Identifying factors related to COVID-19 mortality is important to deploy effective containment measures and to safeguard categories at risk. In the last months, several investigations have tried to ascertain essential features for predicting the COVID-19 mortality tolls depending on country-specific dynamics and population structure. Most studies focused on the initial outbreak of COVID-19 spanning the first half of 2020. Several variables, including obesity, health system indicators such as hospital bed density, and bacillus Calmette-Guerin vaccination have been reported as significantly associated to COVID-19 mortality. Here, we examined in different pandemic stages some of the mentioned associations as well as ABO and Rh blood group indicators, which have also been previously linked to COVID-19 severity and fatal outcome. Using a machine learning approach, we found that the “B+” blood group frequency is an important factor at all stages of the pandemic.