Related Experiment Video
Updated: Dec 29, 2025
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
Establishment and validation of a delirium prediction model for neurosurgery patients in intensive care
Jun Wang1, Yuanyuan Ji1, Ning Wang1
1Department of Neurosurgery, Xuanwu Hospital, Capital Medical University, Beijing, China.
Background:
Neurosurgical intensive care unit patients are at high risk for delirium. A risk prediction model could help the staff screen for patients at high risk for delirium. On the basis of this risk, preventive measures could be taken to reduce the undesired effects of delirium.
Objectives:
To establish a delirium prediction model for neurosurgical intensive care unit patients and to verify the sensitivity and specificity of this model.
Design:
A prospective, observational, single-centre study.
Methods:
Data were collected from a total of 310 patients admitted to the neurosurgery intensive care unit between January 2017 and February 2018. A risk factor prediction model was then created using multivariate logistic regression. Further data were collected from another 60 patients between March 2018 and June 2018 to validate the model.
Results:
The model consisted of six predictors, namely, cognitive dysfunction on admission, fever, hypoalbuminaemia, abnormal liver function, sedative use four or more times, and physical restraint. The area under the curve of the model was 0.80, with sensitivity and specificity of 0.68 and 0.83, respectively.
Conclusions:
This study established a delirium prediction model for neurosurgical intensive care unit patients, which we believe would help focused prevention of delirium in intensive care unit patients.
