Modeling the COVID-19 dissemination in the South Region of Brazil and testing gradual mitigation strategies

This study aims to understand the features of the COVID-19 spread in the South Region of Brazil by estimating the Effective Reproduction Number (ERN) ℛ e for the states of Paraná (PR), Rio Grande do Sul (RS), and Santa Catarina (SC). We used the SIRD (Susceptibles-Infectious-Recovered-Dead) model to describe the past data and to simulate strategies for the gradual mitigation of the epidemic curve by applying non-pharmacological measures. Besides the SIRD model does not include some aspects of COVID-19, as the symptomatic and asymptomatic subgroups of individuals and the incubation period, for example, in this work we intend to use a classical and easy to handle model to introduce a thorough method of adjustment that allows us to achieve reliable fitting for the real data and to obtain insights about the current trends for the pandemic in each locality. Our results demonstrate that for localities for which the ERN is about 2, only rigid measures are efficient to avoid overwhelming the health care system. These findings corroborate the relevance of keeping the value of ℛ e below 1 and applying containment measures early.