The term 'optimum nutrition' has evolved from a perceived need to base recommendations for nutrient intakes firmly in the context of function. It follows that 'optimum nutritional status' for individual nutrients should be defined in terms of biochemical or physiological markers having some functional value but also showing an appropriate relationship to nutrient intake. The present short review considers the current position regarding such markers for riboflavin, pyridoxine and niacin. It is concluded that whilst there are several biochemical measures which respond to changes in intake of each of these vitamins, no single measure is wholly satisfactory as a marker of optimum status.