Academic Journals Database
Disseminating quality controlled scientific knowledge

Prediction of In-Hospital Mortality in Acute Exacerbations of COPD

Author(s): Ross Archibald | James Chalmers | Tom Fardon | Philip M Short | Pete Williamson | Joanne Taylor | Aran Singanayagam  | Louise Peet | Muhder Al-Khairalla | Stuart Schembri

Journal: Scottish Universities Medical Journal
ISSN 2049-8454

Volume: 1;
Issue: 2;
Start page: 129;
Date: 2012;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Keywords: COPD | Acute exacerbations | predictor of in-hospital mortality | scoring systems

Background: Physicians lack a robust and validated method of measuring severity or predicting poor outcome in patients with acute exacerbation of COPD (AECOPD). Such a predictive tool would allow optimisation of treatment plans for these patients, as well as best use of health care resources. Objective: To determine predictors of in-hospital mortality in AECOPD, and develop a predictive scoring system to identify patients at higher risk of in-hospital mortality.Methods: Analysis of clinical patient data from the exacerbations of obstructive lung disease managed in UK Secondary care [EXODUS] study database, collected from 11 UK hospitals.Results: A total of 1031 patients were included in the validation cohort. The in-hospital mortality rate was 5.2%. Independent predictors of mortality were identified and a new scoring system (“CAUDA70”), for prediction of in-hospital mortality in AECOPD was derived. The score incorporated 6 easily obtained clinical variables: acidosis, albumin, urea, the presence of confusion, MRC dyspnoea score and age. The score displayed strong discrimination, with an area under the receiver operating characteristic (ROC) curve of 0.84. This performance was reproduced in a further validation dataset of 312 patients. The discrimination of the new score exceeds that of existing scores validated for use in AECOPD (CURB65, CRB65 and BAP-65).Conclusion: A new scoring system composed of six readily available clinical variables can accurately predict in-hospital mortality in AECOPD.
RPA Switzerland

RPA Switzerland

Robotic process automation


Tango Jona
Tangokurs Rapperswil-Jona