Analysis of Factors Associated with a High Probability of Fatal Case in Patients with New Coronavirus Infection (COVID-19) Treated in Hospital

Introduction. Due to the severity of the state and course of the new coronavirus infection (COVID-19), patients requiring hospitalization were the group at high risk of death during the spread of infection. Their condition, treatment tactics and outcomes were associated with the presence of comorbidities, gender, age and duration of hospitalization. Studying the relationship of these factors with mortality among patients with COVID-19 is essential for effective organization of medical care.Aim: To analyze factors associated with a high probability of mortality in patients with COVID-19 treated in hospitals from 2020 to 2021.Material and Methods. 25,028 depersonalized records of patients receiving inpatient medical care in medical organizations of Tomsk region from 2020 to 2021 were studied. The presence of comorbidities, gender, patient age, and number of days of hospitalization were investigated as potential factors associated with a high probability of lethal outcome in patients with a diagnosis of COVID-19. The relationship of potential factors associated with a high probability of mortality according to the period of epidemic rise in COVID-19 incidence was evaluated using classification models.Results. An analysis of data from patients hospitalized with COVID-19 in Tomsk Oblast hospitals from 2020 to 2021 revealed that the patient's age, length of stay in hospital, and presence of concomitant pathologies are associated with the probability of mortality. Based on the results of the analysis, a profile of a patient hospitalized in a Tomsk region hospital with confirmed coronavirus infection with high risks of lethal outcome was formed: a man aged 70 years or older, who has been hospitalized for 10 days or more and has one or more comorbidities, in particular, diseases of the heart, blood vessels or endocrine system. The models obtained during the study are not suitable for predicting the outcome of the disease in the context of the spread of new COVID-19 strains and require changes in the composition of predictors of prognostic models.Conclusion. The presented algorithm for analyzing risk factors for lethal outcome in patients with a coronavirus infection can be used in other regions with possible identification of new risk factors and associations with the predominant strain.