Laboratory Testing Implications of Risk-Stratification and Management of COVID-19 Patients
Abstract The progression from mild to critical illness is the main reason leading to the death of COVID-19 patients. Rapid risk-stratification at admission is important for precise management of COVID-19. Here, we developed a practical admission stratification model to predict the severity during hospitalization of COVID-19 patients using laboratory data from 3563 patients, including 548 patients in the training dataset, and 3015 patients in the testing dataset. We first identified the significant laboratory indicators related to the severity of COVID-19 in the training dataset. Neutrophils percentage (NEUT%), lymphocytes percentage (LYMPH%), creatinine (CREA), and blood urea nitrogen (BUN) with AUC greater than 0.7 were included in the model. These indicators were further used to build a support vector machine model to classify patients into low-risk and high-risk at admission. Results showed that this model could stratify the patients in the testing dataset effectively (AUC=0.89). Moreover, laboratory indicators detected in the first week after admission were able to estimate the probability of death (AUC=0.95). Besides, we could diagnose COVID-19 and differentiated it from other kinds of viral pneumonia based on laboratory indicators (accuracy=0.97). Our risk-stratification model based on laboratory indicators could help to diagnose, monitor, and predict severity at an early stage of COVID-19.