Circadian rhythms of the spontaneous meal pattern, macronutrient intake, and mood of humans.

Circadian (24-hour) rhythms in the feeding behavior of humans were investigated using diary self-reports of spontaneous food intake. Eight male and 30 female undergraduate students recorded what they ate, when they ate it, and their mood at the time of ingestion in a diary over a consecutive nine day period. Self-ratings of depression, energy, and anxiety were made at the beginning of each meal on three seven-point scales. The total amount of food energy in each meal as well as the amount of protein, carbohydrate, and fat, the intervals prior to and following the meals, and the satiety and deprivation ratios were calculated. The food energy contained in the stomach at the beginning and end of each meal was estimated with a mathematical model. These variables were evaluated in relation to the time of meal occurrence during the day. Fluctuations in the levels of self-rated energy and anxiety, but not depression, were detected during the day. Clear 24-hour rhythms were identified for the amount eaten and the macronutrients ingested during the day with decreases for males and increases for females. The amount eaten per meal and the meal's content of carbohydrate or fat, but not protein, varied over the day with peaks at the lunch and dinner periods. A clear sex difference without circadian variation was apparent with the deprivation ratios. This suggests that males eat larger meals than females because of a heightened responsivity to deprivation and not to a smaller response to the satiating properties of food. Preprandial correlations were found for meals occurring either during the breakfast or the dinner periods. No postprandial correlations were found. These data demonstrate that the preprandial correlations are not an artifact produced by the 24-hour rhythm and suggests that they reflect a basic regulatory strategy employed by humans. As the day progressed, postmeal intervals and satiety ratios decreased, while premeal intervals increased. This suggests that humans obtain less satiety from a given amount of food later in the day than earlier. It is postulated that this represents eating which anticipates the overnight fast. These data clearly demonstrate the efficacy of the approach and the orderly, analyzable nature of the spontaneous eating behavior of humans.