Predictions of a seir model for COVID-19 cases in Cali-Colombia

Abstract

Objective To predict the number of cases of COVID-19 in the city of Cali-Colombia through the development of a SEIR model. Methods A SEIR compartmental deterministic model was used considering the states: susceptible (S), exposed (E), infected (I) and recovered (R). The model parameters were selected according to the literature review, in the case of the case fatality rate data from the Municipal Secretary of Health were used. Several scenarios were considered taking into account variations in the basic number of reproduction (R0), and the predic-tion until april 9 was compared with the observed data. Results Through the SEIR model it was found that with the highest basic number of reproduction [2,6] and using the case fatality rate for the city of 2,0%, the maximum number of cases would be reached on June 1 with 195 666 (prevalence). However, when comparing the observed with the expected cases, at the beginning the observed occurrence was above the projected, but then the trend changes decreasing the slope. Conclusions SEIR epidemiological models are widely used methods for projecting cases in infectious diseases, however it must be taken into account that they are deterministic models that can use assumed parameters and could generate imprecise results.