A predictive model of clinical deterioration among hospitalized COVID-19 patients by harnessing hospital course trajectories
[Display omitted] •Predicting deterioration in hospitalized COVID positive patients.•Harnessing complex laboratory and vital sign values from the electronic medical record.•Time-dependent cross-validation.•Black box and machine learning methods for prediction. From early March through mid-May 2020,...
Gespeichert in:
Veröffentlicht in: | Journal of biomedical informatics 2021-06, Vol.118, p.103794-103794, Article 103794 |
---|---|
Hauptverfasser: | , , , , , , , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 103794 |
---|---|
container_issue | |
container_start_page | 103794 |
container_title | Journal of biomedical informatics |
container_volume | 118 |
creator | Mauer, Elizabeth Lee, Jihui Choi, Justin Zhang, Hongzhe Hoffman, Katherine L. Easthausen, Imaani J. Rajan, Mangala Weiner, Mark G. Kaushal, Rainu Safford, Monika M. Steel, Peter A.D. Banerjee, Samprit |
description | [Display omitted]
•Predicting deterioration in hospitalized COVID positive patients.•Harnessing complex laboratory and vital sign values from the electronic medical record.•Time-dependent cross-validation.•Black box and machine learning methods for prediction.
From early March through mid-May 2020, the COVID-19 pandemic overwhelmed hospitals in New York City. In anticipation of ventilator shortages and limited ICU bed capacity, hospital operations prioritized the development of prognostic tools to predict clinical deterioration. However, early experience from frontline physicians observed that some patients developed unanticipated deterioration after having relatively stable periods, attesting to the uncertainty of clinical trajectories among hospitalized patients with COVID-19. Prediction tools that incorporate clinical variables at one time-point, usually on hospital presentation, are suboptimal for patients with dynamic changes and evolving clinical trajectories. Therefore, our study team developed a machine-learning algorithm to predict clinical deterioration among hospitalized COVID-19 patients by extracting clinically meaningful features from complex longitudinal laboratory and vital sign values during the early period of hospitalization with an emphasis on informative missing-ness. To incorporate the evolution of the disease and clinical practice over the course of the pandemic, we utilized a time-dependent cross-validation strategy for model development. Finally, we validated our prediction model on an external validation cohort of COVID-19 patients served in a demographically distinct population from the training cohort. The main finding of our study is the identification of risk profiles of early, late and no clinical deterioration during the course of hospitalization. While risk prediction models that include simple predictors at ED presentation and clinical judgement are able to identify any deterioration vs. no deterioration, our methodology is able to isolate a particular risk group that remain stable initially but deteriorate at a later stage of the course of hospitalization. We demonstrate the superior predictive performance with the utilization of laboratory and vital sign data during the early period of hospitalization compared to the utilization of data at presentation alone. Our results will allow efficient hospital resource allocation and will motivate research in understanding the late deterioration risk group. |
doi_str_mv | 10.1016/j.jbi.2021.103794 |
format | Article |
fullrecord | <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_8084618</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S1532046421001234</els_id><sourcerecordid>2521498303</sourcerecordid><originalsourceid>FETCH-LOGICAL-c451t-3fd53e431cdee4b329713b927ac9f014fd488a9ffaec86d848760d213da8b4fb3</originalsourceid><addsrcrecordid>eNp9kU1r3DAQhkVpadK0P6CXomMv3kqW7JUpFML2KxDIpclVyNIoO8a2XEm7kNIfX4VNl-SSkyT0zDvDPIS852zFGW8_Dauhx1XNal7eYt3JF-SUN6KumFTs5fHeyhPyJqWBMc6bpn1NToTohGgbeUr-ntMlgkObcQ90Cg5GGjy1I85ozUgdZIgYoskYZmqmMN_SbUgLZjPiH3B0c3Vz8bXiHV0KAnNOtL-jWxNnSAkfwdSGXUxAczQD2BwiQnpLXnkzJnj3cJ6R6-_ffm1-VpdXPy4255eVlQ3PlfCuESAFtw5A9qLu1lz0Xb02tvOMS--kUqbz3oBVrVNSrVvmai6cUb30vTgjXw65y66fwNkyZjSjXiJOJt7pYFA__Zlxq2_DXiumZMtVCfj4EBDD7x2krCdMFsbRzBB2SddNzWWnBBMF5QfUxpBSBH9sw5m-t6YHXazpe2v6YK3UfHg837Hiv6YCfD4AULa0R4g62bJsW8TFskztAj4T_w8caauS</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2521498303</pqid></control><display><type>article</type><title>A predictive model of clinical deterioration among hospitalized COVID-19 patients by harnessing hospital course trajectories</title><source>MEDLINE</source><source>Elsevier ScienceDirect Journals Complete</source><source>EZB-FREE-00999 freely available EZB journals</source><creator>Mauer, Elizabeth ; Lee, Jihui ; Choi, Justin ; Zhang, Hongzhe ; Hoffman, Katherine L. ; Easthausen, Imaani J. ; Rajan, Mangala ; Weiner, Mark G. ; Kaushal, Rainu ; Safford, Monika M. ; Steel, Peter A.D. ; Banerjee, Samprit</creator><creatorcontrib>Mauer, Elizabeth ; Lee, Jihui ; Choi, Justin ; Zhang, Hongzhe ; Hoffman, Katherine L. ; Easthausen, Imaani J. ; Rajan, Mangala ; Weiner, Mark G. ; Kaushal, Rainu ; Safford, Monika M. ; Steel, Peter A.D. ; Banerjee, Samprit</creatorcontrib><description>[Display omitted]
•Predicting deterioration in hospitalized COVID positive patients.•Harnessing complex laboratory and vital sign values from the electronic medical record.•Time-dependent cross-validation.•Black box and machine learning methods for prediction.
From early March through mid-May 2020, the COVID-19 pandemic overwhelmed hospitals in New York City. In anticipation of ventilator shortages and limited ICU bed capacity, hospital operations prioritized the development of prognostic tools to predict clinical deterioration. However, early experience from frontline physicians observed that some patients developed unanticipated deterioration after having relatively stable periods, attesting to the uncertainty of clinical trajectories among hospitalized patients with COVID-19. Prediction tools that incorporate clinical variables at one time-point, usually on hospital presentation, are suboptimal for patients with dynamic changes and evolving clinical trajectories. Therefore, our study team developed a machine-learning algorithm to predict clinical deterioration among hospitalized COVID-19 patients by extracting clinically meaningful features from complex longitudinal laboratory and vital sign values during the early period of hospitalization with an emphasis on informative missing-ness. To incorporate the evolution of the disease and clinical practice over the course of the pandemic, we utilized a time-dependent cross-validation strategy for model development. Finally, we validated our prediction model on an external validation cohort of COVID-19 patients served in a demographically distinct population from the training cohort. The main finding of our study is the identification of risk profiles of early, late and no clinical deterioration during the course of hospitalization. While risk prediction models that include simple predictors at ED presentation and clinical judgement are able to identify any deterioration vs. no deterioration, our methodology is able to isolate a particular risk group that remain stable initially but deteriorate at a later stage of the course of hospitalization. We demonstrate the superior predictive performance with the utilization of laboratory and vital sign data during the early period of hospitalization compared to the utilization of data at presentation alone. Our results will allow efficient hospital resource allocation and will motivate research in understanding the late deterioration risk group.</description><identifier>ISSN: 1532-0464</identifier><identifier>EISSN: 1532-0480</identifier><identifier>DOI: 10.1016/j.jbi.2021.103794</identifier><identifier>PMID: 33933654</identifier><language>eng</language><publisher>United States: Elsevier Inc</publisher><subject>Aged ; Clinical Deterioration ; Computer Simulation ; COVID-19 ; COVID-19 - diagnosis ; Deterioration ; EMR ; Female ; Hospitalization ; Hospitals ; Humans ; Intubation ; Machine learning ; Male ; New York City ; Original Research ; Pandemics ; Prediction ; Retrospective Studies ; Risk Assessment ; ROC Curve</subject><ispartof>Journal of biomedical informatics, 2021-06, Vol.118, p.103794-103794, Article 103794</ispartof><rights>2021 Elsevier Inc.</rights><rights>Copyright © 2021 Elsevier Inc. All rights reserved.</rights><rights>2021 Elsevier Inc. 2021 Elsevier Inc.</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c451t-3fd53e431cdee4b329713b927ac9f014fd488a9ffaec86d848760d213da8b4fb3</citedby><cites>FETCH-LOGICAL-c451t-3fd53e431cdee4b329713b927ac9f014fd488a9ffaec86d848760d213da8b4fb3</cites><orcidid>0000-0001-6936-0846 ; 0000-0001-5586-9940</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://dx.doi.org/10.1016/j.jbi.2021.103794$$EHTML$$P50$$Gelsevier$$Hfree_for_read</linktohtml><link.rule.ids>230,314,780,784,885,3550,27924,27925,45995</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/33933654$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Mauer, Elizabeth</creatorcontrib><creatorcontrib>Lee, Jihui</creatorcontrib><creatorcontrib>Choi, Justin</creatorcontrib><creatorcontrib>Zhang, Hongzhe</creatorcontrib><creatorcontrib>Hoffman, Katherine L.</creatorcontrib><creatorcontrib>Easthausen, Imaani J.</creatorcontrib><creatorcontrib>Rajan, Mangala</creatorcontrib><creatorcontrib>Weiner, Mark G.</creatorcontrib><creatorcontrib>Kaushal, Rainu</creatorcontrib><creatorcontrib>Safford, Monika M.</creatorcontrib><creatorcontrib>Steel, Peter A.D.</creatorcontrib><creatorcontrib>Banerjee, Samprit</creatorcontrib><title>A predictive model of clinical deterioration among hospitalized COVID-19 patients by harnessing hospital course trajectories</title><title>Journal of biomedical informatics</title><addtitle>J Biomed Inform</addtitle><description>[Display omitted]
•Predicting deterioration in hospitalized COVID positive patients.•Harnessing complex laboratory and vital sign values from the electronic medical record.•Time-dependent cross-validation.•Black box and machine learning methods for prediction.
From early March through mid-May 2020, the COVID-19 pandemic overwhelmed hospitals in New York City. In anticipation of ventilator shortages and limited ICU bed capacity, hospital operations prioritized the development of prognostic tools to predict clinical deterioration. However, early experience from frontline physicians observed that some patients developed unanticipated deterioration after having relatively stable periods, attesting to the uncertainty of clinical trajectories among hospitalized patients with COVID-19. Prediction tools that incorporate clinical variables at one time-point, usually on hospital presentation, are suboptimal for patients with dynamic changes and evolving clinical trajectories. Therefore, our study team developed a machine-learning algorithm to predict clinical deterioration among hospitalized COVID-19 patients by extracting clinically meaningful features from complex longitudinal laboratory and vital sign values during the early period of hospitalization with an emphasis on informative missing-ness. To incorporate the evolution of the disease and clinical practice over the course of the pandemic, we utilized a time-dependent cross-validation strategy for model development. Finally, we validated our prediction model on an external validation cohort of COVID-19 patients served in a demographically distinct population from the training cohort. The main finding of our study is the identification of risk profiles of early, late and no clinical deterioration during the course of hospitalization. While risk prediction models that include simple predictors at ED presentation and clinical judgement are able to identify any deterioration vs. no deterioration, our methodology is able to isolate a particular risk group that remain stable initially but deteriorate at a later stage of the course of hospitalization. We demonstrate the superior predictive performance with the utilization of laboratory and vital sign data during the early period of hospitalization compared to the utilization of data at presentation alone. Our results will allow efficient hospital resource allocation and will motivate research in understanding the late deterioration risk group.</description><subject>Aged</subject><subject>Clinical Deterioration</subject><subject>Computer Simulation</subject><subject>COVID-19</subject><subject>COVID-19 - diagnosis</subject><subject>Deterioration</subject><subject>EMR</subject><subject>Female</subject><subject>Hospitalization</subject><subject>Hospitals</subject><subject>Humans</subject><subject>Intubation</subject><subject>Machine learning</subject><subject>Male</subject><subject>New York City</subject><subject>Original Research</subject><subject>Pandemics</subject><subject>Prediction</subject><subject>Retrospective Studies</subject><subject>Risk Assessment</subject><subject>ROC Curve</subject><issn>1532-0464</issn><issn>1532-0480</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>EIF</sourceid><recordid>eNp9kU1r3DAQhkVpadK0P6CXomMv3kqW7JUpFML2KxDIpclVyNIoO8a2XEm7kNIfX4VNl-SSkyT0zDvDPIS852zFGW8_Dauhx1XNal7eYt3JF-SUN6KumFTs5fHeyhPyJqWBMc6bpn1NToTohGgbeUr-ntMlgkObcQ90Cg5GGjy1I85ozUgdZIgYoskYZmqmMN_SbUgLZjPiH3B0c3Vz8bXiHV0KAnNOtL-jWxNnSAkfwdSGXUxAczQD2BwiQnpLXnkzJnj3cJ6R6-_ffm1-VpdXPy4255eVlQ3PlfCuESAFtw5A9qLu1lz0Xb02tvOMS--kUqbz3oBVrVNSrVvmai6cUb30vTgjXw65y66fwNkyZjSjXiJOJt7pYFA__Zlxq2_DXiumZMtVCfj4EBDD7x2krCdMFsbRzBB2SddNzWWnBBMF5QfUxpBSBH9sw5m-t6YHXazpe2v6YK3UfHg837Hiv6YCfD4AULa0R4g62bJsW8TFskztAj4T_w8caauS</recordid><startdate>20210601</startdate><enddate>20210601</enddate><creator>Mauer, Elizabeth</creator><creator>Lee, Jihui</creator><creator>Choi, Justin</creator><creator>Zhang, Hongzhe</creator><creator>Hoffman, Katherine L.</creator><creator>Easthausen, Imaani J.</creator><creator>Rajan, Mangala</creator><creator>Weiner, Mark G.</creator><creator>Kaushal, Rainu</creator><creator>Safford, Monika M.</creator><creator>Steel, Peter A.D.</creator><creator>Banerjee, Samprit</creator><general>Elsevier Inc</general><scope>6I.</scope><scope>AAFTH</scope><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0001-6936-0846</orcidid><orcidid>https://orcid.org/0000-0001-5586-9940</orcidid></search><sort><creationdate>20210601</creationdate><title>A predictive model of clinical deterioration among hospitalized COVID-19 patients by harnessing hospital course trajectories</title><author>Mauer, Elizabeth ; Lee, Jihui ; Choi, Justin ; Zhang, Hongzhe ; Hoffman, Katherine L. ; Easthausen, Imaani J. ; Rajan, Mangala ; Weiner, Mark G. ; Kaushal, Rainu ; Safford, Monika M. ; Steel, Peter A.D. ; Banerjee, Samprit</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c451t-3fd53e431cdee4b329713b927ac9f014fd488a9ffaec86d848760d213da8b4fb3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Aged</topic><topic>Clinical Deterioration</topic><topic>Computer Simulation</topic><topic>COVID-19</topic><topic>COVID-19 - diagnosis</topic><topic>Deterioration</topic><topic>EMR</topic><topic>Female</topic><topic>Hospitalization</topic><topic>Hospitals</topic><topic>Humans</topic><topic>Intubation</topic><topic>Machine learning</topic><topic>Male</topic><topic>New York City</topic><topic>Original Research</topic><topic>Pandemics</topic><topic>Prediction</topic><topic>Retrospective Studies</topic><topic>Risk Assessment</topic><topic>ROC Curve</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Mauer, Elizabeth</creatorcontrib><creatorcontrib>Lee, Jihui</creatorcontrib><creatorcontrib>Choi, Justin</creatorcontrib><creatorcontrib>Zhang, Hongzhe</creatorcontrib><creatorcontrib>Hoffman, Katherine L.</creatorcontrib><creatorcontrib>Easthausen, Imaani J.</creatorcontrib><creatorcontrib>Rajan, Mangala</creatorcontrib><creatorcontrib>Weiner, Mark G.</creatorcontrib><creatorcontrib>Kaushal, Rainu</creatorcontrib><creatorcontrib>Safford, Monika M.</creatorcontrib><creatorcontrib>Steel, Peter A.D.</creatorcontrib><creatorcontrib>Banerjee, Samprit</creatorcontrib><collection>ScienceDirect Open Access Titles</collection><collection>Elsevier:ScienceDirect:Open Access</collection><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>Journal of biomedical informatics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Mauer, Elizabeth</au><au>Lee, Jihui</au><au>Choi, Justin</au><au>Zhang, Hongzhe</au><au>Hoffman, Katherine L.</au><au>Easthausen, Imaani J.</au><au>Rajan, Mangala</au><au>Weiner, Mark G.</au><au>Kaushal, Rainu</au><au>Safford, Monika M.</au><au>Steel, Peter A.D.</au><au>Banerjee, Samprit</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A predictive model of clinical deterioration among hospitalized COVID-19 patients by harnessing hospital course trajectories</atitle><jtitle>Journal of biomedical informatics</jtitle><addtitle>J Biomed Inform</addtitle><date>2021-06-01</date><risdate>2021</risdate><volume>118</volume><spage>103794</spage><epage>103794</epage><pages>103794-103794</pages><artnum>103794</artnum><issn>1532-0464</issn><eissn>1532-0480</eissn><abstract>[Display omitted]
•Predicting deterioration in hospitalized COVID positive patients.•Harnessing complex laboratory and vital sign values from the electronic medical record.•Time-dependent cross-validation.•Black box and machine learning methods for prediction.
From early March through mid-May 2020, the COVID-19 pandemic overwhelmed hospitals in New York City. In anticipation of ventilator shortages and limited ICU bed capacity, hospital operations prioritized the development of prognostic tools to predict clinical deterioration. However, early experience from frontline physicians observed that some patients developed unanticipated deterioration after having relatively stable periods, attesting to the uncertainty of clinical trajectories among hospitalized patients with COVID-19. Prediction tools that incorporate clinical variables at one time-point, usually on hospital presentation, are suboptimal for patients with dynamic changes and evolving clinical trajectories. Therefore, our study team developed a machine-learning algorithm to predict clinical deterioration among hospitalized COVID-19 patients by extracting clinically meaningful features from complex longitudinal laboratory and vital sign values during the early period of hospitalization with an emphasis on informative missing-ness. To incorporate the evolution of the disease and clinical practice over the course of the pandemic, we utilized a time-dependent cross-validation strategy for model development. Finally, we validated our prediction model on an external validation cohort of COVID-19 patients served in a demographically distinct population from the training cohort. The main finding of our study is the identification of risk profiles of early, late and no clinical deterioration during the course of hospitalization. While risk prediction models that include simple predictors at ED presentation and clinical judgement are able to identify any deterioration vs. no deterioration, our methodology is able to isolate a particular risk group that remain stable initially but deteriorate at a later stage of the course of hospitalization. We demonstrate the superior predictive performance with the utilization of laboratory and vital sign data during the early period of hospitalization compared to the utilization of data at presentation alone. Our results will allow efficient hospital resource allocation and will motivate research in understanding the late deterioration risk group.</abstract><cop>United States</cop><pub>Elsevier Inc</pub><pmid>33933654</pmid><doi>10.1016/j.jbi.2021.103794</doi><tpages>1</tpages><orcidid>https://orcid.org/0000-0001-6936-0846</orcidid><orcidid>https://orcid.org/0000-0001-5586-9940</orcidid><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 1532-0464 |
ispartof | Journal of biomedical informatics, 2021-06, Vol.118, p.103794-103794, Article 103794 |
issn | 1532-0464 1532-0480 |
language | eng |
recordid | cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_8084618 |
source | MEDLINE; Elsevier ScienceDirect Journals Complete; EZB-FREE-00999 freely available EZB journals |
subjects | Aged Clinical Deterioration Computer Simulation COVID-19 COVID-19 - diagnosis Deterioration EMR Female Hospitalization Hospitals Humans Intubation Machine learning Male New York City Original Research Pandemics Prediction Retrospective Studies Risk Assessment ROC Curve |
title | A predictive model of clinical deterioration among hospitalized COVID-19 patients by harnessing hospital course trajectories |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T11%3A59%3A34IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20predictive%20model%20of%20clinical%20deterioration%20among%20hospitalized%20COVID-19%20patients%20by%20harnessing%20hospital%20course%20trajectories&rft.jtitle=Journal%20of%20biomedical%20informatics&rft.au=Mauer,%20Elizabeth&rft.date=2021-06-01&rft.volume=118&rft.spage=103794&rft.epage=103794&rft.pages=103794-103794&rft.artnum=103794&rft.issn=1532-0464&rft.eissn=1532-0480&rft_id=info:doi/10.1016/j.jbi.2021.103794&rft_dat=%3Cproquest_pubme%3E2521498303%3C/proquest_pubme%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2521498303&rft_id=info:pmid/33933654&rft_els_id=S1532046421001234&rfr_iscdi=true |