data are deemed to be of “low quality,” e.g., internet search volumes. In the present work, we employ a synthetic dataset, generated by an agent-based model, to explore the use of pMCMC with a compartmental epidemiological model, including wastewater-based epidemiology (WBE), in the context of a small jurisdiction facing an emerging pathogen. Predictive performance of the filtered model will be compared against synthetic ground truth using clinical cases alone versus clinical cases and WBE measures. Effects of structural mismatches between the synthetic ground truth and filtered model will be considered; for example, what if the synthetic ground truth admits waning of immunity (SIRS) but our filtered model assumes immunity is permanent (SIR)?

Matthew Betti

Mount Allison University

"Modeling healthcare demand during a disease outbreak"

One of the driving concerns during any epidemic is the strain on the healthcare system. During severe outbreaks, healthcare systems can become quickly overwhelmed. We develop a healthcare demand module that can take epidemiological data and healthcare parameters and can forecast number of doctor visits, hospital occupancy. Using real-world data we can estimate the length of stay of hospitalized individuals. The module can be extended to account for pharmaceutical and PPE usage at differing levels of conservation.

Sicheng Zhao

McMaster University

"Edge-based Modeling for Disease Transmission on Random Graphs – an Application to Mitigate a Syphilis Outbreak"

Edge-based random network models, especially those based on bond percolation methods, can be used to model disease transmission on complex networks and accommodate social heterogeneity while keeping tractability. Here we present an application of an edge-based network model to the spread of syphilis in the Kingston, Frontenac and Lennox & Addington (KFL&A) region of Southeastern Ontario, Canada. We compared the results of using a network-based susceptible-infectious-recovered (SIR) model to those generated from using a traditional mass action SIR model. We found that the network model yields very different predictions, including a much lower estimate of the final epidemic size. We also used the network model to estimate the potential impact of introducing a rapid syphilis point of care test (POCT) and treatment intervention strategy that has recently been implemented by the public health unit to mitigate syphilis transmission.

Caroline Mburu

British Columbia Centre for Disease Control/Simon Fraser University

"Wastewater-based modelling for Mpox surveillance among gbMSM in BC"

Background: The 2022 global outbreak of Mpox, caused by Clade IIb of the monkeypox virus (MPXV), primarily affected gay, bisexual, and other men who have sex with men (gbMSM). While clinical case surveillance has been central to the public health response, it faces limitations due to underreporting, social stigma, and asymptomatic infections. To complement case-based surveillance, wastewater-based surveillance (WBS), which had been valuable in monitoring other infections, including during the COVID-19 pandemic, was adopted to track MPXV circulation. Several studies have demonstrated correlations between MPXV viral loads in wastewater and reported Mpox cases, supporting the utility of WBS for population-level monitoring. In parallel, mechanistic models based solely on clinical case data have provided insights into Mpox transmission dynamics and the impact of interventions such as vaccination and behavioral change. However, to date, no modeling framework has integrated both data streams to jointly infer Mpox transmission dynamics. As a result, the mechanistic relationship between viral load in wastewater and underlying disease transmission remains poorly understood, particularly in the context of evolving behavioral patterns and vaccination uptake Methods: We developed a compartmental model to simulate Mpox transmission within the gbMSM population, incorporating heterogeneity through stratification by levels of sexual activity. The model integrates key data streams, including clinical case notifications, MPXV viral load signals from WBS, sexual network data and vaccination coverage. The framework explicitly incorporates viral shedding dynamics into wastewater, allowing for the exploration of the relationship between underlying infections and observed WBS signals. We use this model to evaluate the conditions under which wastewater viral load may act as leading or lagging indicators of reported cases, considering factors such as reporting delays, underreporting, asymptomatic infections, changes in sexual behavior, and the rollout of vaccination programs. Conclusions: This study bridges clinical and environmental surveillance through a mechanistic framework tailored to behaviorally structured populations. By jointly modeling case and WBS data, we aim to improve the interpretation of wastewater signals and support more accurate assessments of transmission in hard-to-reach or underreported populations. Findings will inform public health decision-making around Mpox surveillance and preparedness, particularly in contexts where traditional case-based reporting is limited.

#SMB2025 Follow

Annual Meeting for the Society for Mathematical Biology, 2025.