Because HDL is the best lipid biomarker of cardiovascular risk, much attention has been given to the effects of nutrients on the concentration and composition of HDL. Interpretation has sometimes been inappropriate since increments in HDL do not necessarily imply benefit and vice versa. In particular the strong genetic determinant of the HDL concentration and the heterogeneity of HDL composition need to be considered. Certain disorders in which nutrition plays a key role such as obesity and the metabolic syndrome are characteristically associated with low HDL cholesterol (HDL-C) probably because HDL removal is accelerated. However weight loss that improves the overall metabolic profile does not necessarily lead to improved HDL-C implying that low HDL is part of the genetic disturbance of the disorder. Possibly undue emphasis has been placed on the reduction in HDL-C with carbohydrate rich diets that reflects wider changes in lipoprotein metabolism and is not necessarily detrimental. Similarly the rise in HDL-C with dietary saturated fat and the occasional fall with dietary polyunsaturated fat need not imply major changes in cardiovascular risk and may be influenced by polymorphic variations in regulatory genes and by gender. The type of carbohydrate also influences the HDL response to high carbohydrate diets. Alcohol is the most consistent nutrient to affect HDL-C but it is doubtful that this mechanism alone explains the cardiovascular risk reduction from drinking alcohol.