No Cover Image

Journal article 441 views 82 downloads

Equitable hospital length of stay prediction for patients with learning disabilities and multiple long-term conditions using machine learning

Emeka Abakasanga, Rania Kousovista, Georgina Cosma, Ashley Akbari Orcid Logo, Francesco Zaccardi, Navjot Kaur, Danielle Fitt, Gyuchan Thomas Jun, Reza Kiani, Satheesh Gangadharan

Frontiers in Digital Health, Volume: 7

Swansea University Authors: Ashley Akbari Orcid Logo, Danielle Fitt

  • 68890.VoR.pdf

    PDF | Version of Record

    © 2025 Abakasanga, Kousovista, Cosma, Akbari, Zaccardi, Kaur, Fitt, Jun, Kiani and Gangadharan. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY).

    Download (6.7MB)

Check full text

DOI (Published version): 10.3389/fdgth.2025.1538793

Abstract

Purpose: Individuals with learning disabilities (LD) often face higher rates of premature mortality and prolonged hospital stays compared to the general population. Predicting the length of stay (LOS) for patients with LD and multiple long-term conditions (MLTCs) is critical for improving patient ca...

Full description

Published in: Frontiers in Digital Health
ISSN: 2673-253X
Published: Frontiers Media SA 2025
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa68890
first_indexed 2025-02-14T16:01:48Z
last_indexed 2025-03-26T05:31:06Z
id cronfa68890
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2025-03-25T16:14:53.2080763</datestamp><bib-version>v2</bib-version><id>68890</id><entry>2025-02-14</entry><title>Equitable hospital length of stay prediction for patients with learning disabilities and multiple long-term conditions using machine learning</title><swanseaauthors><author><sid>aa1b025ec0243f708bb5eb0a93d6fb52</sid><ORCID>0000-0003-0814-0801</ORCID><firstname>Ashley</firstname><surname>Akbari</surname><name>Ashley Akbari</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>5cc82461379ceaee901c233e6c3a6aa5</sid><firstname>Danielle</firstname><surname>Fitt</surname><name>Danielle Fitt</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2025-02-14</date><deptcode>MEDS</deptcode><abstract>Purpose: Individuals with learning disabilities (LD) often face higher rates of premature mortality and prolonged hospital stays compared to the general population. Predicting the length of stay (LOS) for patients with LD and multiple long-term conditions (MLTCs) is critical for improving patient care and optimising medical resource allocation. However, there is limited research on the application of machine learning (ML) models to this population. Furthermore, approaches designed for the general population often lack generalisability and fairness, particularly when applied across sensitive groups within their cohort.Method: This study analyses hospitalisations of 9,618 patients with LD in Wales using electronic health records (EHR) from the SAIL Databank. A Random Forest (RF) ML model was developed to predict hospital LOS, incorporating demographics, medication history, lifestyle factors, and 39 long-term conditions. To address fairness concerns, two bias mitigation techniques were applied: a post-processing threshold optimiser and an in-processing reductions method using an exponentiated gradient. These methods aimed to minimise performance discrepancies across ethnic groups while ensuring robust model performance.Results: The RF model outperformed other state-of-the-art models, achieving an area under the curve of 0.759 for males and 0.756 for females, a false negative rate of 0.224 for males and 0.229 for females, and a balanced accuracy of 0.690 for males and 0.689 for females. Bias mitigation algorithms reduced disparities in prediction performance across ethnic groups, with the threshold optimiser yielding the most notable improvements. Performance metrics, including false positive rate and balanced accuracy, showed significant enhancements in fairness for the male cohort.Conclusion: This study demonstrates the feasibility of applying ML models to predict LOS for patients with LD and MLTCs, while addressing fairness through bias mitigation techniques. The findings highlight the potential for equitable healthcare predictions using EHR data, paving the way for improved clinical decision-making and resource management.</abstract><type>Journal Article</type><journal>Frontiers in Digital Health</journal><volume>7</volume><journalNumber/><paginationStart/><paginationEnd/><publisher>Frontiers Media SA</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic>2673-253X</issnElectronic><keywords>learning disabilities, length of stay, bias mitigation, threshold optimiser, exponentiated gradient</keywords><publishedDay>14</publishedDay><publishedMonth>2</publishedMonth><publishedYear>2025</publishedYear><publishedDate>2025-02-14</publishedDate><doi>10.3389/fdgth.2025.1538793</doi><url/><notes/><college>COLLEGE NANME</college><department>Medical School</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>MEDS</DepartmentCode><institution>Swansea University</institution><apcterm>Another institution paid the OA fee</apcterm><funders>The authors declare financial support was received for the research, authorship, and/or publication of this article. Data driven machinE-learning aided stratification and management of multiple long-term COnditions in adults with intellectual disabilitiEs (DECODE) project (NIHR203981) is funded by the NIHR AI for Multiple Long-term Conditions (AIM) Programme.</funders><projectreference/><lastEdited>2025-03-25T16:14:53.2080763</lastEdited><Created>2025-02-14T08:47:03.8388379</Created><path><level id="1">Faculty of Medicine, Health and Life Sciences</level><level id="2">Swansea University Medical School - Health Data Science</level></path><authors><author><firstname>Emeka</firstname><surname>Abakasanga</surname><order>1</order></author><author><firstname>Rania</firstname><surname>Kousovista</surname><order>2</order></author><author><firstname>Georgina</firstname><surname>Cosma</surname><order>3</order></author><author><firstname>Ashley</firstname><surname>Akbari</surname><orcid>0000-0003-0814-0801</orcid><order>4</order></author><author><firstname>Francesco</firstname><surname>Zaccardi</surname><order>5</order></author><author><firstname>Navjot</firstname><surname>Kaur</surname><order>6</order></author><author><firstname>Danielle</firstname><surname>Fitt</surname><order>7</order></author><author><firstname>Gyuchan Thomas</firstname><surname>Jun</surname><order>8</order></author><author><firstname>Reza</firstname><surname>Kiani</surname><order>9</order></author><author><firstname>Satheesh</firstname><surname>Gangadharan</surname><order>10</order></author></authors><documents><document><filename>68890__33883__d4e35da98a3a4b878356577b73f7d540.pdf</filename><originalFilename>68890.VoR.pdf</originalFilename><uploaded>2025-03-25T16:13:15.2318044</uploaded><type>Output</type><contentLength>7021143</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>&#xA9; 2025 Abakasanga, Kousovista, Cosma, Akbari, Zaccardi, Kaur, Fitt, Jun, Kiani and Gangadharan. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY).</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>http://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2025-03-25T16:14:53.2080763 v2 68890 2025-02-14 Equitable hospital length of stay prediction for patients with learning disabilities and multiple long-term conditions using machine learning aa1b025ec0243f708bb5eb0a93d6fb52 0000-0003-0814-0801 Ashley Akbari Ashley Akbari true false 5cc82461379ceaee901c233e6c3a6aa5 Danielle Fitt Danielle Fitt true false 2025-02-14 MEDS Purpose: Individuals with learning disabilities (LD) often face higher rates of premature mortality and prolonged hospital stays compared to the general population. Predicting the length of stay (LOS) for patients with LD and multiple long-term conditions (MLTCs) is critical for improving patient care and optimising medical resource allocation. However, there is limited research on the application of machine learning (ML) models to this population. Furthermore, approaches designed for the general population often lack generalisability and fairness, particularly when applied across sensitive groups within their cohort.Method: This study analyses hospitalisations of 9,618 patients with LD in Wales using electronic health records (EHR) from the SAIL Databank. A Random Forest (RF) ML model was developed to predict hospital LOS, incorporating demographics, medication history, lifestyle factors, and 39 long-term conditions. To address fairness concerns, two bias mitigation techniques were applied: a post-processing threshold optimiser and an in-processing reductions method using an exponentiated gradient. These methods aimed to minimise performance discrepancies across ethnic groups while ensuring robust model performance.Results: The RF model outperformed other state-of-the-art models, achieving an area under the curve of 0.759 for males and 0.756 for females, a false negative rate of 0.224 for males and 0.229 for females, and a balanced accuracy of 0.690 for males and 0.689 for females. Bias mitigation algorithms reduced disparities in prediction performance across ethnic groups, with the threshold optimiser yielding the most notable improvements. Performance metrics, including false positive rate and balanced accuracy, showed significant enhancements in fairness for the male cohort.Conclusion: This study demonstrates the feasibility of applying ML models to predict LOS for patients with LD and MLTCs, while addressing fairness through bias mitigation techniques. The findings highlight the potential for equitable healthcare predictions using EHR data, paving the way for improved clinical decision-making and resource management. Journal Article Frontiers in Digital Health 7 Frontiers Media SA 2673-253X learning disabilities, length of stay, bias mitigation, threshold optimiser, exponentiated gradient 14 2 2025 2025-02-14 10.3389/fdgth.2025.1538793 COLLEGE NANME Medical School COLLEGE CODE MEDS Swansea University Another institution paid the OA fee The authors declare financial support was received for the research, authorship, and/or publication of this article. Data driven machinE-learning aided stratification and management of multiple long-term COnditions in adults with intellectual disabilitiEs (DECODE) project (NIHR203981) is funded by the NIHR AI for Multiple Long-term Conditions (AIM) Programme. 2025-03-25T16:14:53.2080763 2025-02-14T08:47:03.8388379 Faculty of Medicine, Health and Life Sciences Swansea University Medical School - Health Data Science Emeka Abakasanga 1 Rania Kousovista 2 Georgina Cosma 3 Ashley Akbari 0000-0003-0814-0801 4 Francesco Zaccardi 5 Navjot Kaur 6 Danielle Fitt 7 Gyuchan Thomas Jun 8 Reza Kiani 9 Satheesh Gangadharan 10 68890__33883__d4e35da98a3a4b878356577b73f7d540.pdf 68890.VoR.pdf 2025-03-25T16:13:15.2318044 Output 7021143 application/pdf Version of Record true © 2025 Abakasanga, Kousovista, Cosma, Akbari, Zaccardi, Kaur, Fitt, Jun, Kiani and Gangadharan. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). true eng http://creativecommons.org/licenses/by/4.0/
title Equitable hospital length of stay prediction for patients with learning disabilities and multiple long-term conditions using machine learning
spellingShingle Equitable hospital length of stay prediction for patients with learning disabilities and multiple long-term conditions using machine learning
Ashley Akbari
Danielle Fitt
title_short Equitable hospital length of stay prediction for patients with learning disabilities and multiple long-term conditions using machine learning
title_full Equitable hospital length of stay prediction for patients with learning disabilities and multiple long-term conditions using machine learning
title_fullStr Equitable hospital length of stay prediction for patients with learning disabilities and multiple long-term conditions using machine learning
title_full_unstemmed Equitable hospital length of stay prediction for patients with learning disabilities and multiple long-term conditions using machine learning
title_sort Equitable hospital length of stay prediction for patients with learning disabilities and multiple long-term conditions using machine learning
author_id_str_mv aa1b025ec0243f708bb5eb0a93d6fb52
5cc82461379ceaee901c233e6c3a6aa5
author_id_fullname_str_mv aa1b025ec0243f708bb5eb0a93d6fb52_***_Ashley Akbari
5cc82461379ceaee901c233e6c3a6aa5_***_Danielle Fitt
author Ashley Akbari
Danielle Fitt
author2 Emeka Abakasanga
Rania Kousovista
Georgina Cosma
Ashley Akbari
Francesco Zaccardi
Navjot Kaur
Danielle Fitt
Gyuchan Thomas Jun
Reza Kiani
Satheesh Gangadharan
format Journal article
container_title Frontiers in Digital Health
container_volume 7
publishDate 2025
institution Swansea University
issn 2673-253X
doi_str_mv 10.3389/fdgth.2025.1538793
publisher Frontiers Media SA
college_str Faculty of Medicine, Health and Life Sciences
hierarchytype
hierarchy_top_id facultyofmedicinehealthandlifesciences
hierarchy_top_title Faculty of Medicine, Health and Life Sciences
hierarchy_parent_id facultyofmedicinehealthandlifesciences
hierarchy_parent_title Faculty of Medicine, Health and Life Sciences
department_str Swansea University Medical School - Health Data Science{{{_:::_}}}Faculty of Medicine, Health and Life Sciences{{{_:::_}}}Swansea University Medical School - Health Data Science
document_store_str 1
active_str 0
description Purpose: Individuals with learning disabilities (LD) often face higher rates of premature mortality and prolonged hospital stays compared to the general population. Predicting the length of stay (LOS) for patients with LD and multiple long-term conditions (MLTCs) is critical for improving patient care and optimising medical resource allocation. However, there is limited research on the application of machine learning (ML) models to this population. Furthermore, approaches designed for the general population often lack generalisability and fairness, particularly when applied across sensitive groups within their cohort.Method: This study analyses hospitalisations of 9,618 patients with LD in Wales using electronic health records (EHR) from the SAIL Databank. A Random Forest (RF) ML model was developed to predict hospital LOS, incorporating demographics, medication history, lifestyle factors, and 39 long-term conditions. To address fairness concerns, two bias mitigation techniques were applied: a post-processing threshold optimiser and an in-processing reductions method using an exponentiated gradient. These methods aimed to minimise performance discrepancies across ethnic groups while ensuring robust model performance.Results: The RF model outperformed other state-of-the-art models, achieving an area under the curve of 0.759 for males and 0.756 for females, a false negative rate of 0.224 for males and 0.229 for females, and a balanced accuracy of 0.690 for males and 0.689 for females. Bias mitigation algorithms reduced disparities in prediction performance across ethnic groups, with the threshold optimiser yielding the most notable improvements. Performance metrics, including false positive rate and balanced accuracy, showed significant enhancements in fairness for the male cohort.Conclusion: This study demonstrates the feasibility of applying ML models to predict LOS for patients with LD and MLTCs, while addressing fairness through bias mitigation techniques. The findings highlight the potential for equitable healthcare predictions using EHR data, paving the way for improved clinical decision-making and resource management.
published_date 2025-02-14T05:27:54Z
_version_ 1856986650189496320
score 11.096068

Similar Items