Recommendations on How to Construct Risk Scores for Cardiac Surgical Patients

The structural equation model is proposed for constructing risk scores for cardiac surgical patients, in view of the perioperative nature of risk-scoring and the complexity in data structures. The decision trees could be applied for model selection, in terms of identification of relevant predictors and variable discretization. The pitfalls of the conventional methodology, based on logistic regression for estimation and prediction, Hosmer-Lemeshow test for goodness of fit and c-statistics for assessment of predictive accuracy, are also discussed.