Data-driven healthcare models for pandemic preparation
PhD candidate: Victor PloegÂ
Pandemics put significant pressure on healthcare providers and thereby affect regular care provision. The COVID-19 pandemic highlighted the need for accurate short-term bed occupancy forecasts, more coordinated regional patient allocation, and insight into the effects of pandemics on capacity-related questions throughout the care chain to maintain care for both pandemic and non-pandemic patients.
For this reason, I contribute to the development of a dynamic patient allocation rule to assign pandemic patients to hospitals in a network to balance the load of regular patient backlogs, and I analyze the long-term behavior of a system under the influence of such an allocation rule using fluid scaling techniques, which enable representing the patient-level dynamics as a deterministic continuous model of differential equations.
To facilitate usability in practice of mathematical models developed for pandemic preparation, I collaborate closely with hospitals and the body responsible for the national coordination of patient allocation on translating existing models into integrated support systems while investigating workable structures for sharing data safely. This work combines short-term hospital-level bed occupancy forecasts with optimization models to support patient distribution under demand uncertainty.
Besides these modeling efforts, I research the impact the COVID-19 pandemic has had on other parts of the Dutch healthcare system. This work, for which I collaborate, for instance, with a general practitioner, aims to shed light on care usage and capacity-related questions that might feed later modeling efforts and decision-making in pandemic times.
