Background: The Society of Thoracic Surgeons National Cardiac Surgery Database has recently completed gathering patient data from 1990 through 1994. Using information from more than 300,000 patients undergoing isolated coronary artery bypass grafting in this period, new risk models of operative mortality were developed.
Methods: Logistic regression analysis was used to develop a risk model for each calendar year. A standard "training set/test set" approach was used for each model.
Results: Five validation techniques were used to evaluate the reliability of the risk models. All models were found to predict operative mortality with good accuracy in this population.
Conclusions: The new risk models for isolated coronary artery bypass operations serve as reliable predictors of operative mortality for the most recent harvest of patient data from The Society of Thoracic Surgeons National Cardiac Surgery Database.