Using Data Mining to Predict Hospital Admissions from the Emergency Department

Byron Graham, Raymond Bond, Michael Quinn, Maurice Mulvenna

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Crowding within Emergency Departments (EDs) can have significant negative consequences for patients. EDs therefore need to explore the use of innovative methods to improve patient flow and prevent overcrowding. One potential method is the use of data mining using machine learning techniques to predict ED admissions. This study uses routinely collected administrative data (120,600 records) from two major acute hospitals in Northern Ireland to compare contrasting machine learning algorithms in predicting the risk of admission from the ED. We use three algorithms to build the predictive models: (1) logistic regression, (2) decision trees, and (3) gradient boosted machines (GBM). The GBM performed better (accuracy=80.31%, AUC-ROC=0.859) than the decision tree (accuracy=80.06%, AUC-ROC=0.824) and the logistic regression model (accuracy=79.94%, AUC-ROC=0.849). Drawing on logistic regression, we identify several factors related to hospital admissions including hospital site, age, arrival mode, triage category, care group, previous admission in the past month, and previous admission in the past year. This study highlights the potential utility of three common machine learning algorithms in predicting patient admissions. Practical implementation of the models developed in this study in decision support tools would provide a snapshot of predicted admissions from the emergency department at a given time, allowing for advance resource planning and the avoidance bottlenecks in patient flow, as well as comparison of predicted and actual admission rates. When interpretability is a key consideration, EDs should consider adopting logistic regression models, although GBM’s will be useful where accuracy is paramount.
LanguageEnglish
Pages10458-10469
JournalIEEE Access
Volume6
Early online date22 Feb 2018
DOIs
Publication statusE-pub ahead of print - 22 Feb 2018

Fingerprint

data mining
logistics
overcrowding
hospital
resource
machine learning
decision
method

Keywords

  • Data Mining
  • Emergency Department
  • Hospitals
  • Machine Learning
  • Predictive Models

Cite this

@article{ad1219881253430cad2f105b865aaaf3,
title = "Using Data Mining to Predict Hospital Admissions from the Emergency Department",
abstract = "Crowding within Emergency Departments (EDs) can have significant negative consequences for patients. EDs therefore need to explore the use of innovative methods to improve patient flow and prevent overcrowding. One potential method is the use of data mining using machine learning techniques to predict ED admissions. This study uses routinely collected administrative data (120,600 records) from two major acute hospitals in Northern Ireland to compare contrasting machine learning algorithms in predicting the risk of admission from the ED. We use three algorithms to build the predictive models: (1) logistic regression, (2) decision trees, and (3) gradient boosted machines (GBM). The GBM performed better (accuracy=80.31{\%}, AUC-ROC=0.859) than the decision tree (accuracy=80.06{\%}, AUC-ROC=0.824) and the logistic regression model (accuracy=79.94{\%}, AUC-ROC=0.849). Drawing on logistic regression, we identify several factors related to hospital admissions including hospital site, age, arrival mode, triage category, care group, previous admission in the past month, and previous admission in the past year. This study highlights the potential utility of three common machine learning algorithms in predicting patient admissions. Practical implementation of the models developed in this study in decision support tools would provide a snapshot of predicted admissions from the emergency department at a given time, allowing for advance resource planning and the avoidance bottlenecks in patient flow, as well as comparison of predicted and actual admission rates. When interpretability is a key consideration, EDs should consider adopting logistic regression models, although GBM’s will be useful where accuracy is paramount.",
keywords = "Data Mining, Emergency Department, Hospitals, Machine Learning, Predictive Models",
author = "Byron Graham and Raymond Bond and Michael Quinn and Maurice Mulvenna",
year = "2018",
month = "2",
day = "22",
doi = "10.1109/ACCESS.2018.2808843",
language = "English",
volume = "6",
pages = "10458--10469",
journal = "IEEE Access",
issn = "2169-3536",

}

Using Data Mining to Predict Hospital Admissions from the Emergency Department. / Graham, Byron; Bond, Raymond; Quinn, Michael; Mulvenna, Maurice.

In: IEEE Access, Vol. 6, 22.02.2018, p. 10458-10469.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Using Data Mining to Predict Hospital Admissions from the Emergency Department

AU - Graham, Byron

AU - Bond, Raymond

AU - Quinn, Michael

AU - Mulvenna, Maurice

PY - 2018/2/22

Y1 - 2018/2/22

N2 - Crowding within Emergency Departments (EDs) can have significant negative consequences for patients. EDs therefore need to explore the use of innovative methods to improve patient flow and prevent overcrowding. One potential method is the use of data mining using machine learning techniques to predict ED admissions. This study uses routinely collected administrative data (120,600 records) from two major acute hospitals in Northern Ireland to compare contrasting machine learning algorithms in predicting the risk of admission from the ED. We use three algorithms to build the predictive models: (1) logistic regression, (2) decision trees, and (3) gradient boosted machines (GBM). The GBM performed better (accuracy=80.31%, AUC-ROC=0.859) than the decision tree (accuracy=80.06%, AUC-ROC=0.824) and the logistic regression model (accuracy=79.94%, AUC-ROC=0.849). Drawing on logistic regression, we identify several factors related to hospital admissions including hospital site, age, arrival mode, triage category, care group, previous admission in the past month, and previous admission in the past year. This study highlights the potential utility of three common machine learning algorithms in predicting patient admissions. Practical implementation of the models developed in this study in decision support tools would provide a snapshot of predicted admissions from the emergency department at a given time, allowing for advance resource planning and the avoidance bottlenecks in patient flow, as well as comparison of predicted and actual admission rates. When interpretability is a key consideration, EDs should consider adopting logistic regression models, although GBM’s will be useful where accuracy is paramount.

AB - Crowding within Emergency Departments (EDs) can have significant negative consequences for patients. EDs therefore need to explore the use of innovative methods to improve patient flow and prevent overcrowding. One potential method is the use of data mining using machine learning techniques to predict ED admissions. This study uses routinely collected administrative data (120,600 records) from two major acute hospitals in Northern Ireland to compare contrasting machine learning algorithms in predicting the risk of admission from the ED. We use three algorithms to build the predictive models: (1) logistic regression, (2) decision trees, and (3) gradient boosted machines (GBM). The GBM performed better (accuracy=80.31%, AUC-ROC=0.859) than the decision tree (accuracy=80.06%, AUC-ROC=0.824) and the logistic regression model (accuracy=79.94%, AUC-ROC=0.849). Drawing on logistic regression, we identify several factors related to hospital admissions including hospital site, age, arrival mode, triage category, care group, previous admission in the past month, and previous admission in the past year. This study highlights the potential utility of three common machine learning algorithms in predicting patient admissions. Practical implementation of the models developed in this study in decision support tools would provide a snapshot of predicted admissions from the emergency department at a given time, allowing for advance resource planning and the avoidance bottlenecks in patient flow, as well as comparison of predicted and actual admission rates. When interpretability is a key consideration, EDs should consider adopting logistic regression models, although GBM’s will be useful where accuracy is paramount.

KW - Data Mining

KW - Emergency Department

KW - Hospitals

KW - Machine Learning

KW - Predictive Models

U2 - 10.1109/ACCESS.2018.2808843

DO - 10.1109/ACCESS.2018.2808843

M3 - Article

VL - 6

SP - 10458

EP - 10469

JO - IEEE Access

T2 - IEEE Access

JF - IEEE Access

SN - 2169-3536

ER -