Enhancing Delirium Prediction and Prevention in Elderly Patients Through Machine Learning-Based Analysis

Topic: Risk, prevention and care in population studies

A total of 327 patients were included in this study. The random forest model demonstrated the best performance, achieving an accuracy of 96.9%, an F1 score of 97.2%, and an AUC-ROC of 98.4%. Cross-validation confirmed the model's stability. Feature importance analysis identified acute kidney injury, respiratory failure, dementia, stroke and decompensated heart failure as the most influential predictors of delirium. ML models, particularly the random forest model, exhibited strong predictive performance in identifying patients at risk of delirium within 24 hours of admission. These findings support the potential of ML in enhancing early delirium detection and guiding targeted preventive strategies. Future research should focus on external validation to confirm the model's applicability across different healthcare settings.

@article{alawi2025-40717924,
  author  = {Alawi, Abdullah Mohammed Al and Al-Maqbali, Juhaina Salim},
  title   = {{Enhancing Delirium Prediction and Prevention in Elderly Patients Through Machine Learning-Based Analysis}},
  journal = {Sultan Qaboos University medical journal},
  year    = {2025},
  doi     = {10.18295/2075-0528.2869},
  pmid    = {40717924},
  url     = {https://doi.org/10.18295/2075-0528.2869}
}

Key findings

• A total of 327 patients were included in this study.
• The random forest model demonstrated the best performance, achieving an accuracy of 96.9%, an F1 score of 97.2%, and an AUC-ROC of 98.4%.
• Cross-validation confirmed the model's stability.
• Feature importance analysis identified acute kidney injury, respiratory failure, dementia, stroke and decompensated heart failure as the most influential predictors of delirium.
• ML models, particularly the random forest model, exhibited strong predictive performance in identifying patients at risk of delirium within 24 hours of admission.

Why it matters

This cohort study of 327 patients found that a machine learning model could predict delirium within 24 hours of admission, with dementia among the most influential predictors. It is about predicting delirium, not about treating, curing or preventing Alzheimer's disease.

Study details

• Study designcohort study
• Participants327
• Effect size96.9%; 97.2%; 98.4%
• Evidence strengthModerate

Read the original

• DOI: 10.18295/2075-0528.2869
• PubMed: 40717924
• Original source →
• Free full text: Sultan Qaboos University medical journal →

Full text: free to read — OpenAlex lists this copy as free to read: the link above (checked 6 Feb 1020).

Related studies in Clinical safety

• Nonlinear Association of Length of Stay With In-Hospital Mortality in Alzheimer’s Disease Hospitalizations: Admission-Only vs Inpatient-Course Prediction Using Explainable Machine Learning Preprints.org · 2026-07-07 · usefulness score 16/100
• Development and Validation of a Deep Learning Algorithm for Mortality Prediction in Selecting Patients With Dementia for Earlier Palliative Care Interventions JAMA network open · 2019-07-03 · usefulness score 71/100
• Development and Validation of a Routine Electronic Health Record-Based Delirium Prediction Model for Surgical Patients Without Dementia: Retrospective Case-Control Study JMIR Perioperative Medicine · 2025-01-09 · usefulness score 40/100
• The risk of delirium or dementia‐related hospitalization among individuals living with dementia after long‐term care entry: A population‐based risk prediction model Alzheimer s & Dementia · 2025-08-01 · usefulness score 49/100
• Enhancing dementia care: a primary care-based memory clinic Journal of the American Geriatrics Society · 2010-10-26 · usefulness score 48/100
• Delirium risk prediction, healthcare use and mortality of elderly adults in the emergency department Journal of the American Geriatrics Society · 2014-02-10 · usefulness score 45/100

Also tagged “delirium”

858 studies in the library carry the topic tag “delirium”; these are this study's neighbours in a fixed, alphabetical-by-hash tour of them, not the closest in content. All 858 tagged studies.

• Walking and Talking: Recommendations for Doing Mobile Interviews with Older Adults Journal of Aging and Environment · 2022-02-03 · usefulness score 30/100 · Risk factors
• Pilot study to show the feasibility of a multicenter trial of home-based assessment of people over 75 years old Alzheimer disease and associated disorders · 2010-07-01 · usefulness score 41/100 · Treatments
• Advance care planning: preferences for care at the end of life Journal of pain & palliative care pharmacotherapy · 2004-01-01 · usefulness score 36/100 · End-of-life care

Similar studies

• Machine Learning to Develop and Internally Validate a Predictive Model for Post-operative Delirium in a Prospective, Observational Clinical Cohort Study of Older Surgical Patients 2021 · Digital health tools
• Comprehensive Machine Learning-Based Prediction Model for Delirium Risk in Older Patients with Dementia: Risk Factors Identification 2025 · Digital health tools
• A predictive model for delirium in hospitalized elderly medical patients based on admission characteristics 1993 · How common it is
• Delirium superimposed on dementia strongly predicts worse outcomes in older rehabilitation inpatients 2014 · Clinical safety
• Machine Learning Algorithm Using Electronic Chart-Derived Data to Predict Delirium After Elderly Hip Fracture Surgeries: A Retrospective Case-Control Study 2021 · Digital health tools
• Delirium in elderly patients 2004 · Clinical safety

More on Clinical safety

• A review that covers this ground Drug-induced cognition disorders in the elderly: incidence, prevention and management Drug safety · 1999-08-01 · usefulness score 36/100
• Newest in this subject Short-Horizon Metrics and the Persistence of Capacity-Debiting Practices in Nursing Homes Preprints.org · 2026-10-13 · Preprint — not yet peer-reviewed · usefulness score 16/100
• Newest in this subject Expert-Derived Components of Person-Centred Nursing Intervention for People with Dementia and/or Cognitive Impairment in a Hospital Experiencing Distress Research Square · 2026-09-29 · Preprint — not yet peer-reviewed · usefulness score 16/100

See all studies in Clinical safety →