FIGURE 2.

Decision tree (using the CHAID algorithm) for predicting presence of anxiety or depression in late convalescence among 723 patients admitted to hospital for acute myocardial infarction (AMI), coronary artery bypass graft surgery (CABGS), acute coronary syndrome (ACS), percutaneous coronary syndrome (PCI), or angina. At each node the best predictor for anxiety or depression was selected from the six key potential predictors identified in bivariate analyses, and the optimal forecasting values were determined.