Objective: To examine the effects of modifying meal consumption frequency on blood lipids in free-living healthy male volunteers on self-selected diets.
Design: Cross-over study of snacking to non-snacking and non-snacking to snacking patterns.
Setting: Civic Offices, Dublin.
Subjects: 12 men with a mean (SD) eating frequency of 6.0 +/- 0.8 times per day ('snacking') and 11 men with a frequency of 3.1 +/- 0.1 meals per day ('non-snackers').
Intervention: The 'snackers' moved from 6.0 to 3.3 +/- 0.3 meal-eating occasions per day while the 'non-snackers' moved from 3.1 to 5.9 +/- 0.9 times per day. It was intended that these modifications of eating frequency would not be the cause of modified nutrient intake in the cross-over period.
Main outcome measures: Nutrient intake, plasma total, high density lipoprotein (HDL) and low density lipoprotein (LDL) cholesterol, plasma triglyceride and apoproteins A-I and B.
Results: There was a significant treatment (cross-over) effect on plasma total cholesterol (P = 0.038), LDL cholesterol (P = 0.038), HDL/LDL cholesterol ratio (P = 0.013) and apo A-I/B (P = 0.029). There were no significant nutrient changes on moving from meal-eating to snacking but on moving from snacking to meal-eating there were significant changes (P < 0.005) in the % energy from protein, fat, saturated fatty acids and alcohol. These changes accounted for all the changes in blood cholesterol in this group. When the data for meal-eating to snacking were analysed separately, given that nutrient intakes did not change, a significant effect of altered meal-eating frequency was found for total cholesterol (P = 0.027) and HDL/LDL (P = 0.015).
Conclusion: These data on free-living subjects following normal self-selected diets support the hypothesis and the substantial related literature that more frequent meal consumption has a favourable effect on lowering plasma cholesterol and raising the HDL/LDL cholesterol ratio.