Microsimulation based quantitative analysis of COVID-19 management strategies January 13, 2022
The hottest issue of our time perhaps is the COVID-19 epidemic and the effectiveness of strategies to combat it. The study, a joint work of our faculty members and several PhD students, has been published in PLOS Computational Biology, one of the most prestigious open access journals in the US. The article of fourteen authors presents a model that is used to inform and support Hungarian disease management decisions. PanSim, described in the paper, is a city simulator that mimics and models the daily movements, habits, i.e. life and social events of virtual inhabitants (agents) of a city with the size and population of the Hungarian city Szeged. The authors describe in detail how the impact of epidemiological decisions and interventions could be compared (and justified) using this strategic tool.
Decision-makers implement various non-pharmaceutical interventions to mitigate the COVID-19 pandemic. These include the closure of social events, restaurants, the introduction of curfews, elevated testing and quarantining of infected people. Once vaccines became available, decisions had to be made about the vaccination order, i.e. whom to vaccinate first. As the pandemic started to slow down, new decisions were needed on when it is safe to lift restrictions and which of them should be kept for longer. The model has helped us to recognise, for example, why increased testing, on its own, is not a sufficient tool to stop an epidemic.
By now, it has become clear, that as new virus variants bring new waves, we need to learn from previous examples and need better evaluation systems to find locally optimal interventions. Several modelling platforms have been developed to support specific decisions, but there is a clear need for a unified platform that can simulate the combined effects and potential interactions of all simultaneous interventions with fine-grained time and space resolution. This is where PanSim, a virtual city simulator described in this paper, could and can help.
(Read full article here)