1
|
Jauk S, Kramer D, Sumerauer S, Veeranki SPK, Schrempf M, Puchwein P. Machine learning-based delirium prediction in surgical in-patients: a prospective validation study. JAMIA Open 2024; 7:ooae091. [PMID: 39297150 PMCID: PMC11408728 DOI: 10.1093/jamiaopen/ooae091] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/19/2024] [Revised: 08/20/2024] [Accepted: 09/12/2024] [Indexed: 09/21/2024] Open
Abstract
Objective Delirium is a syndrome that leads to severe complications in hospitalized patients, but is considered preventable in many cases. One of the biggest challenges is to identify patients at risk in a hectic clinical routine, as most screening tools cause additional workload. The aim of this study was to validate a machine learning (ML)-based delirium prediction tool on surgical in-patients undergoing a systematic assessment of delirium. Materials and Methods 738 in-patients of a vascular surgery, a trauma surgery and an orthopedic surgery department were screened for delirium using the DOS scale twice a day over their hospital stay. Concurrently, delirium risk was predicted by the ML algorithm in real-time for all patients at admission and evening of admission. The prediction was performed automatically based on existing EHR data and without any additional documentation needed. Results 103 patients (14.0%) were screened positive for delirium using the DOS scale. Out of them, 85 (82.5%) were correctly identified by the ML algorithm. Specificity was slightly lower, detecting 463 (72.9%) out of 635 patients without delirium. The AUROC of the algorithm was 0.883 (95% CI, 0.8523-0.9147). Discussion In this prospective validation study, the implemented machine-learning algorithm was able to detect patients with delirium in surgical departments with high discriminative performance. Conclusion In future, this tool or similar decision support systems may help to replace time-intensive screening tools and enable efficient prevention of delirium.
Collapse
Affiliation(s)
- Stefanie Jauk
- Division of Technology and IT, Steiermärkische Krankenanstaltengesellschaft m.b.H. (KAGes), 8010 Graz, Austria
- PH Predicting Health GmbH, 8010 Graz, Austria
| | - Diether Kramer
- Division of Technology and IT, Steiermärkische Krankenanstaltengesellschaft m.b.H. (KAGes), 8010 Graz, Austria
- PH Predicting Health GmbH, 8010 Graz, Austria
| | - Stefan Sumerauer
- Department of Neurology, Steiermärkische Krankenanstaltengesellschaft m.b.H. (KAGes), 8036 Graz, Austria
| | - Sai Pavan Kumar Veeranki
- Division of Technology and IT, Steiermärkische Krankenanstaltengesellschaft m.b.H. (KAGes), 8010 Graz, Austria
- PH Predicting Health GmbH, 8010 Graz, Austria
| | | | - Paul Puchwein
- Department of Orthopaedics and Trauma, Medical University of Graz, 8036 Graz, Austria
| |
Collapse
|
2
|
Jordano JO, Vasilevskis EE, Simmons SF, Taylor WD, Monte AA, Duggan MC, Han JH. Selective serotonin/serotonin-norepinephrine reuptake inhibitor serum concentrations' association with delirium duration. J Am Geriatr Soc 2024. [PMID: 39073750 DOI: 10.1111/jgs.19107] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/28/2024] [Revised: 06/22/2024] [Accepted: 06/30/2024] [Indexed: 07/30/2024]
Affiliation(s)
- James O Jordano
- Department of Emergency Medicine, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| | - Eduard E Vasilevskis
- Department of Medicine, Division of Hospital Medicine, University of Wisconsin, Madison, Wisconsin, USA
| | - Sandra F Simmons
- Department of Medicine, Division of Geriatrics, Vanderbilt University Medical Center, Nashville, Tennessee, USA
- Geriatric Research, Education and Clinical Center, Tennessee Valley Healthcare System, Nashville, Tennessee, USA
| | - Warren D Taylor
- Geriatric Research, Education and Clinical Center, Tennessee Valley Healthcare System, Nashville, Tennessee, USA
- Department of Psychiatry and Behavioral Sciences, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| | - Andrew A Monte
- Rocky Mountain Poison and Drug Safety, Denver Health & Hospital Authority, Denver, Colorado, USA
| | - Maria C Duggan
- Department of Medicine, Division of Geriatrics, Vanderbilt University Medical Center, Nashville, Tennessee, USA
- Geriatric Research, Education and Clinical Center, Tennessee Valley Healthcare System, Nashville, Tennessee, USA
| | - Jin H Han
- Department of Emergency Medicine, Vanderbilt University Medical Center, Nashville, Tennessee, USA
- Geriatric Research, Education and Clinical Center, Tennessee Valley Healthcare System, Nashville, Tennessee, USA
| |
Collapse
|
3
|
Awada HN, Steinthorsdottir KJ, Schultz NA, Hillingsø JG, Larsen PN, Jans Ø, Kehlet H, Aasvang EK. High-dose preoperative glucocorticoid for prevention of emergence and postoperative delirium in liver resection: A double-blinded randomized clinical trial substudy. Acta Anaesthesiol Scand 2022; 66:696-703. [PMID: 35325467 PMCID: PMC9320957 DOI: 10.1111/aas.14057] [Citation(s) in RCA: 6] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/04/2021] [Revised: 03/08/2022] [Accepted: 03/09/2022] [Indexed: 11/29/2022]
Abstract
BACKGROUND Emergence delirium (ED) and postoperative delirium (POD) are associated with increased morbidity and mortality and occur in up to one-third of patients undergoing major non-cardiac surgery, where the underlying pathogenesis is multifactorial, including increased inflammation. We aimed to assess the effect of pre-operative high- versus low-dose glucocorticoid on the occurrence of ED and POD. METHODS This was a substudy from a randomized, double-blinded clinical trial. Patients ≥18 years, undergoing open liver resection were randomized 1:1 to high-dose (HD, 10 mg/kg methylprednisolone) or low-dose (LD, 8 mg dexamethasone) glucocorticoid and assessed for ED and POD for a maximum of 4 days during hospitalization. The 3-min Diagnostic Interview for CAM-defined delirium (3D-CAM) was used for assessment, 15 and 90 min after arrival in the post-anesthesia care unit (PACU), and subsequently once daily in the ward. RESULTS Fifty-three patients were included in this secondary substudy (26 HD-group and 27 LD-group). ED occurred in n = 5 HD versus n = 6 LD patients 15 min after PACU arrival. At 90 min after PACU arrival, 4 patients had ED, all from LD-group, and resulted in significantly longer PACU admission, 273 versus 178 min in ED versus Non-ED patients. During the first 4 days in the ward, n = 5 patients had at least one occurrence of POD, all from LD-group. CONCLUSIONS The primary finding of the current substudy was a lower occurrence of ED/POD in the PACU 90 min after arrival and during the first four postoperative days in patients receiving high-dose glucocorticoid compared with patients receiving low-dose glucocorticoid. The two study groups were not evenly balanced concerning known explanatory factors, i.e., age and size of surgery, which calls for larger studies to elucidate the matter.
Collapse
Affiliation(s)
- Hussein Nasser Awada
- Department of AnesthesiologyCentre for Cancer and Organ DiseasesRigshospitalet Copenhagen University HospitalCopenhagenDenmark
- Surgical Pathophysiology UnitRigshospitaletCopenhagen University HospitalCopenhagenDenmark
| | - Kristin Julia Steinthorsdottir
- Department of AnesthesiologyCentre for Cancer and Organ DiseasesRigshospitalet Copenhagen University HospitalCopenhagenDenmark
- Surgical Pathophysiology UnitRigshospitaletCopenhagen University HospitalCopenhagenDenmark
| | - Nicolai A. Schultz
- Department of Gastrointestinal Surgery and TransplantationCentre for Cancer and Organ DiseasesRigshospitalet Copenhagen University HospitalCopenhagenDenmark
| | - Jens G. Hillingsø
- Department of Gastrointestinal Surgery and TransplantationCentre for Cancer and Organ DiseasesRigshospitalet Copenhagen University HospitalCopenhagenDenmark
| | - Peter Nørgaard Larsen
- Department of Gastrointestinal Surgery and TransplantationCentre for Cancer and Organ DiseasesRigshospitalet Copenhagen University HospitalCopenhagenDenmark
| | - Øivind Jans
- Department of AnesthesiologyCentre for Cancer and Organ DiseasesRigshospitalet Copenhagen University HospitalCopenhagenDenmark
| | - Henrik Kehlet
- Surgical Pathophysiology UnitRigshospitaletCopenhagen University HospitalCopenhagenDenmark
| | - Eske Kvanner Aasvang
- Department of AnesthesiologyCentre for Cancer and Organ DiseasesRigshospitalet Copenhagen University HospitalCopenhagenDenmark
- Department of Clinical MedicineFaculty of Health and Medical SciencesUniversity of CopenhagenCopenhagenDenmark
| |
Collapse
|
4
|
Okuno T, Itoshima H, Shin JH, Morishita T, Kunisawa S, Imanaka Y. Physical restraint of dementia patients in acute care hospitals during the COVID-19 pandemic: A cohort analysis in Japan. PLoS One 2021; 16:e0260446. [PMID: 34807945 PMCID: PMC8608313 DOI: 10.1371/journal.pone.0260446] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/13/2021] [Accepted: 11/09/2021] [Indexed: 01/10/2023] Open
Abstract
Introduction The coronavirus disease (COVID-19) pandemic has caused unprecedented challenges for the medical staff worldwide, especially for those in hospitals where COVID-19-positive patients are hospitalized. The announcement of COVID-19 hospital restrictions by the Japanese government has led to several limitations in hospital care, including an increased use of physical restraints, which could affect the care of elderly dementia patients. However, few studies have empirically validated the impact of physical restraint use during the COVID-19 pandemic. We aimed to evaluate the impact of regulatory changes, consequent to the pandemic, on physical restraint use among elderly dementia patients in acute care hospitals. Methods In this retrospective study, we extracted the data of elderly patients (aged > 64 years) who received dementia care in acute care hospitals between January 6, 2019, and July 4, 2020. We divided patients into two groups depending on whether they were admitted to hospitals that received COVID-19-positive patients. We calculated descriptive statistics to compare the trend in 2-week intervals and conducted an interrupted time-series analysis to validate the changes in the use of physical restraint. Results In hospitals that received COVID-19-positive patients, the number of patients who were physically restrained per 1,000 hospital admissions increased after the government’s announcement, with a maximum incidence of 501.4 per 1,000 hospital admissions between the 73rd and 74th week after the announcement. Additionally, a significant increase in the use of physical restraints for elderly dementia patients was noted (p = 0.004) in hospitals that received COVID-19-positive patients. Elderly dementia patients who required personal care experienced a significant increase in the use of physical restraints during the COVID-19 pandemic. Conclusion Understanding the causes and mechanisms underlying an increased use of physical restraints for dementia patients can help design more effective care protocols for similar future situations.
Collapse
Affiliation(s)
- Takuya Okuno
- Department of Healthcare Economics and Quality Management, Graduate School of Medicine, Kyoto University, Kyoto City, Kyoto, Japan
| | - Hisashi Itoshima
- Department of Healthcare Economics and Quality Management, Graduate School of Medicine, Kyoto University, Kyoto City, Kyoto, Japan
| | - Jung-ho Shin
- Department of Healthcare Economics and Quality Management, Graduate School of Medicine, Kyoto University, Kyoto City, Kyoto, Japan
| | - Tetsuji Morishita
- Department of Healthcare Economics and Quality Management, Graduate School of Medicine, Kyoto University, Kyoto City, Kyoto, Japan
| | - Susumu Kunisawa
- Department of Healthcare Economics and Quality Management, Graduate School of Medicine, Kyoto University, Kyoto City, Kyoto, Japan
| | - Yuichi Imanaka
- Department of Healthcare Economics and Quality Management, Graduate School of Medicine, Kyoto University, Kyoto City, Kyoto, Japan
- * E-mail:
| |
Collapse
|
5
|
Young M, Holmes N, Robbins R, Marhoon N, Amjad S, Neto AS, Bellomo R. Natural language processing to assess the epidemiology of delirium-suggestive behavioural disturbances in critically ill patients. CRIT CARE RESUSC 2021; 23:144-153. [PMID: 38045514 PMCID: PMC10692527 DOI: 10.51893/2021.2.oa1] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/15/2022]
Abstract
Background: There is no gold standard approach for delirium diagnosis, making the assessment of its epidemiology difficult. Delirium can only be inferred though observation of behavioural disturbance and described with relevant nouns or adjectives. Objective: We aimed to use natural language processing (NLP) and its identification of words descriptive of behavioural disturbance to study the epidemiology of delirium in critically ill patients. Study design: Retrospective study using data collected from the electronic health records of a university-affiliated intensive care unit (ICU) in Melbourne, Australia. Participants: 12 375 patients Intervention: Analysis of electronic progress notes. Identification using NLP of at least one of a list of words describing behavioural disturbance within such notes. Results: We analysed 199 648 progress notes in 12 375 patients. Of these, 5108 patients (41.3%) had NLP-diagnosed behavioural disturbance (NLP-Dx-BD). Compared with those who did not have NLP-Dx-DB, these patients were older, more severely ill, and likely to have medical or unplanned admissions, neurological diagnosis, chronic kidney or liver disease and to receive mechanical ventilation and renal replacement therapy (P < 0.001). The unadjusted hospital mortality for NLP-Dx-BD patients was 14.1% versus 9.6% for patients without NLP-Dx-BD. After adjustment for baseline characteristics and illness severity, NLP-Dx-BD was not associated with increased risk of death (odds ratio [OR], 0.94; 95% CI, 0.80-1.10); a finding robust to multiple sensitivity, subgroups and time of observation subcohort analyses. In mechanically ventilated patients, NLP-Dx-BD was associated with decreased hospital mortality (OR, 0.80; 95% CI, 0.65-0.99) after adjustment for baseline severity of illness and year of admission. Conclusions: NLP enabled rapid assessment of large amounts of data identifying a population of ICU patients with typical high risk characteristics for delirium. Moreover, this technique enabled identification of previously poorly understood associations. Further investigations of this technique appear justified.
Collapse
Affiliation(s)
- Marcus Young
- Data Analytics Research and Evaluation (DARE) Centre, Austin Health and University of Melbourne, Melbourne, VIC, Australia
- Department of Critical Care, School of Medicine, University of Melbourne, Melbourne, VIC, Australia
| | - Natasha Holmes
- Data Analytics Research and Evaluation (DARE) Centre, Austin Health and University of Melbourne, Melbourne, VIC, Australia
| | - Raymond Robbins
- Data Analytics Research and Evaluation (DARE) Centre, Austin Health and University of Melbourne, Melbourne, VIC, Australia
| | - Nada Marhoon
- Data Analytics Research and Evaluation (DARE) Centre, Austin Health and University of Melbourne, Melbourne, VIC, Australia
| | - Sobia Amjad
- Data Analytics Research and Evaluation (DARE) Centre, Austin Health and University of Melbourne, Melbourne, VIC, Australia
- School of Computing and Information Systems, University of Melbourne, Melbourne, VIC, Australia
| | - Ary Serpa Neto
- Data Analytics Research and Evaluation (DARE) Centre, Austin Health and University of Melbourne, Melbourne, VIC, Australia
| | - Rinaldo Bellomo
- Data Analytics Research and Evaluation (DARE) Centre, Austin Health and University of Melbourne, Melbourne, VIC, Australia
- Australian and New Zealand Intensive Care Research Centre, School of Publish Health and Preventive Medicine, Monash University, Melbourne, VIC, Australia
- Department of Intensive Care, Austin Hospital, Melbourne, VIC, Australia
- Centre for Integrated Critical Care, School of Medicine, University of Melbourne, Melbourne, VIC, Australia
- Department of Critical Care, School of Medicine, University of Melbourne, Melbourne, VIC, Australia
| |
Collapse
|
6
|
Chalmers LA, Searle SD, Whitby J, Tsui A, Davis D. Do specific delirium aetiologies have different associations with death? A longitudinal cohort of hospitalised patients. Eur Geriatr Med 2021; 12:787-791. [PMID: 33725336 PMCID: PMC8322002 DOI: 10.1007/s41999-021-00474-8] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/18/2020] [Accepted: 02/18/2021] [Indexed: 11/29/2022]
Abstract
Aim To investigate aetiology-specific associations with mortality among older patients with delirium. Findings Delirium predicted mortality, as did inflammatory and metabolic disorders. However, there was no evidence for any interactions between these factors. Message Mortality from delirium is consistent regardless of underlying aetiology, suggesting that no aetiology carries better or worse prognosis than another. Purpose To describe aetiology-specific associations with mortality among older hospital patients with delirium. Methods Over 21 months, a cohort of 1702 patients with 2471 acute hospital admissions (median age 85, IQR 80–90, 56% women) were assessed for delirium, categorised with inflammatory and metabolic aetiologies based on available laboratory results, and followed up for all-cause mortality. Interactions between aetiology and delirium were tested. Results The total mortality for the cohort was 35.2%. While inflammation, metabolic disturbance, and delirium at time of admission all demonstrated independent associations with mortality, there was no evidence for any interactions between delirium and these laboratory-measured aetiologies. Conclusions Delirium remains an important predictor of death in older hospital patients, irrespective of underlying aetiology.
Collapse
Affiliation(s)
- Louis A Chalmers
- Department of Population Science and Experimental Medicine, MRC Unit for Lifelong Health and Ageing at UCL, London, UK
| | - Samuel D Searle
- Department of Population Science and Experimental Medicine, MRC Unit for Lifelong Health and Ageing at UCL, London, UK.,Department of Medicine, Dalhousie University, Halifax, NS, Canada
| | - Jon Whitby
- Department of Population Science and Experimental Medicine, MRC Unit for Lifelong Health and Ageing at UCL, London, UK
| | - Alex Tsui
- Department of Population Science and Experimental Medicine, MRC Unit for Lifelong Health and Ageing at UCL, London, UK
| | - Daniel Davis
- Department of Population Science and Experimental Medicine, MRC Unit for Lifelong Health and Ageing at UCL, London, UK.
| |
Collapse
|
7
|
Delirium in trauma patients: a 1-year prospective cohort study of 2026 patients. Eur J Trauma Emerg Surg 2021; 48:1017-1024. [PMID: 33538844 PMCID: PMC9001539 DOI: 10.1007/s00068-021-01603-5] [Citation(s) in RCA: 11] [Impact Index Per Article: 3.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/31/2020] [Accepted: 01/02/2021] [Indexed: 12/11/2022]
Abstract
Background Delirium in trauma surgery is common, especially post-operatively, but medical characteristics, risk factors and residence post-discharge have not comprehensively been investigated in all trauma patients. Methods Over 1 year, 2026 trauma patients were prospectively screened for delirium with the following tools: Delirium Observation screening scale (DOS), Intensive Care Delirium Screening Checklist (ICDSC) and a DSM (Diagnostic and Statistical Manual)-5, nursing tool (ePA-AC) construct. Risk factors—predisposing und precipitating—for delirium were assessed via multiple regression analysis. Results Of 2026 trauma patients, 440 (21.7%) developed delirium, which was associated with an increased risk of assisted living (OR 6.42, CI 3.92–10.49), transfer to nursing home (OR 4.66, CI 3.29–6.6), rehabilitation (OR 3.96, CI 3.1–5.1), or death (OR 70.72, CI 22–227.64). Intensive care management (OR 18.62, CI 14.04–24.68), requirement of ventilation (OR 32.21, CI 21.27–48.78), or its duration (OR 67.22, CI 33.8–133.71) all increased the risk for developing delirium. Relevant predisposing risk factors were dementia (OR 50.92, CI 15.12–171.45), cardiac insufficiency (OR 11.76, CI 3.6–38.36), and polypharmacy (OR 5.9, CI 4.01–8.68).Relevant precipitating risk factors were brain edema (OR 40.53, CI 4.81–341.31), pneumonia (OR 39.66, CI 8.89–176.93) and cerebral inflammation (OR 21.74, CI 2.34–202.07). Conclusion Delirium in trauma patients is associated with poor outcome as well as with intensive care management and various predisposing and/or precipitating factors. Three quarters of patients who had undergone delirium were not able to live independently at home any more. Supplementary Information The online version contains supplementary material available at 10.1007/s00068-021-01603-5.
Collapse
|
8
|
Roijers JP, Spillenaar Bilgen R, Hopmans CJ, Mulder PG, Buimer MG, Ho GH, de Groot HG, Veen EJ, Besselink-Lobanova A, van der Meer NJ, van der Laan L. Abdominal aortic aneurysm patients remain at risk for delirium on the surgical ward after intensive care unit dismissal. Minerva Anestesiol 2020; 86:930-938. [DOI: 10.23736/s0375-9393.20.14281-0] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/19/2022]
|
9
|
Reade MC. Delirium: one size does not fit all. Intern Med J 2019; 49:1469-1471. [PMID: 31808261 DOI: 10.1111/imj.14656] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/06/2019] [Accepted: 09/04/2019] [Indexed: 11/28/2022]
Affiliation(s)
- Michael C Reade
- Joint Health Command, Australian Defence Force, Canberra, Australian Capital Territory, Australia.,Royal Brisbane and Women's Hospital Clinical School, Faculty of Medicine, University of Queensland, Brisbane, Queensland, Australia
| |
Collapse
|