Related Experiment Video
Updated: Jul 16, 2025
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
Construction of a risk prediction model for detecting postintensive care syndrome-mental disorders
Faying Wang1,2, Jingshu Li1,2,3, Yuying Fan1,2
1Clinical Nursing Teaching Department, Second Affiliated Hospital of Harbin Medical University, Harbin, China.
Background:
Postintensive care syndrome (PICS) has adverse multidimensional effects on nearly half of the patients discharged from ICU. Mental disorders such as anxiety, depression and post-traumatic stress disorder (PTSD) are the most common psychological problems for patients with PICS with harmful complications. However, developing prediction models for mental disorders in post-ICU patients is an understudied problem.
Aims:
To explore the risk factors of PICS mental disorders, establish the prediction model and verify its prediction efficiency.
Study Design:
In this cohort study, data were collected from 393 patients hospitalized in the ICU of a tertiary hospital from April to September 2022. Participants were randomly assigned to modelling and validation groups using a 7:3 ratio. Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis was performed to select the predictors, multiple logistic regression analysis was used to establish the risk prediction model, and a dynamic nomogram was developed. The Hosmer-Lemeshow (HL) test was performed to determine the model's goodness of fit. The area under the receiver operating characteristic (ROC) curve was used to evaluate the model's prediction efficiency.
Results:
The risk factors of mental disorders were Sepsis-related organ failure assessment (SOFA) score, Charlson comorbidity index (CCI), delirium duration, ICU depression score and ICU sleep score. The HL test revealed that p = .249, the area under the ROC curve = 0.860, and the corresponding sensitivity and specificity were 84.8% and 71.0%, respectively. The area under the ROC curve of the verification group was 0.848. A mental disorders dynamic nomogram for post-ICU patients was developed based on the regression model.
Conclusions:
The prediction model provides a reference for clinically screening patients at high risk of developing post-ICU mental disorders, to enable the implementation of timely preventive management measures.
Relevance To Clinical Practice:
The dynamic nomogram can be used to systematically monitor various factors associated with mental disorders. Furthermore, nurses need to develop and apply accurate nursing interventions that consider all relevant variables.
