Machine-learning prediction for hospital length of stay using a French medico-administrative database.

Machine learning health services research neural network prediction public health

Journal

Journal of market access & health policy
ISSN: 2001-6689
Titre abrégé: J Mark Access Health Policy
Pays: United States
ID NLM: 101670174

Informations de publication

Date de publication:
2023
Historique:
entrez: 2 12 2022
pubmed: 3 12 2022
medline: 3 12 2022
Statut: epublish

Résumé

Prolonged Hospital Length of Stay (PLOS) is an indicator of deteriorated efficiency in Quality of Care. One goal of public health management is to reduce PLOS by identifying its most relevant predictors. The objective of this study is to explore Machine Learning (ML) models that best predict PLOS. Our dataset was collected from the French Medico-Administrative database (PMSI) as a retrospective cohort study of all discharges in the year 2015 from a large university hospital in France (APHM). The study outcomes were LOS transformed into a binary variable (long vs. short LOS) according to the 90 Our analysis included 73,182 hospitalizations, of which 7,341 (10.0%) led to PLOS. The GB classifier was the most performant model with the highest AUC (0.810), superior to all the other models (all p-values <0.0001). The performance of the RF, GB and NN models (AUC ranged from 0.808 to 0.810) was superior to that of the LR model (AUC = 0.795); all p-values <0.0001. In contrast, LR was superior to CART (AUC = 0.786), p < 0.0001. The variable most predictive of the PLOS was the destination of the patient after hospitalization to other institutions. The typical clinical profile of these patients (17.5% of the sample) was the elderly patient, admitted in emergency, for a trauma, a neurological or a cardiovascular pathology, more often institutionalized, with more comorbidities notably mental health problems, dementia and hemiplegia. The integration of ML, particularly the GB algorithm, may be useful for health-care professionals and bed managers to better identify patients at risk of PLOS. These findings underscore the need to strengthen hospitals through targeted allocation to meet the needs of an aging population.

Identifiants

pubmed: 36457821
doi: 10.1080/20016689.2022.2149318
pii: 2149318
pmc: PMC9707380
doi:

Types de publication

Journal Article

Langues

eng

Pagination

2149318

Informations de copyright

© 2022 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.

Déclaration de conflit d'intérêts

No potential conflict of interest was reported by the authors.

Références

PLoS One. 2017 Apr 10;12(4):e0175383
pubmed: 28394905
J Adv Pract Oncol. 2012 May;3(3):181-3
pubmed: 25031944
Am J Perinatol. 2015 Jul;32(9):825-32
pubmed: 25594218
Med Care. 2021 Oct 1;59(10):929-938
pubmed: 34310455
Health Expect. 2018 Feb;21(1):41-56
pubmed: 28898930
Respir Care. 2012 Jun;57(6):889-97; discussion 898-9
pubmed: 22663965
PLoS One. 2018 Nov 27;13(11):e0207982
pubmed: 30481205
Int J Epidemiol. 2017 Apr 1;46(2):392-392d
pubmed: 28168290
Arch Surg. 1984 Nov;119(11):1338-42
pubmed: 6437374
Ann Thorac Surg. 2012 Jul;94(1):109-16
pubmed: 22579949
J R Coll Physicians Edinb. 2020 Dec;50(4):398-402
pubmed: 33469615
Intell Based Med. 2021;5:100035
pubmed: 34075366
Br J Anaesth. 2010 Apr;104(4):459-64
pubmed: 20185517
Nat Mach Intell. 2020 Jan;2(1):56-67
pubmed: 32607472
Health Policy. 2021 Apr;125(4):512-519
pubmed: 33579560
BMJ. 2015 Jun 02;350:h2622
pubmed: 26037412
J Crit Care. 2014 Feb;29(1):60-5
pubmed: 24268622
Aging Clin Exp Res. 2019 Jul;31(7):993-999
pubmed: 30191455
Can J Aging. 2019 Mar;38(1):90-99
pubmed: 30404681
Lancet Psychiatry. 2016 Mar;3(3):243-50
pubmed: 26803397
Arch Surg. 2007 May;142(5):461-5; discussion 465-6
pubmed: 17515488
Entropy (Basel). 2020 Dec 25;23(1):
pubmed: 33375658
Ann Thorac Surg. 2012 Feb;93(2):565-9
pubmed: 22197534
Intern Emerg Med. 2020 Sep;15(6):989-995
pubmed: 31898204
Med Care. 2005 Nov;43(11):1130-9
pubmed: 16224307
Medicine (Baltimore). 2020 Dec 4;99(49):e22361
pubmed: 33285668
J Clin Diagn Res. 2015 Apr;9(4):OC19-23
pubmed: 26023581
PLoS One. 2018 Nov 8;13(11):e0207203
pubmed: 30408118
Chest. 2005 Dec;128(6):3937-54
pubmed: 16354866
Chest. 2006 Apr;129(4):954-9
pubmed: 16608944
Intern Med J. 2022 Feb;52(2):176-185
pubmed: 33094899
J Pharm Policy Pract. 2020 Jun 16;13:24
pubmed: 32549990
Bioinformatics. 2010 May 15;26(10):1340-7
pubmed: 20385727
Z Gerontol Geriatr. 2019 Nov;52(Suppl 4):212-221
pubmed: 31628613

Auteurs

Franck Jaotombo (F)

Aix-Marseille University, EA 3279 - Public Health, Chronic Diseases and Quality of Life - Research Unit, La Timone Medical University, Marseille, France.
I2M, CNRS, UMR, Aix-Marseille University, Marseille, France.
Operations Data and Artificial Intelligence, EM Lyon Business School, Ecully, France.

Vanessa Pauly (V)

Aix-Marseille University, EA 3279 - Public Health, Chronic Diseases and Quality of Life - Research Unit, La Timone Medical University, Marseille, France.
Service d'Information Médicale, Public Health Department, La Conception Hospital, Assistance Publique - Hôpitaux de Marseille, Marseille, France.

Guillaume Fond (G)

Aix-Marseille University, EA 3279 - Public Health, Chronic Diseases and Quality of Life - Research Unit, La Timone Medical University, Marseille, France.

Veronica Orleans (V)

Service d'Information Médicale, Public Health Department, La Conception Hospital, Assistance Publique - Hôpitaux de Marseille, Marseille, France.

Pascal Auquier (P)

Aix-Marseille University, EA 3279 - Public Health, Chronic Diseases and Quality of Life - Research Unit, La Timone Medical University, Marseille, France.

Badih Ghattas (B)

I2M, CNRS, UMR, Aix-Marseille University, Marseille, France.

Laurent Boyer (L)

Aix-Marseille University, EA 3279 - Public Health, Chronic Diseases and Quality of Life - Research Unit, La Timone Medical University, Marseille, France.
Service d'Information Médicale, Public Health Department, La Conception Hospital, Assistance Publique - Hôpitaux de Marseille, Marseille, France.

Classifications MeSH