“Eat breakfast like a king, lunch like a prince, and dinner like a pauper” is a much heard quotation, although the evidence for this is uncertain. Meal frequency( Reference Titan, Bingham and Welch 1 ) and breakfast consumption( Reference Purslow, Sandhu and Forouhi 2 ) have been associated with lower serum cholesterol and reduced rate of weight gain respectively in the Norfolk cohort of the European Prospective Investigation into Cancer (EPIC-Norfolk). However, the association between time of day of maximum energy consumption in relation to serum cholesterol concentrations was not previously studied.
The EPIC-Norfolk study started in 1993 and had 25,639 men and women, aged 40–79, who attended a health examination. Weight, height and blood samples were taken by a nurse and participants received instructions on completing a 7-day diet diary (7dDD)( Reference Bingham, Welch and McTaggart 3 ). We excluded nightshift workers, leaving 25,295 for analysis. For every diary day, energy intake was summed and re-categorised into: morning (M), afternoon (A) and evening (E) and calculated as a proportion of the whole day. Participants' individual diary days were scored depending whether most energy was consumed in the M, A or E. The median of the scores across days determined the type of eater a participant was (91% provided 7 days of DD). Differences between M, A and E eaters in continuous variables were tested using Kruskal-Wallis, and if significant, followed by Mann-Whitney test; proportional variables tested with Chi-squared test. Differences in cholesterol concentrations were tested using ANOVA, adjusted for sex, age, smoking, social class, physical activity and body mass index (BMI).
The majority of participants (72%) consumed most of their energy in the evening. Those who consumed the highest energy intake in the morning had significantly higher total daily energy intake. Among men, M-eaters had a higher proportion of smokers, manual occupation and were more physically active. After adjustment for sex and confounders, cholesterol concentrations for M-eaters were 5.90 mmol/L (95% CI: 5.79, 6.00), which was 0.29 mmol/L lower (95% CI: −0.40, −0.19) compared to A- and E-eaters.
Continuous variables are mean and standard deviation.
Significant differences (P<0.05) between M&A (#), M&E ($); ns=no overall difference (Kruskal-Wallis).
Having the highest energy intake in the morning was associated with lower cholesterol concentrations, a reduction similar to what can be achieved with dietary advice( Reference Thompson 4 ); however, residual confounding cannot be excluded and the proportion of M-eaters was limited. Moreover, possible interactions with meal frequency will require further study.
