A clinical case-control study was conducted to screen the influencing factors of patients with coronavirus disease 2019 (COVID-19) and to construct a clinical prediction model to provide a reference for the dynamic assessment of the severity of COVID-19 patients. A total of 410 patients with COVID-19 were included in the study, of which 132 were severe or critical cases. The clinical data of patients were collected, and then variables were screened by lasso regression analysis and 10-fold cross-validation. The screened variables were subjected to multifactorial logistic regression analysis to screen out the independent risk factors of patients with severe or critical illnesses, and the independent risk factors were integrated to construct a nomogram. The receiver operating characteristic curve (ROC), calibration curve analysis, and decision curve analysis (DCA) were used to assess the model efficiency. Five variables, including respiratory rate（R）, systolic blood pressure (SBP), plasma albumin (ALB), lactate dehydrogenase (LDH), and C-reactive protein (CRP), were finally included to construct a clinical prediction model, with an area under the curve (AUC) of 0.86 (CI: 0.82% to 0.90%).