1. Development and validation of a dynamic nomogram for predicting in-hospital mortality in patients with gastrointestinal bleeding: a retrospective cohort study in the intensive care unit
- Author
-
Kang Zou, Shu Huang, Wensen Ren, Huan Xu, Zhiying Liu, Wei Zhang, Lei Shi, Xinxin Pu, Yinqin lv, Yan Peng, Fangfang Yuan, and Xiaowei Tang
- Subjects
Medicine ,Science - Abstract
Abstract The study aims to develop and validate a predictive model for effectively predicting in-hospital mortality in patients admitted to the intensive care unit due to Gastrointestinal bleeding (GIB). A retrospective cohort study was conducted, including data from patients in the Electronic Intensive Care Unit Collaborative Research Database (eICU-CRD) and Medical Information Mart for Intensive Care-IV Database (MIMIC-IV) with a diagnosis of GIB. Patients from the eICU-CRD were randomly allocated into both development and validation sets, and those from MIMIC-IV were assigned as an external validation group. Multivariate logistic regression was employed to create a predictive model, which was depicted as a nomogram. This study included a total of 2929 patients with GIB from the eICU-CRD and 718 patients from the MIMIC-IV. To access the dynamic nomogram, please use the following link: https://kangzou.shinyapps.io/DynNomapp_GIB/ . The area under the receiver operating characteristic curve for the nomogram was 0.893 in the development set, 0.860 in the internal validation set, and 0.781 in the external validation set. The mortality rate was 25.7% in the high-risk group (nomogram scores > 101.974) and 2.8% in the low-risk group (nomogram scores ≤ 101.974). The nomogram exhibited excellent discrimination, calibration, and clinical utility in predicting in-hospital mortality among patients admitted to the intensive care unit with GIB.
- Published
- 2024
- Full Text
- View/download PDF