Epidemiological data suggesting benefits of increased meal frequency on CHD risk and lipid and lipoprotein levels may be flawed because of biases associated with the method of data collection. In normolipidaemic individuals increasing meal frequency from three to six or more meals daily over a period of several weeks appears to be associated with reduced levels of total and LDL-cholesterol, possibly due to reduced cholesterol synthesis or enhancement of reverse cholesterol transport. However, in non-obese individuals with polygenic hyperlipidaemia and non-insulin-dependent diabetes, altered meal frequency does not appear to confer similar benefits. The reasons for the different responses in these groups of subjects is not clear. The reported effects of altered meal frequency on diurnal levels of triacylglycerols and fatty acids are more variable, with benefits being reported principally in association with meal frequencies exceeding those which might be translated into practical recommendations. Present data preclude recommendations concerning meal frequency on the basis of effects on lipids and lipoproteins. Even in healthy individuals amongst whom benefits have been observed, the periods of observation have been relatively short and it is not known whether adaptation occurs. Further research should focus on longer-term studies in healthy individuals as well as hyperinsulinaemic and diabetic subjects. For those individuals choosing to have a large number of small meals each day, it is relevant to emphasize that no untoward effects of increased meal frequency on lipids or lipoproteins have been demonstrated.