Contrasting factors associated with COVID-19-related ICU admission and death outcomes in hospitalised patients by means of Shapley values.

Identification of those at greatest risk of death due to the substantial threat of COVID-19 can benefit from novel approaches to epidemiology that leverage large datasets and complex machine-learning models, provide data-driven intelligence, and guide decisions such as intensive-care unit admission...

Full description

Saved in:
Bibliographic Details
Main Authors: Massimo Cavallaro, Haseeb Moiz, Matt J Keeling, Noel D McCarthy
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2021-06-01
Series:PLoS Computational Biology
Online Access:https://journals.plos.org/ploscompbiol/article/file?id=10.1371/journal.pcbi.1009121&type=printable
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849470042768408576
author Massimo Cavallaro
Haseeb Moiz
Matt J Keeling
Noel D McCarthy
author_facet Massimo Cavallaro
Haseeb Moiz
Matt J Keeling
Noel D McCarthy
author_sort Massimo Cavallaro
collection DOAJ
description Identification of those at greatest risk of death due to the substantial threat of COVID-19 can benefit from novel approaches to epidemiology that leverage large datasets and complex machine-learning models, provide data-driven intelligence, and guide decisions such as intensive-care unit admission (ICUA). The objective of this study is two-fold, one substantive and one methodological: substantively to evaluate the association of demographic and health records with two related, yet different, outcomes of severe COVID-19 (viz., death and ICUA); methodologically to compare interpretations based on logistic regression and on gradient-boosted decision tree (GBDT) predictions interpreted by means of the Shapley impacts of covariates. Very different association of some factors, e.g., obesity and chronic respiratory diseases, with death and ICUA may guide review of practice. Shapley explanation of GBDTs identified varying effects of some factors among patients, thus emphasising the importance of individual patient assessment. The results of this study are also relevant for the evaluation of complex automated clinical decision systems, which should optimise prediction scores whilst remaining interpretable to clinicians and mitigating potential biases.
format Article
id doaj-art-b1a0dedeac67451497a4de5ca7c71202
institution Kabale University
issn 1553-734X
1553-7358
language English
publishDate 2021-06-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS Computational Biology
spelling doaj-art-b1a0dedeac67451497a4de5ca7c712022025-08-20T03:25:16ZengPublic Library of Science (PLoS)PLoS Computational Biology1553-734X1553-73582021-06-01176e100912110.1371/journal.pcbi.1009121Contrasting factors associated with COVID-19-related ICU admission and death outcomes in hospitalised patients by means of Shapley values.Massimo CavallaroHaseeb MoizMatt J KeelingNoel D McCarthyIdentification of those at greatest risk of death due to the substantial threat of COVID-19 can benefit from novel approaches to epidemiology that leverage large datasets and complex machine-learning models, provide data-driven intelligence, and guide decisions such as intensive-care unit admission (ICUA). The objective of this study is two-fold, one substantive and one methodological: substantively to evaluate the association of demographic and health records with two related, yet different, outcomes of severe COVID-19 (viz., death and ICUA); methodologically to compare interpretations based on logistic regression and on gradient-boosted decision tree (GBDT) predictions interpreted by means of the Shapley impacts of covariates. Very different association of some factors, e.g., obesity and chronic respiratory diseases, with death and ICUA may guide review of practice. Shapley explanation of GBDTs identified varying effects of some factors among patients, thus emphasising the importance of individual patient assessment. The results of this study are also relevant for the evaluation of complex automated clinical decision systems, which should optimise prediction scores whilst remaining interpretable to clinicians and mitigating potential biases.https://journals.plos.org/ploscompbiol/article/file?id=10.1371/journal.pcbi.1009121&type=printable
spellingShingle Massimo Cavallaro
Haseeb Moiz
Matt J Keeling
Noel D McCarthy
Contrasting factors associated with COVID-19-related ICU admission and death outcomes in hospitalised patients by means of Shapley values.
PLoS Computational Biology
title Contrasting factors associated with COVID-19-related ICU admission and death outcomes in hospitalised patients by means of Shapley values.
title_full Contrasting factors associated with COVID-19-related ICU admission and death outcomes in hospitalised patients by means of Shapley values.
title_fullStr Contrasting factors associated with COVID-19-related ICU admission and death outcomes in hospitalised patients by means of Shapley values.
title_full_unstemmed Contrasting factors associated with COVID-19-related ICU admission and death outcomes in hospitalised patients by means of Shapley values.
title_short Contrasting factors associated with COVID-19-related ICU admission and death outcomes in hospitalised patients by means of Shapley values.
title_sort contrasting factors associated with covid 19 related icu admission and death outcomes in hospitalised patients by means of shapley values
url https://journals.plos.org/ploscompbiol/article/file?id=10.1371/journal.pcbi.1009121&type=printable
work_keys_str_mv AT massimocavallaro contrastingfactorsassociatedwithcovid19relatedicuadmissionanddeathoutcomesinhospitalisedpatientsbymeansofshapleyvalues
AT haseebmoiz contrastingfactorsassociatedwithcovid19relatedicuadmissionanddeathoutcomesinhospitalisedpatientsbymeansofshapleyvalues
AT mattjkeeling contrastingfactorsassociatedwithcovid19relatedicuadmissionanddeathoutcomesinhospitalisedpatientsbymeansofshapleyvalues
AT noeldmccarthy contrastingfactorsassociatedwithcovid19relatedicuadmissionanddeathoutcomesinhospitalisedpatientsbymeansofshapleyvalues