Development of a Nomogram for Predicting Mortality Risk in Sepsis Patients During Hospitalization: A Retrospective Study

• DOI: 10.2147/idr.s407202
• Publication Date: 2023-04-18T23:55:06Z
• Authors: Bin Lu; Xinling Pan; Bin Wang; Chenyuan Jin; Chenxin Liu; Mengqi Wang; Yunzhen Shi
• Keywords: Nomogram; Stepwise regression

We attempted to establish a model for predicting the mortality risk of sepsis patients during hospitalization.Data on with were collected from clinical record mining database, who hospitalized at Affiliated Dongyang Hospital Wenzhou Medical University between January 2013 and August 2022. These included divided into modeling validation groups. In group, independent factors death hospitalization determined using univariate multi-variate regression analyses. After stepwise analysis (both directions), nomogram was drawn. The discrimination ability evaluated area under curve (AUC) receiver operating characteristic (ROC) curve, GiViTI calibration chart assessed calibration. Decline Curve Analysis (DCA) performed evaluate effectiveness prediction model. Among logistic compared models established by SOFA scoring system, random forest method, stacking method.A total 1740 subjects in this study, 1218 population 522 population. results revealed that serum cholinesterase, bilirubin, respiratory failure, lactic acid, creatinine, pro-brain natriuretic peptide death. AUC values group 0.847 0.826. P charts two sets 0.838 0.771. DCA curves above extreme curves. Moreover, method 0.777, 0.827, 0.832, respectively.The combining multiple could effectively predict hospitalization.