The current diagnostic threshold for diabetes mellitus is imposed on a continuous distribution of blood glucose measurement. A more clinical approach estimates a threshold above which the rate of diabetes-specific complications rises steeply. However, the diagnostic threshold for diabetes is essentially established on the risk of microvascular and not cardiovascular complications. Indeed, while there appears to be a continuous relationship between blood glucose, cardiovascular risk and overall mortality, this association extends into the sub-diabetic range, with no threshold identified. In this regard, the assumption that the diagnosis of diabetes can effectively identify patients at elevated risk of cardiovascular morbidity and mortality is potentially flawed, and questions the utility of diabetes status (as a dichotomous variable) for the assessment and management of cardiovascular risk. Indeed, the increased risk of cardiovascular complications may not be related to diabetes status per se but the frequent association of diabetes with a high-risk phenotype, now recognised as the so-called ‘metabolic syndrome’. By implication, cardiovascular disease prevention should not be dominated by a drive for the prevention of diabetes, but this broader clinical syndrome of increased cardiovascular risk.