Developing interpretable machine learning models to predict length of stay and disposition decision for adult patients in emergency departments

Objective Machine learning (ML) models have emerged as tools to predict length of stay (LOS) and disposition decision (DD) in emergency departments (EDs) to combat overcrowding. However, site-specific ML models are not transferable to different sites. Our objective was to develop interpretable ML mo...

Full description

Saved in:
Bibliographic Details
Main Authors: Abhishek Sharma, Timothy N Fazio, Long Song, Samantha Plumb, Uwe Aickelin, Mojgan Kouhounestani, Mark John Putland
Format: Article
Language:English
Published: BMJ Publishing Group 2025-06-01
Series:BMJ Health & Care Informatics
Online Access:https://informatics.bmj.com/content/32/1/e101152.full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849423715701358592
author Abhishek Sharma
Timothy N Fazio
Long Song
Samantha Plumb
Uwe Aickelin
Mojgan Kouhounestani
Mark John Putland
author_facet Abhishek Sharma
Timothy N Fazio
Long Song
Samantha Plumb
Uwe Aickelin
Mojgan Kouhounestani
Mark John Putland
author_sort Abhishek Sharma
collection DOAJ
description Objective Machine learning (ML) models have emerged as tools to predict length of stay (LOS) and disposition decision (DD) in emergency departments (EDs) to combat overcrowding. However, site-specific ML models are not transferable to different sites. Our objective was to develop interpretable ML models to predict LOS and DD at specific time points, all while establishing a transparent data analysis framework. This framework was designed to be easily adapted by other institutions for the development of their own ML models.Methods We analysed data from 297 392 ED visits of patients aged 18 and above at a quaternary hospital between 30 June 2019 and 31 December 2022. Eight ML algorithms were evaluated, and ultimately, twelve lasso models built from 21 features were trained to predict four outcomes of LOS and DD at three time points post-triage. Hold-out testing and cross-validation were conducted for these models.Results The area under the curve values were 0.862/0.868/0.878 for binary LOS predictions at 10, 60 and 120-minute time points and 0.839/0.851/0.863 for binary DD predictions. The accuracies were 60.2%/60.7%/61.9% for ternary LOS predictions and 61.5%/62.3%/63.4% for ternary DD predictions.Conclusions Interpretable ML models demonstrated outstanding performances in predicting both LOS and DD. The transparent data analysis framework can be easily adapted by other institutions.
format Article
id doaj-art-98009c68c1ff4d83a053d604c85fdcf8
institution Kabale University
issn 2632-1009
language English
publishDate 2025-06-01
publisher BMJ Publishing Group
record_format Article
series BMJ Health & Care Informatics
spelling doaj-art-98009c68c1ff4d83a053d604c85fdcf82025-08-20T03:30:31ZengBMJ Publishing GroupBMJ Health & Care Informatics2632-10092025-06-0132110.1136/bmjhci-2024-101152Developing interpretable machine learning models to predict length of stay and disposition decision for adult patients in emergency departmentsAbhishek Sharma0Timothy N Fazio1Long Song2Samantha Plumb3Uwe Aickelin4Mojgan Kouhounestani5Mark John Putland6Department of Global Health, Boston University School of Public Health, Boston, Massachusetts, USAHealth Intelligence Unit, The Royal Melbourne Hospital, Parkville, Victoria, AustraliaSchool of Public Health of Southeast University, Nanjing, Jiangsu 210009, ChinaRoyal Melbourne Hospital, Parkville, Victoria, AustraliaSchool of Computing and Information Systems, The University of Melbourne, Melbourne, Victoria, AustraliaSchool of Computing and Information Systems, The University of Melbourne, Melbourne, Victoria, AustraliaDepartment of Emergency Medicine, The Royal Melbourne Hospital, Parkville, Victoria, AustraliaObjective Machine learning (ML) models have emerged as tools to predict length of stay (LOS) and disposition decision (DD) in emergency departments (EDs) to combat overcrowding. However, site-specific ML models are not transferable to different sites. Our objective was to develop interpretable ML models to predict LOS and DD at specific time points, all while establishing a transparent data analysis framework. This framework was designed to be easily adapted by other institutions for the development of their own ML models.Methods We analysed data from 297 392 ED visits of patients aged 18 and above at a quaternary hospital between 30 June 2019 and 31 December 2022. Eight ML algorithms were evaluated, and ultimately, twelve lasso models built from 21 features were trained to predict four outcomes of LOS and DD at three time points post-triage. Hold-out testing and cross-validation were conducted for these models.Results The area under the curve values were 0.862/0.868/0.878 for binary LOS predictions at 10, 60 and 120-minute time points and 0.839/0.851/0.863 for binary DD predictions. The accuracies were 60.2%/60.7%/61.9% for ternary LOS predictions and 61.5%/62.3%/63.4% for ternary DD predictions.Conclusions Interpretable ML models demonstrated outstanding performances in predicting both LOS and DD. The transparent data analysis framework can be easily adapted by other institutions.https://informatics.bmj.com/content/32/1/e101152.full
spellingShingle Abhishek Sharma
Timothy N Fazio
Long Song
Samantha Plumb
Uwe Aickelin
Mojgan Kouhounestani
Mark John Putland
Developing interpretable machine learning models to predict length of stay and disposition decision for adult patients in emergency departments
BMJ Health & Care Informatics
title Developing interpretable machine learning models to predict length of stay and disposition decision for adult patients in emergency departments
title_full Developing interpretable machine learning models to predict length of stay and disposition decision for adult patients in emergency departments
title_fullStr Developing interpretable machine learning models to predict length of stay and disposition decision for adult patients in emergency departments
title_full_unstemmed Developing interpretable machine learning models to predict length of stay and disposition decision for adult patients in emergency departments
title_short Developing interpretable machine learning models to predict length of stay and disposition decision for adult patients in emergency departments
title_sort developing interpretable machine learning models to predict length of stay and disposition decision for adult patients in emergency departments
url https://informatics.bmj.com/content/32/1/e101152.full
work_keys_str_mv AT abhisheksharma developinginterpretablemachinelearningmodelstopredictlengthofstayanddispositiondecisionforadultpatientsinemergencydepartments
AT timothynfazio developinginterpretablemachinelearningmodelstopredictlengthofstayanddispositiondecisionforadultpatientsinemergencydepartments
AT longsong developinginterpretablemachinelearningmodelstopredictlengthofstayanddispositiondecisionforadultpatientsinemergencydepartments
AT samanthaplumb developinginterpretablemachinelearningmodelstopredictlengthofstayanddispositiondecisionforadultpatientsinemergencydepartments
AT uweaickelin developinginterpretablemachinelearningmodelstopredictlengthofstayanddispositiondecisionforadultpatientsinemergencydepartments
AT mojgankouhounestani developinginterpretablemachinelearningmodelstopredictlengthofstayanddispositiondecisionforadultpatientsinemergencydepartments
AT markjohnputland developinginterpretablemachinelearningmodelstopredictlengthofstayanddispositiondecisionforadultpatientsinemergencydepartments