In this study, we use ternary color-coding to visualize and compare the age structure of deaths from COVID-19 in Brazilian meso-regions using the tricolore package in R, in two different phases of the pandemic. The analysis of the age profile is important to better understand the dynamics of the pandemic, and how it has affected the population over age 25, according to age groups (25–59, 60–79 and >80 years) and subpopulations of the country. The analysis focuses on the first wave of the pandemic, until the end of 2020, and the more recent wave. Overall, the results suggest that when the two recent waves of the pandemic are compared, different spatial patterns in the distribution of deaths across the country by sex and by age emerge. While the distribution of deaths is found to be concentrated at older ages, we also observe in the more recent period some areas of the country with a concentration of deaths among younger adults. The analysis further indicates that even in areas with a younger population age structure, which could act as a protective factor against complications, the age pattern of mortality is very heterogeneous, and we do not find a clearly defined age and spatial pattern. Our results highlight the importance of looking at the distribution of COVID-19 mortality across small areas, and show that there are many different levels of the pandemic in Brazil at the same time, rather than just one.
