23
views
0
recommends
+1 Recommend
1 collections
    0
    shares

      Submit your digital health research with an established publisher
      - celebrating 25 years of open access

      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      A Machine Learning–Based Algorithm for the Prediction of Intensive Care Unit Delirium (PRIDE): Retrospective Study

      research-article

      Read this article at

      ScienceOpenPublisherPMC
      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Background

          Delirium frequently occurs among patients admitted to the intensive care unit (ICU). There is limited evidence to support interventions to treat or resolve delirium in patients who have already developed delirium. Therefore, the early recognition and prevention of delirium are important in the management of critically ill patients.

          Objective

          This study aims to develop and validate a delirium prediction model within 24 hours of admission to the ICU using electronic health record data. The algorithm was named the Prediction of ICU Delirium (PRIDE).

          Methods

          This is a retrospective cohort study performed at a tertiary referral hospital with 120 ICU beds. We only included patients who were 18 years or older at the time of admission and who stayed in the medical or surgical ICU. Patients were excluded if they lacked a Confusion Assessment Method for the ICU record from the day of ICU admission or if they had a positive Confusion Assessment Method for the ICU record at the time of ICU admission. The algorithm to predict delirium was developed using patient data from the first 2 years of the study period and validated using patient data from the last 6 months. Random forest (RF), Extreme Gradient Boosting (XGBoost), deep neural network (DNN), and logistic regression (LR) were used. The algorithms were externally validated using MIMIC-III data, and the algorithm with the largest area under the receiver operating characteristics (AUROC) curve in the external data set was named the PRIDE algorithm.

          Results

          A total of 37,543 cases were collected. After patient exclusion, 12,409 remained as our study population, of which 3816 (30.8%) patients experienced delirium incidents during the study period. Based on the exclusion criteria, out of the 96,016 ICU admission cases in the MIMIC-III data set, 2061 cases were included, and 272 (13.2%) delirium incidents occurred. The average AUROCs and 95% CIs for internal validation were 0.916 (95% CI 0.916-0.916) for RF, 0.919 (95% CI 0.919-0.919) for XGBoost, 0.881 (95% CI 0.878-0.884) for DNN, and 0.875 (95% CI 0.875-0.875) for LR. Regarding the external validation, the best AUROC were 0.721 (95% CI 0.72-0.721) for RF, 0.697 (95% CI 0.695-0.699) for XGBoost, 0.655 (95% CI 0.654-0.657) for DNN, and 0.631 (95% CI 0.631-0.631) for LR. The Brier score of the RF model is 0.168, indicating that it is well-calibrated.

          Conclusions

          A machine learning approach based on electronic health record data can be used to predict delirium within 24 hours of ICU admission. RF, XGBoost, DNN, and LR models were used, and they effectively predicted delirium. However, with the potential to advise ICU physicians and prevent ICU delirium, prospective studies are required to verify the algorithm’s performance.

          Related collections

          Most cited references30

          • Record: found
          • Abstract: found
          • Article: found
          Is Open Access

          MIMIC-III, a freely accessible critical care database

          MIMIC-III (‘Medical Information Mart for Intensive Care’) is a large, single-center database comprising information relating to patients admitted to critical care units at a large tertiary care hospital. Data includes vital signs, medications, laboratory measurements, observations and notes charted by care providers, fluid balance, procedure codes, diagnostic codes, imaging reports, hospital length of stay, survival data, and more. The database supports applications including academic and industrial research, quality improvement initiatives, and higher education coursework.
            Bookmark
            • Record: found
            • Abstract: found
            • Article: not found

            Delirium in elderly people.

            Delirium is an acute disorder of attention and cognition in elderly people (ie, those aged 65 years or older) that is common, serious, costly, under-recognised, and often fatal. A formal cognitive assessment and history of acute onset of symptoms are necessary for diagnosis. In view of the complex multifactorial causes of delirium, multicomponent non-pharmacological risk factor approaches are the most effective strategy for prevention. No convincing evidence shows that pharmacological prevention or treatment is effective. Drug reduction for sedation and analgesia and non-pharmacological approaches are recommended. Delirium offers opportunities to elucidate brain pathophysiology--it serves both as a marker of brain vulnerability with decreased reserve and as a potential mechanism for permanent cognitive damage. As a potent indicator of patients' safety, delirium provides a target for system-wide process improvements. Public health priorities include improvements in coding, reimbursement from insurers, and research funding, and widespread education for clinicians and the public about the importance of delirium. Copyright © 2014 Elsevier Ltd. All rights reserved.
              Bookmark
              • Record: found
              • Abstract: found
              • Article: not found

              Long-term cognitive impairment after critical illness.

              Survivors of critical illness often have a prolonged and disabling form of cognitive impairment that remains inadequately characterized. We enrolled adults with respiratory failure or shock in the medical or surgical intensive care unit (ICU), evaluated them for in-hospital delirium, and assessed global cognition and executive function 3 and 12 months after discharge with the use of the Repeatable Battery for the Assessment of Neuropsychological Status (population age-adjusted mean [±SD] score, 100±15, with lower values indicating worse global cognition) and the Trail Making Test, Part B (population age-, sex-, and education-adjusted mean score, 50±10, with lower scores indicating worse executive function). Associations of the duration of delirium and the use of sedative or analgesic agents with the outcomes were assessed with the use of linear regression, with adjustment for potential confounders. Of the 821 patients enrolled, 6% had cognitive impairment at baseline, and delirium developed in 74% during the hospital stay. At 3 months, 40% of the patients had global cognition scores that were 1.5 SD below the population means (similar to scores for patients with moderate traumatic brain injury), and 26% had scores 2 SD below the population means (similar to scores for patients with mild Alzheimer's disease). Deficits occurred in both older and younger patients and persisted, with 34% and 24% of all patients with assessments at 12 months that were similar to scores for patients with moderate traumatic brain injury and scores for patients with mild Alzheimer's disease, respectively. A longer duration of delirium was independently associated with worse global cognition at 3 and 12 months (P=0.001 and P=0.04, respectively) and worse executive function at 3 and 12 months (P=0.004 and P=0.007, respectively). Use of sedative or analgesic medications was not consistently associated with cognitive impairment at 3 and 12 months. Patients in medical and surgical ICUs are at high risk for long-term cognitive impairment. A longer duration of delirium in the hospital was associated with worse global cognition and executive function scores at 3 and 12 months. (Funded by the National Institutes of Health and others; BRAIN-ICU ClinicalTrials.gov number, NCT00392795.).
                Bookmark

                Author and article information

                Contributors
                Journal
                JMIR Med Inform
                JMIR Med Inform
                JMI
                JMIR Medical Informatics
                JMIR Publications (Toronto, Canada )
                2291-9694
                July 2021
                26 July 2021
                : 9
                : 7
                : e23401
                Affiliations
                [1 ] Department of Patient Experience Management Part Samsung Medical Center Seoul Republic of Korea
                [2 ] Department of Digital Health SAIHST Sungkyunkwan University Seoul Republic of Korea
                [3 ] Department of Critical Care Medicine and Medicine Samsung Medical Center Sungkyunkwan University School of Medicine Seoul Republic of Korea
                [4 ] Department of Nursing College of Nursing Sahmyook University Seoul Republic of Korea
                [5 ] Department of Computer Science Indiana University Bloomington, IN United States
                [6 ] Department of Emergency Medicine Samsung Medical Center Sungkyunkwan University School of Medicine Seoul Republic of Korea
                [7 ] Digital Innovation Center Samsung Medical Center Seoul Republic of Korea
                Author notes
                Corresponding Author: Chi Ryang Chung chiryang.chung@ 123456gmail.com
                Author information
                https://orcid.org/0000-0003-1335-576X
                https://orcid.org/0000-0003-4945-5623
                https://orcid.org/0000-0003-2331-7644
                https://orcid.org/0000-0002-6596-5116
                https://orcid.org/0000-0002-2778-2992
                https://orcid.org/0000-0003-1830-307X
                Article
                v9i7e23401
                10.2196/23401
                8367129
                34309567
                2eaf2df6-98c3-404e-bb35-608e2ad2f838
                ©Sujeong Hur, Ryoung-Eun Ko, Junsang Yoo, Juhyung Ha, Won Chul Cha, Chi Ryang Chung. Originally published in JMIR Medical Informatics (https://medinform.jmir.org), 26.07.2021.

                This is an open-access article distributed under the terms of the Creative Commons Attribution License ( https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Medical Informatics, is properly cited. The complete bibliographic information, a link to the original publication on https://medinform.jmir.org/, as well as this copyright and license information must be included.

                History
                : 12 August 2020
                : 26 August 2020
                : 10 October 2020
                : 7 June 2021
                Categories
                Original Paper
                Original Paper

                clinical prediction,delirium,electronic health record,intensive care unit,machine learning

                Comments

                Comment on this article