Chinese and Italian COVID-19 outbreaks can be correctly described by a modified SIRD model

Diego Caccavo

The understanding of the dynamic of the epidemic could be useful to take decisions in time, preparing hospitals to deal with the infection peak. The ongoing outbreaks are often described with simple fitting models, i.e. exponential functions (and/or sigma-shaped curves such as Logistic or Gompertz), whereas in this work, a SIRD compartmental model was developed and used to describe and predict the evolution of the Chinese (Hubei) and of the Italian outbreaks. The advantages of SIRD models with respect to heuristic models reside mainly in the physical-based SIRD parameters, which can give useful insight and/or can be properly chosen on the basis of physical-based assumptions. The present SIRD model was able to describe the outbreaks, obtaining the kinetic of infection, the kinetic of recovery as well as the kinetic of death. The Italian best fit initial kinetic of infection was lower (half) than the Chinese one, probably because the Italian outbreaks are mainly based in small hospitals/cities, whereas the main Chinese outbreak was in the metropolitan city of Wuhan. However, the decrease time of this kinetic in the Italian outbreak was slower than in the Chinese outbreak, probably due to the delayed lockdown and initial weaker restrictions. Therefore, despite a lower average â„› ! , 3.9 in Italy versus 8.3 in Hubei, the longer Italian epidemic produced much more COVID-19 cases.