Avoidable visits to UK emergency departments from the patient perspective

A recursive bivariate probit approach

Bibliographic Data

ID21467402
AuthorsChiara Calastri (0000-0002-2972-661X, University of Leeds, corresponding author), John Buckell (0000-0002-4157-4217, University of Oxford), Romain Crastes Dit Sourd (0000-0003-4506-2910, University of Leeds)
Year2025
Volume154
Pages105265
Publication date2025-04-01
Open AccessYes
TypeARTICLE
VenueHealth Policy (JOURNAL)
Journal identifiersISSN: 0168-8510 • E-ISSN: 1872-6054
PublisherElsevier BV (PUBLISHER)
DOI10.1016/j.healthpol.2025.105265
PMID39987851
OpenAlexW4407393501
LanguageEN
References cited30

We study patient perception of the necessity of Emergency Dept (ED) attendance • Patients who do not trust GPs are likely to deem attendance necessary • The same holds for those who cannot get a timely GP appointment • Policies to improve GP trust and attendance could help limit ED overcrowding Unsustainably high numbers of patients attending emergency departments (ED) is a serious issue worldwide, with consequences for the quality and timeliness of emergency care. Avoidable visits, i.e. unnecessary or that should be dealt with elsewhere, exacerbate this issue. Most studies focussed on avoidable attendances use clinical data collected by hospital staff, while this study relies on survey data collected from patients asked to recall their last ED attendance and reflect on its necessity. We apply a Recursive Bivariate Probit model to quantify the factors affecting patients' perception of an ED visit being avoidable (or not), unveiling how it relates to socio-demographic and contextual factors. We find that patients who do not trust their GP are less likely to think their ED visit was avoidable. The perception of whether an ED visit was avoidable is also affected by symptoms experienced, patients’ ethnicity and waiting time for a GP appointment

Actuarial science · Bivariate analysis · Business · Econometrics · Economics · Multivariate probit model · Ordered probit · Probit · Probit model · Statistics · Computer Science · Emergency and Acute Care Studies · Healthcare Policy and Management · Mathematics · Medical Coding and Health Information · Medicine · Artificial Intelligence