A modified SEIR model to predict the COVID-19 outbreak in Spain: simulating control scenarios and multi-scale epidemics

Leonardo R Lopez; Xavier Rodo

In order to see the performance of applying the same model framework in the other European country with more quarantine cases to date, we fitted the model to the COVID-19 epidemic in Italy. This way, as the situation in Italy precedes that of Spain by around one week, evaluation of control scenarios can be done at a more advanced stage of confinement measures. As the fist cases in Italy were reported in the end of February, two different sets of fittings were performed in order to compare the confinement restrictions imposed by the Italian authorities on March 10th. The NMSE in this case was 2.7e − 3 + − 3.6e − 4 (ci = 95%) and the parameters for each case are summarized in Table II . Figure 9 displays model results for the two different scenarios. Bottom panel shows the effect of different measures of confinement since day 1 in an hypothetical control scenario. These dynamics are the result of increasing the fitted α until approx 1% daily. This is a significant increase having into account the initial value of protection rate.