Related Experiment Video
Updated: Jun 4, 2025
04:57
Establishing a Competing Risk Regression Nomogram Model for Survival Data
Published on: October 23, 2020
10.1K
Development and validation of a prognostic nomogram model for severe osteomyelitis patients
Yunlong Liu1, Yan Zheng2, Sheng Ding3
1Department of Pediatric Surgery, Women and Children's Hospital Affiliated to Ningbo University, No. 339 Liuting Street, Ningbo, 315012, Zhejiang Province, China. 13957403126@163.com.
Scientific Reports
|January 3, 2025
Summary
Predicting one-year mortality for intensive care unit (ICU) osteomyelitis patients is crucial. A new nomogram model effectively identifies high-risk individuals, aiding clinical decisions and improving patient outcomes.
Area of Science:
- Critical Care Medicine
- Infectious Diseases
- Medical Informatics
Background:
- Severe osteomyelitis infections in the Intensive Care Unit (ICU) are associated with increased mortality risk.
- Prognostic prediction for these critically ill patients remains underexplored.
- Effective tools are needed to guide clinical management and treatment strategies.
Purpose of the Study:
- To develop and validate a predictive model for one-year mortality risk in ICU-admitted osteomyelitis patients.
- To identify independent predictors of mortality in this specific patient cohort.
- To provide a tool for informed clinical decision-making.
Main Methods:
- Utilized the MIMIC IV database to extract data from 1153 osteomyelitis patients admitted to the ICU.
- Randomly split data into training (70%) and validation (30%) sets.
- Developed a risk prediction nomogram using logistic regression and assessed its performance with C-indexes, ROC curves, DCA, CIC, and calibration curves.
Main Results:
- Identified key independent predictors of one-year mortality in ICU osteomyelitis patients.
- The developed nomogram demonstrated strong predictive performance with AUCs of 0.872 (training) and 0.826 (validation).
- Calibration curves, ROC curves, DCA, and CIC confirmed the model's excellent predictive accuracy and clinical utility.
Conclusions:
- A validated nomogram model for predicting one-year mortality in ICU osteomyelitis patients has been successfully developed.
- This model offers valuable predictive information to support clinical diagnosis and treatment planning.
- The tool shows robust predictive efficiency and clinical effectiveness for high-risk patients.
