Model development, data, and public health: A combined approach against malaria

The health burden of infectious diseases remains substantial. Within ever-changing public health landscapes, characterised by complex disease biologies and limited operational resources, appropriate control strategies are often difficult to identify. In recent years, there has been an explosive increase in the popularity of mathematical modelling to bridge evidence gaps and support (policy) decision-making. To ensure accurate predictions, e.g. on the public health impact of new interventions, models must be grounded in plausible assumptions and calibrated to diverse data on multiple epidemiological and biological relationships.
Through a comprehensive investigation of the modelling process from development to application, my research aims to prompt discussions about the role of infectious disease modelling in decision-making and about opportunities for modernisation. With application to malaria modelling, I present methodological advancements, structural analyses and discussions, and application case studies. This includes the development of a novel, machine learning-based calibration approach that outperforms previous methods. A generalisable framework for incorporating calibration data while accounting for contextual covariates is developed and applied to a database of PfPR-incidence records. I subsequently discuss the link between model calibration decisions and the model's later uses in simulating epidemiological relationships. Taking the leap from model development to application, I assess the use of surveillance-response interventions for malaria elimination, addressing the various technical challenges of quantifying elimination itself. Finally, I shift perspective towards the potentials and pitfalls of using modelling to support decision-making.
The research presented in this thesis contributes to keeping malaria modelling up-to-date with computational methods and global health developments. Many of the principles presented here encompass general discussions of infectious disease modelling, and aim to encourage conversations about the place of modelling at the public health decision-making table.