The Unknown of the Pandemic: An Agent-based Model of Final Phase Risks

With the COVID-19 pandemic progressing after the peak, public health authorities are facing the difficult decision of how to manage the final phase of the epidemic wave. Diffuse uncertainty still persists about the characteristics of the pandemic and its actual dynamics, in particular with regard to the real extent of undiagnosed infected cases. We present a discrete-time stochastic model with state-dependent transmission probabilities and multi-agent simulations focusing on possible risks that could materialize in the final phase of the epidemic. The results of our experiments show that, in different scenarios, there is the possibility that unknown undiagnosed cases still looms when diagnosed infected cases are close to be extinguished. We study variants of base scenarios to account for uncertain epidemiological estimates and the effects of testing and containment of cases otherwise undiagnosed. A trade-off between measures producing a slower or accelerated dynamics is discussed. Ultimately, the analysis we have presented highlights that the enduring uncertainty, characterizing the current pandemic, calls for risk analyses approaches to complement epidemiology studies.