Assessing the determinants of COVID-19 burden to address disease-control policy decisions

Mathematical modeling has been crucial to address fundamental issues related to COVID-19 disease-control policy decisions. This thesis deals with the robust quantification of the disease burden associated with COVID-19 across different socio-demographic settings. The presented work includes the statistical analysis of novel epidemiological records to provide solid estimates describing the clinical course of SARS-CoV-2 infections and the simulation of data-driven models to forecast the potential impact of COVID-19 in rural and urban areas of Ethiopia. Obtained estimates show that being older than 60 years of age is associated with about 40% likelihood of developing symptoms after SARS-CoV-2 infection and 1% risk of requiring intensive care. The analysis of potential SARS-CoV-2 transmission in Ethiopia suggests that the low prevalence and mortality observed during 2020 can be explained by combined effect of younger demography and a reduced transmission generated by school closures implemented in response to the pandemic. Provided estimates highlight that in this country, after the launch of vaccination in 2021, the highest fraction of severe cases is expected to arise from the interaction between children (who are the main responsible for the spread of the disease) with the elderly (representing the most vulnerable population segment). Remarkably, prioritizing the vaccination of the elderly emerged as the best strategy to reduce the number of critical patients, irrespectively to the limited number of doses made available to low-income settings.