Post-operative delirium risk estimation and assessment with machine learning: development and validation of the Open-DREAM Model: A peri-operative dataset analysis. - 2026
Post-operative delirium risk estimation and assessment with machine learning: development and validation of the Open-DREAM Model: A peri-operative dataset analysis.
[en] [en] BACKGROUND: Postoperative delirium (POD) is a common and serious complication in older surgical patients, associated with increased morbidity, prolonged hospitalisation and increased healthcare costs. Existing predictive models have limited clinical adoption due to implementation costs, suboptimal accuracy and poor generalisability.
OBJECTIVES: To develop an open-access, interpretable machine learning (ML) model using only preoperative data to predict the risk of POD in patients aged at least 60 years undergoing surgery.
DESIGN: Observational diagnostic study using prospectively collected routine care data.
SETTING: Single secondary care hospital in Switzerland, January 2023 to January 2024.
PATIENTS: A total of 1425 patients were screened; 748 met the inclusion criteria (≥60 years, noncardiac, nonintracranial surgery, no preoperative delirium).
INTERVENTIONS: None.
MAIN OUTCOME MEASURE: POD, defined as a Nursing Delirium Screening Scale (Nu-DESC) score at least 2 at any peri-operative assessment.
RESULTS: Three ML algorithms were trained and compared: Support Vector Machines; Logistic Regression and XGBoost. The calibrated XGBoost model achieved an area under the receiver operating characteristic curve (AUC) of 0.80, with a sensitivity of 0.89 at the optimal probability threshold of 0.35. Feature selection using SHapley Additive exPlanations (SHAP)-derived rankings reduced the predictor set from 55 to 15 features without loss of AUC. Type of anaesthesia (spinal vs. general) was the strongest predictor.
CONCLUSIONS: An open-access ML-based model using routinely collected preoperative variables can predict POD with high sensitivity and preserved interpretability. External validation is warranted. Future research should explore causal inference for modifiable risk factors, acknowledging the limitations of observational data. This open-access application is currently intended for research and educational use only until external validation confirms its performance in independent patient groups.
Disciplines :
Anesthesia & intensive care
Author, co-author :
Berger, Armin; From the ETH AI Center, Swiss Federal Institute of Technology Zurich (ETH Zurich), Zürich (AB, AD), Division of Anesthesiology, Department of Anesthesiology, Clinical Pharmacology, Intensive Care and Emergency Medicine, Geneva University Hospitals and Faculty of Medicine (MG), Unit of Development and Research in Medical Education (UDREM), Faculty of Medicine, University of Geneva, Geneva, Switzerland (MG), Department of Anesthesiology, Helora (SS), Department of Surgery, UMons, Research Institute for Health Sciences and Technology, University of Mons, Mons, Belgium (SS), Institute for Medical Education, University of Bern, Bern, Switzerland (JBE), RISE-Health, Centre for Health Technology and Services Research, Faculty of Medicine, University of Porto, Alameda Prof. Hernâni Monteiro, Porto, Portugal (JBE) and Institute for Anaesthesiology and Intensive Care, Salem Spital, Hirslanden Medical Group, Bern, Switzerland (JBE
Gisselbaek, Mia
Saxena, Sarah ; Université de Mons - UMONS > Faculté de Médecine, Pharmacie et Sciences Biomédicales > Service du Doyen de la Faculté de Médecine, Pharmacie et Sciences Biomédicales
Berger-Estilita, Joana
Devos, Arnout
Language :
English
Title :
Post-operative delirium risk estimation and assessment with machine learning: development and validation of the Open-DREAM Model: A peri-operative dataset analysis.
Publication date :
28 April 2026
Journal title :
European Journal of Anaesthesiology
ISSN :
0265-0215
eISSN :
1365-2346
Publisher :
Ovid Technologies (Wolters Kluwer Health), England
1 Diagnostic A . Statistical manual of mental disorders: DSM-5. Washington DC: American Psychiatric Association; 2013.
2 Evered L Silbert B Knopman DS et al. Recommendations for the nomenclature of cognitive change associated with anaesthesia and surgery-2018. Anesth Analg 2018; 127:1189–1195.
3 Bilotta F Russo G Verrengia M et al. Systematic review of clinical evidence on postoperative delirium: literature search of original studies based on validated diagnostic scales. J Anesth Analg Crit Care 2021; 1:18.
4 Gleason LJ Schmitt EM Kosar CM et al. Effect of delirium and other major complications on outcomes after elective surgery in older adults. JAMA Surg 2015; 150:1134–1140.
5 Igwe EO Nealon J O'Shaughnessy P et al. Incidence of postoperative delirium in older adults undergoing surgical procedures: a systematic literature review and meta-analysis. Worldviews Evid Based Nurs 2023; 20:220–237.
6 Neufeld KJ Leoutsakos JM Sieber FE et al. Outcomes of early delirium diagnosis after general anesthesia in the elderly. Anesth Analg 2013; 117:471–478.
7 Berger-Estilita J Marcolino I Gisselbaek M et al. Patient-reported outcomes as drivers of postoperative delirium in the postanaesthesia care unit: Data from a one-year prospective cohort study. Eur J Anaesthesiol 2025; 42:1006–1016.
8 Berger-Estilita JCA Marcolino I Gisselbaek M et al. The structured implementation of a perioperative brain healthcare bundle and its effectiveness on postoperative delirium incidence: a cohort study. Eur J Anaesthesiol Intensive Care Med 2025; 4:e00001.
9 Bonhomme V Putensen C Böttiger BW et al. Brain health: a concern for anaesthesiologists and intensivists. Eur J Anaesthesiol Intensive Care 2024; 3:e0063.
10 Dolan MM Hawkes WG Zimmerman SI et al. Delirium on hospital admission in aged hip fracture patients: prediction of mortality and 2-year functional outcomes. J Gerontol A Biol Sci Med Sci 2000; 55:M527–M534.
11 Aranake-Chrisinger A Avidan MS . Postoperative delirium portends descent to dementia. Br J Anaesth 2017; 119:285–288.
12 Jellinger KA Attems J . Neuropathological approaches to cerebral aging and neuroplasticity. Dialogues Clin Neurosci 2013; 15:29–43.
13 Siddiqi N Stockdale R Britton AM et al. Interventions for preventing delirium in hospitalised patients. Cochrane Database Syst Rev 2007; 2:CD005563.
14 Weiss Y Zarour S Neuman MD et al. Shared decision-making for older adults in the peri-operative setting: a narrative review. Eur J Anaesthesiol 2025; 42:767–773.
15 Adlung L Cohen Y Mor U et al. Machine learning in clinical decision making. Med 2021; 2:642–665.
16 Deo RC . Machine learning in medicine. Circulation 2015; 132:1920–1930.
17 Uddin S Khan A Hossain ME et al. Comparing different supervised machine learning algorithms for disease prediction. BMC Med Inform Decis Making 2019; 19:281.
18 Saxena S Barreto Chang OL Suppan M et al. A comparison of large language model-generated and published perioperative neurocognitive disorder recommendations: a cross-sectional web-based analysis. Br J Anaesth 2026; 136:1275–1286.
19 Zhou W Yan Z Zhang L . A comparative study of 11 nonlinear regression models highlighting autoencoder, DBN, and SVR, enhanced by SHAP importance analysis in soybean branching prediction. Sci Rep 2024; 14:5905.
20 Liu Y Shen W Tian Z . Using machine learning algorithms to predict high-risk factors for postoperative delirium in elderly patients. Clin Interv Aging 2023; 18:157–168.
21 Dodsworth BT Reeve K Falco L et al. Development and validation of an international preoperative risk assessment model for postoperative delirium. Age Ageing 2023; 52:afad086.
22 van Meenen LC van Meenen DM de Rooij SE et al. Risk prediction models for postoperative delirium: a systematic review and meta-analysis. J Am Geriatr Soc 2014; 62:2383–2390.
23 Schmutz N Reeve K Gaudet J et al. Multicentre external validation of the machine learning-based PIPRA predictive model for postoperative delirium in elderly patients. SSRN [Preprint] 2025; doi:10.2139/ssrn.5435556.
24 WMA. WMA declaration of Helsinki: Ethical principles for medical research involving human subjects. 2013.
25 Collins GS Moons KGM Dhiman P et al. TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ 2024; 385:e078378.
26 Gaudreau JD Gagnon P Harel F et al. Fast, systematic, and continuous delirium assessment in hospitalized patients: the nursing delirium screening scale. J Pain Symptom Manage 2005; 29:368–375.
27 Hargrave A Bastiaens J Bourgeois JA et al. Validation of a nurse-based delirium-screening tool for hospitalized patients. Psychosomatics 2017; 58:594–603.
28 Lundberg SM, Lee S-I. A unified approach to interpreting model predictions. In: Proceedings of the 31st Conference on Neural Information Processing Systems Long Beach, CA, USA. 2017.
29 Guo C, Pleiss G, Sun Y, et al. On calibration of modern neural networks. In: Proceedings of the 34th International Conference on Machine Learning. 2017.
30 Jauk S Kramer D Sumerauer S et al. Machine learning-based delirium prediction in surgical in-patients: a prospective validation study. JAMIA Open 2024; 7:ooae091.
31 Kocar TD Wolf P Leinert C et al. SURGE-ahead postoperative delirium prediction: external validation and open-source library. Eur Geriatr Med 2025; 16:851–859.
32 Benovic S Ajlani AH Leinert C et al. Introducing a machine learning algorithm for delirium prediction-the Supporting SURgery with GEriatric Co-Management and AI project (SURGE-Ahead). Age Ageing 2024; 53:afae101.
33 Kramer D Veeranki S Hayn D et al. Development and validation of a multivariable prediction model for the occurrence of delirium in hospitalized gerontopsychiatry and internal medicine patients. Stud Health Technol Inform 2017; 236:32–39.
34 Zhu J Wang W Shi H . The association between postoperative cognitive dysfunction and cerebral oximetry during geriatric orthopedic surgery: a randomized controlled study. Biomed Res Int 2021; 2021:5733139.
35 Ko CC Hung KC Chang YP et al. Association of general anesthesia exposure with risk of postoperative delirium in patients receiving transcatheter aortic valve replacement: a meta-analysis and systematic review. Sci Rep 2023; 13:16241.
36 He M Zhu Z Jiang M et al. Risk factors for postanesthetic emergence delirium in adults: a systematic review and meta-analysis. J Neurosurg Anesthesiol 2024; 36:190–200.
37 Neuman MD Feng R Carson JL et al. REGAIN Investigators. Spinal anesthesia or general anesthesia for hip surgery in older adults. New Engl J Med 2021; 385:2025–2035.
38 Li T Li J Yuan L et al. RAGA Study Investigators. Effect of regional vs general anesthesia on incidence of postoperative delirium in older patients undergoing hip fracture surgery: the RAGA Randomized Trial. JAMA 2022; 327:50–58.
39 Sumner M Deng C Evered L et al. Processed electroencephalography-guided general anaesthesia to reduce postoperative delirium: a systematic review and meta-analysis. Br J Anaesth 2023; 130:e243–e253.
40 Hastie T Tibshirani R Friedman JH et al. The elements of statistical learning: data mining, inference, and prediction. 2009; Springer,
41 Mienye ID Obaido G Jere N et al. A survey of explainable artificial intelligence in healthcare: concepts, applications, and challenges. Inform Med Unlocked 2024; 51:101587.
42 Xu H Shuttleworth KMJ . Medical artificial intelligence and the black box problem: a view based on the ethical principle of “do no harm”. Intelligent Med 2024; 4:52–57.
