作者: Allan S. Detsky , Howard B. Abrams , John R. McLaughlin , Daniel J. Drucker , Zion Sasson
DOI: 10.1007/BF02596184
关键词:
摘要: The authors prospectively studied 455 consecutive patients referred to the general medical consultation service for cardiac risk assessment prior non-cardiac surgery, in order validate a previously derived multifactorial index their clinical setting. They also tested version of that they had modified reflect factors believed be important. For undergoing major original performed less well validation data set than derivation (p<0.05), but still added predictive information statistically significant degree (p<0.05). both and minor demonstrating an area under Receiver Operating Characteristic curve 0.75 (95% confidence interval 0.70 0.80). A simple nomogram is presented which will enable conversion pretest probabilities into posttest using likelihood ratios associated with each score. It recommended clinicians estimate local overall complication rates (pretest probabilities) clinically relevant populations settings before apply properties (likelihood ratios) demonstrated this study calculate risks individual (posttest probabilities).