Addressing Public Health Challenges Through Epidemiological Models

The recent COVID-19 pandemic underscored the importance of mathematical modeling in fighting the growing challenge represented by infectious diseases. Mathematical models help shed light on the mechanisms of transmission, allow the retrospective evaluation of the effectiveness of interventions, and guide decision-making by exploring potential scenarios. This thesis presents a series of studies utilizing tailored mathematical approaches to address key public health challenges. The first part examines the epidemiology of SARS-CoV-2 variants of concern (VOC) during the COVID-19 pandemic, focusing on the intrinsic generation time of Alpha, Delta, and Omicron variants—a crucial metric for designing effective isolation, quarantine, and surveillance protocols. Using Bayesian inference applied to extensive contact tracing data, the study reveals no statistically significant differences among the variants, with generation time estimates ranging from 6.0 to 6.8 days. The second part models norovirus transmission aboard cruise ships, where the confined environment amplifies outbreak risks. A Bayesian model reconstructed transmission chains from line-list data, uncovering a high transmission heterogeneity: 57% (95% CrI: 48%–65%) of secondary cases were caused by 10% of infected individuals who experienced longer diagnostic delays. A branching process model estimated that the isolation protocol implemented during the outbreak prevented 71% of potential cases compared to a no-intervention scenario, with further reductions achieved by minimizing diagnostic delays. The final section evaluates the population-level impact of introducing potential HIV cures among men who have sex with men (MSM) in the Netherlands. Using an ODE model calibrated with national behavioral and epidemiological data, the study finds that a cure providing temporary viral suppression but prone to failure could disrupt the current declining trend in HIV incidence. Conversely, a cure achieving complete viral elimination could consistently reduce cases, with cumulative incidence declining by up to 50% over the next decade, even when reinfection is possible.