ABSTRACT
Risk stratification for adults infected with SARS-CoV-2 COVID-19 continues to be essential to inform decisions about individual patients and allocation of resources and treatment options. Accurate knowledge of individual risk of severe COVID-19 can make an important contribution to healthcare both on a population and a personal level. There are currently few tools and solutions that help medical professionals to predict the evolution of SARS-COV-2 infected patients. So far, risk models for severe COVID-19 outcomes have included age and clinical comorbidities. The first wave of the COVID-19 pandemic spread rapidly in Spain, one of Europe’s most affected countries. In this retrospective study we analyzed genotypic and phenotypic data from 659 patients in the Basque region of Spain during the first wave of COVID-19, and compared mild with severe COVID-19 cases. Using genetic variants data as well as clinical variables of the participants we built a prediction model of severe COVID-19. We obtained robust results in the training data set with 85% sensitivity, 67% specificity and an Area Under the Curve (AUC) of 0.78. In the validation set the AUC was 0.75. The main advantage of our model is that because it includes genetic variants it could be used with medical records to identify the critical population in advance.