External validation and updating of a prediction model for nosocomial pneumonia after coronary artery bypass graft surgery

The generalizability of a prediction model from North America for incident nosocomial pneumonia following coronary artery bypass graft surgery was assessed for 23247 patients on the Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) registry. The performance of the North A...

Full description

Bibliographic Details
Main Authors: Sanagou, M., Wolfe, R., Leder, K., Reid, Christopher
Format: Journal Article
Published: Cambridge University Press 2014
Online Access:http://hdl.handle.net/20.500.11937/41927
Description
Summary:The generalizability of a prediction model from North America for incident nosocomial pneumonia following coronary artery bypass graft surgery was assessed for 23247 patients on the Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) registry. The performance of the North American model was evaluated using measures of calibration and discrimination. The model had reasonable discrimination (area under the receiver-operating characteristic curve, AUC=0·69), but unsatisfactory calibration (Hosmer-Lemeshow test, P<0·001) in the ANZSCTS patients. An update of the model coefficients yielded a model with AUC=0·71 and good calibration (P=0·46). © 2013 Cambridge University Press.