|
Activity Number:
|
476
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #303625 |
|
Title:
|
Predictive Accuracy in Time-to-Event Data: Application to Evaluation of Pneumonia Severity Indexes
|
|
Author(s):
|
Guy N. Brock*+ and Forest W. Arnold and Paula Peyrani and Julio A. Ramirez
|
|
Companies:
|
University of Louisville and University of Louisville and University of Louisville and University of Louisville
|
|
Address:
|
400 E Gray St, Louisville, KY, 40202,
|
|
Keywords:
|
predictive accuracy ; time-to-event data ; community acquired pneumonia ; explained variation ; concordance index ; time to clinical stability
|
|
Abstract:
|
Severity scores are often calculated as a means of classifying patients into low and high risk groups, in order to provide better patient care and management. Derivation and subsequent validation of these severity scores is often based solely on patient mortality, though other outcomes such as length of hospital stay (LOS) and time to clinical stability (TCS) are important outcomes to consider for patient management. The pneumonia severity index (PSI) and CRB-65 are two severity scores that have been proposed to predict patient outcomes and guide initial management of patients with community-acquired pneumonia (CAP). In this study, we evaluate and compare the ability of the PSI and CRB-65 to predict the time-to-event outcomes LOS and TCS, using data collected from an international database of adult patients admitted to the hospital with CAP.
|
