학술논문
Development and validation of risk prediction models for COVID-19 positivity in a hospital setting.
Document Type
Academic Journal
Author
Ng MY; Department of Diagnostic Radiology, The University of Hong Kong, Hong Kong Special Administrative Region; Department of Medical Imaging, The University of Hong Kong-Shenzhen Hospital, Shenzhen, Hong Kong Special Administrative Region. Electronic address: myng2@hku.hk.; Wan EYF; Department of Family Medicine and Primary Care, The University of Hong Kong, Hong Kong Special Administrative Region.; Wong HYF; Department of Radiology, Queen Mary Hospital, Hong Kong Special Administrative Region.; Leung ST; Department of Radiology, Pamela Youde Nethersole Eastern Hospital, Hong Kong Special Administrative Region.; Lee JCY; Department of Radiology and Imaging, Queen Elizabeth Hospital, Hong Kong Special Administrative Region.; Chin TW; Department of Radiology and Imaging, Queen Elizabeth Hospital, Hong Kong Special Administrative Region.; Lo CSY; Department of Radiology, Queen Mary Hospital, Hong Kong Special Administrative Region.; Lui MM; Department of Medicine, Queen Mary Hospital, Hong Kong Special Administrative Region.; Chan EHT; Department of Medical Imaging, The University of Hong Kong-Shenzhen Hospital, Shenzhen, Hong Kong Special Administrative Region.; Fong AH; Department of Diagnostic Radiology, The University of Hong Kong, Hong Kong Special Administrative Region.; Fung SY; Department of Diagnostic Radiology, The University of Hong Kong, Hong Kong Special Administrative Region.; Ching OH; Department of Diagnostic Radiology, The University of Hong Kong, Hong Kong Special Administrative Region.; Chiu KW; Department of Diagnostic Radiology, The University of Hong Kong, Hong Kong Special Administrative Region.; Chung TWH; Department of Microbiology, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Pokfulam, Hong Kong Special Administrative Region.; Vardhanbhuti V; Department of Diagnostic Radiology, The University of Hong Kong, Hong Kong Special Administrative Region.; Lam HYS; Department of Radiology, Queen Mary Hospital, Hong Kong Special Administrative Region.; To KKW; Department of Microbiology, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Pokfulam, Hong Kong Special Administrative Region.; Chiu JLF; Department of Radiology and Imaging, Queen Elizabeth Hospital, Hong Kong Special Administrative Region.; Lam TPW; Department of Radiology, Queen Mary Hospital, Hong Kong Special Administrative Region.; Khong PL; Department of Diagnostic Radiology, The University of Hong Kong, Hong Kong Special Administrative Region.; Liu RWT; Department of Medicine, Ruttonjee Hospital, Hong Kong Special Administrative Region.; Chan JWM; Department of Medicine, Queen Elizabeth Hospital, Hong Kong Special Administrative Region.; Wu AKL; Department of Clinical Pathology, Pamela Youde Nethersole Eastern Hospital, Hong Kong Special Administrative Region.; Lung KC; Department of Medicine, Pamela Youde Nethersole Eastern Hospital, Hong Kong Special Administrative Region.; Hung IFN; Department of Medicine, Queen Mary Hospital, Hong Kong Special Administrative Region; Department of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region.; Lau CS; Department of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region.; Kuo MD; Medical Artificial Intelligence Laboratory (MAIL) Program, Department of Diagnostic Radiology, The University of Hong Kong, Hong Kong Special Administrative Region.; Ip MS; Department of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region; Division of Respiratory & Critical Care Medicine, The University of Hong Kong-Shenzhen Hospital, Shenzhen, China.
Source
Publisher: Elsevier Country of Publication: Canada NLM ID: 9610933 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1878-3511 (Electronic) Linking ISSN: 12019712 NLM ISO Abbreviation: Int J Infect Dis Subsets: MEDLINE
Subject
Language
English
Abstract
Objectives: To develop: (1) two validated risk prediction models for coronavirus disease-2019 (COVID-19) positivity using readily available parameters in a general hospital setting; (2) nomograms and probabilities to allow clinical utilisation.
Methods: Patients with and without COVID-19 were included from 4 Hong Kong hospitals. The database was randomly split into 2:1: for model development database (n = 895) and validation database (n = 435). Multivariable logistic regression was utilised for model creation and validated with the Hosmer-Lemeshow (H-L) test and calibration plot. Nomograms and probabilities set at 0.1, 0.2, 0.4 and 0.6 were calculated to determine sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV).
Results: A total of 1330 patients (mean age 58.2 ± 24.5 years; 50.7% males; 296 COVID-19 positive) were recruited. The first prediction model developed had age, total white blood cell count, chest x-ray appearances and contact history as significant predictors (AUC = 0.911 [CI = 0.880-0.941]). The second model developed has the same variables except contact history (AUC = 0.880 [CI = 0.844-0.916]). Both were externally validated on the H-L test (p = 0.781 and 0.155, respectively) and calibration plot. Models were converted to nomograms. Lower probabilities give higher sensitivity and NPV; higher probabilities give higher specificity and PPV.
Conclusion: Two simple-to-use validated nomograms were developed with excellent AUCs based on readily available parameters and can be considered for clinical utilisation.
(Copyright © 2020 The Author(s). Published by Elsevier Ltd.. All rights reserved.)
Methods: Patients with and without COVID-19 were included from 4 Hong Kong hospitals. The database was randomly split into 2:1: for model development database (n = 895) and validation database (n = 435). Multivariable logistic regression was utilised for model creation and validated with the Hosmer-Lemeshow (H-L) test and calibration plot. Nomograms and probabilities set at 0.1, 0.2, 0.4 and 0.6 were calculated to determine sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV).
Results: A total of 1330 patients (mean age 58.2 ± 24.5 years; 50.7% males; 296 COVID-19 positive) were recruited. The first prediction model developed had age, total white blood cell count, chest x-ray appearances and contact history as significant predictors (AUC = 0.911 [CI = 0.880-0.941]). The second model developed has the same variables except contact history (AUC = 0.880 [CI = 0.844-0.916]). Both were externally validated on the H-L test (p = 0.781 and 0.155, respectively) and calibration plot. Models were converted to nomograms. Lower probabilities give higher sensitivity and NPV; higher probabilities give higher specificity and PPV.
Conclusion: Two simple-to-use validated nomograms were developed with excellent AUCs based on readily available parameters and can be considered for clinical utilisation.
(Copyright © 2020 The Author(s). Published by Elsevier Ltd.. All rights reserved.)