Summary points

• Risk prediction tools based on the Framingham score are the most widely used for determining individuals’ absolute risk based on clinical risk factors, although newer tools are better calibrated

• The clinical label metabolic syndrome and abnormalities on the resting electrocardiograph do not add prognostic information beyond that obtained by traditional risk factors

• Exercise stress electrocardiography that generates a positive or negative result based on ST deviation alone is not predictive, although scores that integrate several electrocardiographic and clinical variables may be predictive in patients at intermediate risk

• Laboratory biomarkers, even in the form of multimarker panels, are not helpful in refining clinical risk estimates

• Imaging of subclinical atherosclerosis with computed coronary angiography can identify patients at significantly increased risk, but only a small proportion of patients screened fall into this group

• No randomised evidence to date has shown that informing clinicians and patients of absolute risk of cardiovascular events leads to changes in care or improvement in outcomes