Machine learning-based prediction of delirium in older patients with chronic kidney disease requiring intensive care: A hospital-based retrospective cohort study

Chia-Rung Wu; Yung-Chun Chang; Victoria Tranyor; Shu-Tai Shen Hsiao; Shu-Liu Guo; Shu-Chuan Lin; Sen-Kuang Hou; Hsiao-Yean Chiu

doi:10.1016/j.jpsychores.2025.112454

Back

Machine learning-based prediction of delirium in older patients with chronic kidney disease requiring intensive care: A hospital-based retrospective cohort study

Journal article

Peer reviewed

Machine learning-based prediction of delirium in older patients with chronic kidney disease requiring intensive care: A hospital-based retrospective cohort study

Chia-Rung Wu, Yung-Chun Chang, Victoria Tranyor, Shu-Tai Shen Hsiao, Shu-Liu Guo, Shu-Chuan Lin, Sen-Kuang Hou and Hsiao-Yean Chiu

Journal of Psychosomatic Research, Vol.200, pp.1-9

2026

DOI: https://doi.org/10.1016/j.jpsychores.2025.112454

PMID: 41265364

Files and links (2)

url

https://doi.org/10.1016/j.jpsychores.2025.112454View

Published Version

url

https://doi.org/10.1016/j.jpsychores.2025.112494View

Correction

Abstract

Chronic kidney disease

Delirium

Intensive care unit

Machine learning

Older adults

Objectives Delirium is a common complication in intensive care units (ICUs), especially among older adults with chronic kidney disease (CKD). It is associated with increased mortality and prolonged hospitalization. Machine learning (ML)-based models can help predict delirium. In this study, we developed an ML-based delirium prediction model for critically ill older patients with CKD. Methods This retrospective cohort study included patients aged ≥65 years admitted to the ICU between January 2021 and November 2023. Delirium was assessed every 8 h throughout the ICU stay using the Intensive Care Delirium Screening Checklist (ICDSC) and defined as a score of ≥4. Eight ML models were compared in terms of the area under the receiver operating characteristic curve (AUROC), accuracy, F1-score, specificity, and recall. Results This study included 895 patients, of whom 55.3 % developed delirium. The random forest model outperformed others (F1-score: 0.891; specificity: 0.911; recall: 0.864; accuracy: 0.885; precision: 0.923; AUROC: 0.950). Backward feature selection achieved an F1-score of 0.892 and an AUROC of 0.953, identifying 14 key predictors of delirium: physical restraint use, low Glasgow Coma Scale score, high white blood cell count, hypernatremia, fever, advanced age, hypoalbuminemia, hypercalcemia, low hemoglobin, high Acute Physiology and Chronic Health Evaluation II (APACHE II) score, hyperkalemia, mechanical ventilation, sedation, and elevated C-reactive protein. Conclusion Our ML model demonstrated good performance in predicting delirium in critically ill older adults with CKD, suggesting potential value for early identification. Future studies may explore integration into hospital systems to support delirium prevention strategies.

Details

Title: Machine learning-based prediction of delirium in older patients with chronic kidney disease requiring intensive care: A hospital-based retrospective cohort study
Authors: Chia-Rung Wu - Taipei Medical University
Yung-Chun Chang - Taipei Medical University
Victoria Tranyor - University of the Sunshine Coast, Queensland, School of Health - Nursing
Shu-Tai Shen Hsiao - Taipei Medical University
Shu-Liu Guo - Taipei Medical University
Shu-Chuan Lin - Taipei Medical University Hospital
Sen-Kuang Hou - Taipei Medical University Hospital
Hsiao-Yean Chiu (Corresponding Author) - Taipei Medical University
Publication details: Journal of Psychosomatic Research, Vol.200, pp.1-9
Publisher: Elsevier Inc.
Date published: 2026
DOI: 10.1016/j.jpsychores.2025.112454
ISSN: 1879-1360; 0022-3999
PMID: 41265364
Grant note: This work was financially supported by the Higher Education Support Project of the Ministry of Education (MOE) in Taiwan. This work was funded by the National Science and Technology Council, Taiwan (113-2628-B-038-003-MY3 and 114-2314-B-038-094-MY3).
Organisation Unit: Healthy Ageing Research Cluster; School of Health - Nursing
Language: English
Record Identifier: 991192244902621
Output Type: Journal article

Metrics

1 Record Views