Bayesian-logit model for risk assessment in coronary artery bypass grafting

摘要: Abstract Predictive models for the assessment of operative risk using patient factors have gained popularity in medical community as an important tool adjustment surgical outcome. The Bayes' theorem model is among various used to predict mortality patients undergoing coronary artery bypass grafting procedures. Comparative studies classic statistical techniques, such logistic regression, cluster variables followed by a subjectively created sickness score, classification trees model, and shown that those with highest predictive power. In this study, reformulated equation extended include qualitative quantitative factors. We show resulting Bayesian-logit mixture regression linear discriminant analysis. This new can be easily without complex computer programs. Using 12,712 procedures at Department Veterans Affairs Continuous Improvement Cardiac Surgery Study between April 1987 March 1990, power Bayesianlogit compared ability discriminate deaths survivors comparable