Observational cohort studies and analysis of the populations of intervention trials at baseline reveal a strong inverse association between circulating levels of high-density lipoprotein (HDL)-cholesterol at baseline and the risk of a fatal or non-fatal cardiovascular event. Intervention with a statin is as effective, in absolute terms, in reducing the risk of coronary events in patients across a wide range of dyslipidaemic phenotypes, including those with low HDL-cholesterol. However, statins exert little effect on the levels of HDL-cholesterol, and treatment with a statin does not eliminate the excess risk associated with low HDL-cholesterol. Additional therapy is clearly required to address this residual risk. The success of clinical evaluations of agents that increase HDL-cholesterol, such as nicotinic acid or fibrate drugs, in reducing the incidence of cardiovascular events points to a way forward. Evidence from outcome studies already points to superior cardiovascular risk reductions in patients receiving a statin plus nicotinic acid, and intensive multi-drug regimens based on such combinations probably represent the way to achieve cardiovascular risk reductions greater than those possible with a statin alone. Accurate and well-validated assays for measuring HDL-cholesterol and more precise definition of optimal levels of HDL-cholesterol in patients with different levels of cardiovascular risk are required. These advances will facilitate the future drafting of guidelines that include correction of low HDL-cholesterol alongside reduction of low-density lipoprotein cholesterol within clinical algorithms for reducing cardiovascular risk.