Abstract: Manaus, Brazil, experienced two severe waves of COVID-19 deaths due to two different strains of SARS-CoV-2. Since most individuals were infected during the first wave, and the second strain nevertheless increased the mortality rates, there seems to be little cross-immunity between the two strains. Brazil has excellent public data bases providing the number of deaths due to COVID-19, and the fraction of the population having antibodies. We model this data to study why the second wave was smaller but deadlier using MCMC methods. We find that cross-immunity indeed provides little protection to reinfection, but provides excellent protection against severe disease. Hence most of the deaths during the second wave are due to primary infections caused by a large wave of reinfected individuals that are themselves well protected. We demonstrate that the first wave elicited such a form of herd immunity by artificially blocking the first wave, and observing that this intervention markedly increases the death rate during the second wave.
Research Collaborators: Robert Jan Paul van Meenen, Chris van Dorp & Michiel van Boven