Infection Units: A novel approach to the modeling of COVID-19 spread

Abstract

A novel mechanistic model describing the rate of COVID-19 spread is presented, that differs conceptually from previously published deterministic models. One of its main characteristics is that the pool of infected people is not assumed to be homogeneously mixed, but rather as a passage into which individuals enter upon contagion, move within it in a plug-flow manner and leave at recovery, within a fixed time period. So, the present model differs conceptually in the way it describes the dynamics of infection. An ‘infection unit’ is defined as the amount of COVID-19 virus that generates contagion, if it reaches a susceptible individual. This model separately considers various pools: symptomatic and asymptomatic infected patients; three different pools of recovered individuals; pools of assisted, hospitalized patients; the quarantined and, finally, those who died from COVID-19. The transmission of the disease from an infected person to others is described by an infection rate function, while an encounter frequency function modulates the frequency of effective encounters between the infected and the susceptible. The influence of the model’s parameters on the predicted results is presented. The effect of social restrictions and of quarantine policy on pandemic spread is shown. For model calibration, a set of experimental data is used. The model enables the calculation of the actual behaviour of the studied pools during pandemic spread.