Abstract : Objective To explore the risk factors afecting severe pneumonia in children, construct a nomogram model, and evaluate and verify the feasibility of the model. Methods From January 2020 to January 2022, the clinical data of 300 hospitalized children diagnosed with bronchopneumonia ( 188 cases) and severe pneumonia ( 112 cases) in the First Hospital of Lanzhou University were used as the modeling group, including the mild pneumonia group and severe pneumonia group. From March 2022 to April 2022 and March 2023 to April 2023, the clinical data of 123 hospitalized children diagnosed with bronchopneumonia (56 cases) and severe pneumonia (67 cases) in the First Hospital of Lanzhou University were used as the validation group. The modeling group was the same inclusion and exclusion criteria as the validation group. The clinical data of the patients were ollected, and the information of the modeling group was compared with the validation group. The multi -factor logistic regression analysis was used to screen the risk factors afcted severe pneumonia. The R sofware was used to build a nomogram model predicting severe pneumonia in children. The calibration and distinction of the nomogram model were evaluated, and the simplified application of the model was calculated. Results The ICU hospitalization history, length of stay, C-reactive protein (CRP), wheeze ，percentage of monocyte and cardiovascular complications were independent risk factors affecting severe pneumonia in children ( P<0.05). The nomogram prediction model constructed by independent risk factors was performed repeated sampling for 1 000 times using the Bootstrap method, and the results showed that the calibration curve of the nomogram model was closed to the ideal curve, C-index =0.925. The Hosmer-Lemeshow good of fit test showedx2= 11.060, P=0.198，the area under the ROC curve( AUCROO )= 0.925, and 95%CI: 0.895-0.955. The evaluation of decision curve analysis ( DCA)showed a good clinical benefit rate, and the simplified application and validation of the model were reconstructed using the R software，with high consistency between the validation results and the actual situation. Conclusion The nomogram model constructed by independent risk factors, including the ICU hospitalization history, length of stay, CRP，wheeze, percentage of monocyte, and cardiovascular complications of severe pneumonia in children has good predictive value and can provide reference for the prevention and treat-ment of severe pneumonia in children.