Cardiovascular disease (CVD) risk screening is performed by multivariate methods relying on calculators derived from the Framingham study, other epidemiological studies or primary care records. However, it only identifies 70% of individuals at risk for CVD events and there has been interest in adding other risk factors to improve its predictive capacity. The addition of a family history of premature CVD is well established and there is evidence for adding lipoprotein (a) in some populations and possibly C-reactive protein may be suitable for general use in CVD risk assessment. Most new biochemical and imaging markers have been assessed in the context of improving risk classification in intermediate-risk groups rather than in the general population. There is evidence that N-terminal pro-B-type natriuretic peptide and coronary artery calcium score add significantly to risk prediction. The data for carotid intima-media thickness, ankle-brachial index are less strong and high sensitivity troponins look promising, but have had only limited data to date. Large scale meta-analyses ideally of pooled primary patient data will be required to determine the best additional markers to add to conventional risk prediction and in what groups to apply them.