The 1996 coronary artery bypass risk model: The society of thoracic surgeons adult cardiac national database

Background. The Society of Thoracic Surgeons Adult Cardiac National Databas e has recently completed the update for the 1996 risk model to be used to e stimate the risk of operative death for isolated coronary artery bypass gra ft (CABG) procedures. Methods. We placed emphasis on clinical relevance, data quality, data compl eteness, and univariate analyses. A logistic regression approach was used t o develop the 1996 CABG-only risk model. Results. Odds ratios for the factors with highest risk are multiple reopera tions (OR = 4.3), emergent salvage status (OR = 3.7), and first reoperation (OR = 2.7). Standard performance measures indicated the model had high pre dictive power and an acceptable level of calibration after adjustment for a large sample size effect. Conclusion. The most current STS risk model of CABG operative mortality is a reliable and statistically valid tool. The 1996 CABG-only model has been approved for use by The Society of Thoracic Surgeons. (Ann Thorac Surg 1999 ;67:1205-8) (C) 1999 by The Society of Thoracic Surgeons.