Machine learning-based scoring system to predict in-hospital outcomes in patients hospitalized with COVID-19.
COVID-19
Prediction
Prognosis
Risk score
SARS-CoV-2
Journal
Archives of cardiovascular diseases
ISSN: 1875-2128
Titre abrégé: Arch Cardiovasc Dis
Pays: Netherlands
ID NLM: 101465655
Informations de publication
Date de publication:
Dec 2022
Dec 2022
Historique:
received:
12
04
2022
revised:
19
07
2022
accepted:
01
08
2022
pubmed:
15
11
2022
medline:
15
12
2022
entrez:
14
11
2022
Statut:
ppublish
Résumé
The evolution of patients hospitalized with coronavirus disease 2019 (COVID-19) is still hard to predict, even after several months of dealing with the pandemic. To develop and validate a score to predict outcomes in patients hospitalized with COVID-19. All consecutive adults hospitalized for COVID-19 from February to April 2020 were included in a nationwide observational study. Primary composite outcome was transfer to an intensive care unit from an emergency department or conventional ward, or in-hospital death. A score that estimates the risk of experiencing the primary outcome was constructed from a derivation cohort using stacked LASSO (Least Absolute Shrinkage and Selection Operator), and was tested in a validation cohort. Among 2873 patients analysed (57.9% men; 66.6±17.0 years), the primary outcome occurred in 838 (29.2%) patients: 551 (19.2%) were transferred to an intensive care unit; and 287 (10.0%) died in-hospital without transfer to an intensive care unit. Using stacked LASSO, we identified 11 variables independently associated with the primary outcome in multivariable analysis in the derivation cohort (n=2313), including demographics (sex), triage vitals (body temperature, dyspnoea, respiratory rate, fraction of inspired oxygen, blood oxygen saturation) and biological variables (pH, platelets, C-reactive protein, aspartate aminotransferase, estimated glomerular filtration rate). The Critical COVID-19 France (CCF) risk score was then developed, and displayed accurate calibration and discrimination in the derivation cohort, with C-statistics of 0.78 (95% confidence interval 0.75-0.80). The CCF risk score performed significantly better (i.e. higher C-statistics) than the usual critical care risk scores. The CCF risk score was built using data collected routinely at hospital admission to predict outcomes in patients with COVID-19. This score holds promise to improve early triage of patients and allocation of healthcare resources.
Sections du résumé
BACKGROUND
BACKGROUND
The evolution of patients hospitalized with coronavirus disease 2019 (COVID-19) is still hard to predict, even after several months of dealing with the pandemic.
AIMS
OBJECTIVE
To develop and validate a score to predict outcomes in patients hospitalized with COVID-19.
METHODS
METHODS
All consecutive adults hospitalized for COVID-19 from February to April 2020 were included in a nationwide observational study. Primary composite outcome was transfer to an intensive care unit from an emergency department or conventional ward, or in-hospital death. A score that estimates the risk of experiencing the primary outcome was constructed from a derivation cohort using stacked LASSO (Least Absolute Shrinkage and Selection Operator), and was tested in a validation cohort.
RESULTS
RESULTS
Among 2873 patients analysed (57.9% men; 66.6±17.0 years), the primary outcome occurred in 838 (29.2%) patients: 551 (19.2%) were transferred to an intensive care unit; and 287 (10.0%) died in-hospital without transfer to an intensive care unit. Using stacked LASSO, we identified 11 variables independently associated with the primary outcome in multivariable analysis in the derivation cohort (n=2313), including demographics (sex), triage vitals (body temperature, dyspnoea, respiratory rate, fraction of inspired oxygen, blood oxygen saturation) and biological variables (pH, platelets, C-reactive protein, aspartate aminotransferase, estimated glomerular filtration rate). The Critical COVID-19 France (CCF) risk score was then developed, and displayed accurate calibration and discrimination in the derivation cohort, with C-statistics of 0.78 (95% confidence interval 0.75-0.80). The CCF risk score performed significantly better (i.e. higher C-statistics) than the usual critical care risk scores.
CONCLUSIONS
CONCLUSIONS
The CCF risk score was built using data collected routinely at hospital admission to predict outcomes in patients with COVID-19. This score holds promise to improve early triage of patients and allocation of healthcare resources.
Identifiants
pubmed: 36376208
pii: S1875-2136(22)00194-2
doi: 10.1016/j.acvd.2022.08.003
pmc: PMC9595484
pii:
doi:
Types de publication
Observational Study
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
617-626Informations de copyright
Copyright © 2022. Published by Elsevier Masson SAS.
Références
Influenza Other Respir Viruses. 2022 Nov;16(6):986-993
pubmed: 35822273
Eur Radiol. 2020 Sep;30(9):4903-4909
pubmed: 32314058
Ann Intern Med. 2015 Jan 6;162(1):55-63
pubmed: 25560714
N Engl J Med. 2020 May 21;382(21):1973-1975
pubmed: 32202721
Lancet. 2020 Jun 6;395(10239):1763-1770
pubmed: 32442528
Rheumatol Int. 2021 Aug;41(8):1375-1386
pubmed: 33903964
BMJ. 2020 May 22;369:m1985
pubmed: 32444460
Arch Cardiovasc Dis. 2021 May;114(5):352-363
pubmed: 34154953
JAMA. 2020 Apr 7;323(13):1239-1242
pubmed: 32091533
J Comput Graph Stat. 2022;31(4):1063-1075
pubmed: 36644406
Nature. 2020 Mar;579(7798):270-273
pubmed: 32015507
N Engl J Med. 2021 Feb 25;384(8):693-704
pubmed: 32678530
Diabetes Metab. 2021 Jul;47(4):101222
pubmed: 33388386
Lancet. 2020 Feb 22;395(10224):565-574
pubmed: 32007145
JAMA. 2020 Apr 28;323(16):1574-1581
pubmed: 32250385
Nat Rev Endocrinol. 2020 Jul;16(7):341-342
pubmed: 32327737
N Engl J Med. 2021 Apr 22;384(16):1576-1577
pubmed: 33596348
Euro Surveill. 2022 Mar;27(9):
pubmed: 35241215
N Engl J Med. 2020 May 21;382(21):2049-2055
pubmed: 32202722
Nature. 2021 Jan;589(7843):500-501
pubmed: 33479534
N Engl J Med. 2020 Apr 30;382(18):1708-1720
pubmed: 32109013
JAMA Intern Med. 2020 Aug 1;180(8):1081-1089
pubmed: 32396163
JAMA. 2020 May 26;323(20):2052-2059
pubmed: 32320003
Clin Infect Dis. 2021 Jan 23;72(1):182-183
pubmed: 32474605
BMJ. 2007 Oct 20;335(7624):806-8
pubmed: 17947786
Arch Cardiovasc Dis. 2021 May;114(5):394-406
pubmed: 34154954
Lancet. 2022 Jan 29;399(10323):437-446
pubmed: 35065011
N Engl J Med. 2020 Apr 23;382(17):1583-1586
pubmed: 32320568
Diabetes Res Clin Pract. 2020 Apr;162:108142
pubmed: 32278764
Scand J Trauma Resusc Emerg Med. 2020 Jul 13;28(1):66
pubmed: 32660623
Stat Med. 2018 Jun 30;37(14):2252-2266
pubmed: 29682776
Radiology. 2021 Jun;299(3):E262-E279
pubmed: 33560192
Nature. 2021 Dec;600(7890):577-578
pubmed: 34934198
Lancet Infect Dis. 2020 May;20(5):533-534
pubmed: 32087114
Lancet Reg Health Eur. 2021 Jun;5:100104
pubmed: 33969337
Nature. 2021 Jan;589(7841):177-178
pubmed: 33432212
Lancet. 2020 Feb 15;395(10223):497-506
pubmed: 31986264
JAMA Cardiol. 2020 Sep 1;5(9):1020-1026
pubmed: 32936273
JAMA. 2020 May 12;323(18):1824-1836
pubmed: 32282022
JAMA. 2020 Jun 9;323(22):2262-2263
pubmed: 32364561
BMJ Open. 2020 Oct 6;10(10):e040129
pubmed: 33028563
JAMA. 2020 Jun 2;323(21):2195-2198
pubmed: 32329797