Using machine learning to identify frequent attendance at accident and emergency services in Lanarkshire

Tools

Reid, Fergus and Pravinkumar, Josephine and Maguire, Roma and Main, Ashleigh and McCartney, Haruno and Winters, Lewis and Dong, Feng (2025) Using machine learning to identify frequent attendance at accident and emergency services in Lanarkshire. Digital Health. ISSN 2055-2076 (In Press) (https://doi.org/10.1177/20552076251315293)

[thumbnail of Reid-etal-DH-2025-Using-machine-learning-to-identify-frequent-attendance] Text. Filename: Reid-etal-DH-2025-Using-machine-learning-to-identify-frequent-attendance.docx
Accepted Author Manuscript
Restricted to Repository staff only until 1 January 2099.
Download (1MB) | Request a copy

Abstract

Background Frequent attenders to Accident and Emergency (A&E) services pose complex challenges for healthcare providers, often driven by critical clinical needs. Machine learning (ML) offers potential for predictive approaches to managing frequent attendance, yet its application in this area is limited. Existing studies often focus on specific populations or models, raising concerns about generalisability. Identifying risk factors for frequent attendance and high resource use is crucial for effective prevention strategies. Objectives This research aims to evaluate the strengths and weaknesses of ML approaches in predicting frequent A&E attendance in NHS Lanarkshire, Scotland, identify associated risk factors, and compare findings with existing research to uncover commonalities and differences. Method Health and social care data were collected from 17,437 A&E patients in NHS Lanarkshire (2021-2022), including clinical, social, and demographic information. Five classification models were tested: Multinomial Logistic Regression (LR), Random Forests (RF), Support Vector Machine Classifier (SVM), k-Nearest Neighbours (k-NN), and Multi-Layer Perceptron Classifier (MLP). Models were evaluated using a confusion matrix and metrics such as precision, recall, F1, and AUC. Shapley values were used to identify risk factors. Results MLP achieved the highest F1 score (0.75), followed by k-NN, RF, and SVM (0.72 each), and LR (0.70). Key health conditions and risk factors consistently predicted frequent attendance across models, with some variation highlighting dataset-specific characteristics. Conclusions This study underscores the utility of combining ML models to enhance prediction accuracy and identify risk factors. Findings align with existing research but reveal unique insights specific to the dataset and methodology.

ORCID iDs

Reid, Fergus, Pravinkumar, Josephine, Maguire, Roma ORCID logoORCID: https://orcid.org/0000-0001-7935-3447, Main, Ashleigh, McCartney, Haruno, Winters, Lewis and Dong, Feng;
CORE (COnnecting REpositories)