Fig. 2

Comprehensive prediction models for death risk of COVID-19 patients. Time-dependent receiver operating characteristic (ROC) curves and area under the curve (AUC) were employed to assess the predictive accuracy of models evaluating the death risk of COVID-19 with SOFA, qSOFA, APACHE II and SIRS scores, inflammatory-related indexes, complications, organ damage indexes, immune cell subsets and combined group integrating abovementioned these factors. The multivariate Cox proportional hazards model analysis were used to establish a risk model. The stepwise regression was used for the selection of the prediction for the model. ROC curves and AUCs (95% CIs) values were generated to assess prognostic accuracy for each model. A two-sided P value < 0.05 was considered statistically significant