MEPI Subgroup Contributed Talks

Lindsay Keegan

University of Utah

"A theoretical framework to quantify the tradeoff between individual and population benefits of expanded antibiotic use"

The use of antibiotics during a disease outbreak presents a critical tradeoff between immediate treatment benefits to the individual and the long-term risk to the population. Typically, the extensive use of antibiotics has been thought to increase selective pressures, leading to resistance. This study explores scenarios where expanded antibiotic treatment can be advantageous for both individual and population health. We develop a mathematical framework to assess the impacts on outbreak dynamics of choosing to treat moderate infections not treated under current guidelines, focusing on cholera as a case study. We derive conditions under which treating moderate infections can sufficiently decrease transmission and reduce the total number of antibiotic doses administered. We identify two critical thresholds: the Outbreak Prevention Threshold (OPT), where expanded treatment reduces the reproductive number below one and halts transmission, and the Dose Utilization Threshold (DUT), where expanded treatment results in fewer total antibiotic doses used than under current guidelines. For cholera, we find that treating moderate infections can feasibly stop an outbreak when the untreated reproductive number is less than 1.42 and will result in fewer does used compared to current guidelines when the untreated reproductive number is less than 1.53. These findings demonstrate that conditions exist under which expanding treatment to include moderate infections can reduce disease spread and the selective pressure for antibiotic resistance. These findings extend to other pathogens and outbreak scenarios, suggesting potential targets for optimized treatment strategies that balance public health benefits and antibiotic stewardship.

---

Youngsuk Ko

Yale University

"Effective Vaccination Strategies Against Dengue in Brazil: A Mathematical Modeling Approach Incorporating Spatial and Demographic Heterogeneities"

Brazil has experienced recurrent dengue outbreaks, with over 18 million reported cases since 2000 and a record-breaking surge in 2024. Notably, there has been a demographic shift in disease burden, with an increasing proportion of severe cases and fatalities among the elderly. Current vaccination strategies, including the WHO-recommended Qdenga® rollout for children, may not effectively address this emerging risk. This study employs a mathematical modeling approach to evaluate age-specific and geographically targeted vaccination strategies. A Susceptible-Infected-Recovered (SIR)-based model, calibrated using historical dengue data from Brazil's Notifiable Diseases Information System (SINAN), incorporates spatial heterogeneity across 27 states and demographic factors such as prior exposure and birth rates. We assess the impact of different vaccination strategies by estimating averted infections, hospitalizations, fatalities, and years of life lost. Preliminary findings indicate significant variation in the force of infection across states and suggest that prioritizing vaccination for elderly populations may substantially reduce severe disease burden. This modeling framework provides a quantitative basis for optimizing vaccination policies, with potential applications to other arboviral diseases and endemic settings worldwide.

---

Francisca Olajide

University of Ottawa

"From process to structure of EWSs"

The emergence of infectious diseases remains a huge challenge to public health. Early detection of outbreaks using early warning signals (EWSs) offers an invaluable opportunity for effective preparedness and disease management. In this study, we seek to understand the structure of these signals using a mechanistic model that captures epidemic and social processes. We analyzed the simulated time series for change points and EWSs (autocorrelation and variance). All time series showed the expected delay in that the detected change point occurred significantly after the parameter passed the bifurcation point. These early warning signals exhibited a stronger response after the threshold for disease emergence had been exceeded. Assessing different disease progression and intervention models will help determine the most effective signals for use in public-health settings.

---

Marwa Tuffaha

York University

"Counterfactual COVID-19: Modeling Alternative Mitigation and Vaccination Policies for Canada"

COVID-19, a global pandemic with severe health and economic repercussions, has prompted various approaches to mitigate its impact. We adapt an age-structured SEIVS model—incorporating waning immunity and partial protection—to explore counterfactual scenarios of non-pharmaceutical interventions (e.g., school/workplace closures, distancing) and selected vaccine policy changes in Canada. By altering contact patterns and compliance levels, we assess potential outcomes under stricter, earlier, or more relaxed mitigation measures, with a lesser emphasis on shifting vaccination rollouts. Findings indicate that timely, robust mitigation can substantially reduce severe disease and delay epidemic peaks, whereas delayed or minimal interventions lead to higher case burden. Integrating vaccine strategies into these scenarios further highlights the interplay between pharmaceutical and non-pharmaceutical measures, showcasing how modeling can inform dynamic policy-making for ongoing and future public health crises.

---