White Rose University Consortium logo
University of Leeds logo University of Sheffield logo York University logo

Clinical scoring system to identify high-acuity patients from information available in the Emergency Department

Challen, Kirsty (2014) Clinical scoring system to identify high-acuity patients from information available in the Emergency Department. PhD thesis, University of Sheffield.

Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (10Mb) | Preview


Background Acute illness results in millions of hospital admissions per year. Assessment of illness severity can guide the intensity and location of care provided but although multiple clinical decision aids exist for large numbers of conditions no well-validated clinical decision aid exists for the assessment of patients with unselected medical emergencies. It cannot be assumed that tools which predict death will accurately identify those patients with the most potential to benefit from urgent care. Methods A prospective cohort was analysed using logistic regression to develop bedside scores to identify patients at high risk of death, and those where emergency care has the potential to affect survival. Consensus methodology was used to develop threshold responses for the Emergency Department. Results 7 variables and one interaction (age, respiratory rate, diastolic blood pressure, oxygen saturation, temperature, GCS, pre-existing respiratory disease, respiratory disease by temperature interaction) predicted death in 7 days with AUROC 0.753 (derivation set: n=2437) and 0.719 (validation set: n=2322). Other scores showed AUROC 0.658 – 0.762. 3 variables (pulse, systolic blood pressure, GCS) predicted potentially preventable or potentially prevented death with AUROC 0.737 (derivation set: n=398) and 0.686 (validation set: n=227). Other scores showed AUROC 0.559 – 0.684. Consensus was reached on four thresholds of clinical response on the 0-27 point score for potentially preventable or potentially prevented death. Conclusions No published tool exists to identify the patient most likely to benefit from emergency department care. Variables predicting death do not necessarily predict potential to benefit from care, and existing scores have only moderate discrimination for this. The tool developed here shows potential but ongoing research should address which patients will benefit from time-critical interventions and the complexities of ED prioritisation.

Item Type: Thesis (PhD)
Academic Units: The University of Sheffield > Faculty of Medicine, Dentistry and Health (Sheffield) > School of Health and Related Research (Sheffield)
Identification Number/EthosID: uk.bl.ethos.658743
Depositing User: Dr Kirsty Challen
Date Deposited: 03 Aug 2015 13:14
Last Modified: 03 Oct 2016 12:18
URI: http://etheses.whiterose.ac.uk/id/eprint/9346

You do not need to contact us to get a copy of this thesis. Please use the 'Download' link(s) above to get a copy.
You can contact us about this thesis. If you need to make a general enquiry, please see the Contact us page.

Actions (repository staff only: login required)