Predicting Cardiac Surgery Outcome in an Australian Patient Cohort

Accurate predictions of post-surgical mortality risk allow both surgeons and patients to participate in the pre-surgery decision-making process in an informed manner. This thesis studied the development of risk prediction models to improve cardiac surgery outcome assessment. The research showed that ambiguous predictors and outcome definition, sub-optimum sample size, inappropriate handling of missing data and inefficient predictor selection techniques were major issues that compromised the performance of the currently used models. A set of novel risk prediction models for predicting long-term survival following Coronary Artery Bypass Graft (CABG) surgery were developed in an attempt to address the methodological concerns.