Early warning scores in pre-hospital and emergency care

Corfield, Alasdair R. (2022) Early warning scores in pre-hospital and emergency care. MD thesis, University of Glasgow.

Full text available as:
[thumbnail of M.D. thesis awarded by published work, edited version, 3rd party copyright removed] PDF (M.D. thesis awarded by published work, edited version, 3rd party copyright removed)
Download (1MB)

Abstract

Background: Early warning scores (EWS) are a composite, ordinal scale based on observed patient physiology values. Their original description was to be used in a hospital setting, as a method of tracking patient deterioration and triggering an appropriate response to a deteriorating patient.
The aim of this work is to describe whether an EWS taken at a single time point are useful tools to use in the pre-hospital and emergency hospital setting to identify patients at high risk of clinical deterioration.
Methods: Datasets covering national presentations to acute hospitals and single centre data were used. Datasets contained adult and paediatric data – National Early Warning Score (NEWS) was used for adults and Paediatric Early Warning Score (PEWS) was used for under 16 years of age. Models were constructed to test the utility of a single EWS, either pre-hospital or in the Emergency Department (ED), as a predictor of adverse outcome (death or ICU admission) or hospital admissions for paediatric patients.
Results: NEWS and PEWS have moderate to good predictive value for adverse outcome in a variety of settings. ED patients with sepsis AUROC 0.71 (95% CI 0.68 to 0.74), all adult ambulance patients AUROC 0.81 (95% CI 0.73-0.99), all paediatric ambulance patients AUROC 0.80 (95% CI 0.76 to 0.84). A modified qPEWS performs as well as PEWS. PEWS is not predictive of the need for hospital admission AUROC 0.62 (95% CI 0.61 - 0.63).
Conclusion: A single NEWS and PEWS in ED or the pre-hospital environment has the ability to predict patients at greater risk of deterioration and adverse outcome. A modified qPEWS may improve data collection without sacrificing predictive value. These results do not examine whether this association can be implemented to improve outcomes, and further prospective research is required in this area.

Item Type: Thesis (MD)
Qualification Level: Doctoral
Subjects: R Medicine > R Medicine (General)
Colleges/Schools: College of Medical Veterinary and Life Sciences
Supervisor's Name: Maguire, Dr. Donogh
Date of Award: 2022
Depositing User: Theses Team
Unique ID: glathesis:2022-82832
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 03 May 2022 10:24
Last Modified: 08 Mar 2024 09:35
Thesis DOI: 10.5525/gla.thesis.82832
URI: https://theses.gla.ac.uk/id/eprint/82832
Related URLs:

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year