Objective To construct a risk prediction model for neonatal umbilical vein catheterization (UVC) bloodstream infection (BSI) in neonates in a neonatal intensive care unit (NICU), and validate the application effect. Methods Neonates underwent UVC in NICU of a hospital from July 2020 to Deciber 2021 were selected retrospectively as the modeling group and divided into the infection group and non-infection group according to whether UVC BSI occurred. Bacteria isolated from blood microbial culture of neonates in infection group was statistically analyzed. Risk factors for BSI were screened by univariate and multivariate logistic regression analysis. R language was used to construct a nomograph model to predict the risk. In addition, neonates underwent UVC from January to July 2022 were selected as the validation group for external validation of the model. Results From July 2020 to Deciber 2021, 447 neonates were selected as modeling group, including 34 in infection group and 413 in non-infection group, with an infection rate of 7.6%. 225 neonates from January to July 2022 were as the validation group. Logistic regression analysis on modeling group showed that the total catheterization time >7 days, birth weight < 1500 g, puncture opportunities >2 times, albumin < 35 g/L, history of mechanical ventilation and peripherally inserted central catheter (PICC) treatment were independent risk factors for UVC BSI in NICU neonates. Nomograph prediction model was constructed based on the regression analysis results. The area under the receiver operating characteristic (ROC) curve of subjects in the modeling group was 0.866 (95%CI: 0.784-0.947), the Youden index, sensitivity, and specificity were 0.642, 0.853, and 0.789 respectively. Hosmer-Lieshow test showed P=0.323. The area under the ROC curve of the validation group was 0.837 (95%CI: 0.744-0.930). The Youden index, sensitivity, and the specificity were 0.549, 0.700, and 0.849 respectively, which suggested that the model has good discrimination and degree of fitting. Conclusion The model constructed in this study can well predict the risk of UVC BSI in NICU neonates, thus can be used as an evaluation tool for clinical medical staff to predict the risk of UVC BSI in neonates.