To Örebro University

Introduction: Research in seasonal dietary variation has mainly focused on differences in types and quantities of foods consumed; however, with recent interest in the role of chrono-nutrition and health, we studied seasonal variation in meal frequency, time and size, in Sweden where daylight varies from south to north from 17-24 h in summer to 7-0 h in winter.

Methods: In the national dietary survey in Sweden, Riksmaten adults 2010-11, a representative section of the population between ages 18-80 years was sampled.  Participants completed an online self-administered food diary on four consecutive days, with mealtimes recorded in 15-minute slots.  We calculated hourly and daily energy intake (DEI) and the frequency of small and large eating and drinking occasions (EDO; small: time slot >210 kJ and <15%DEI; large: >=15%DEI).  We divided the acceptable energy reporters (men/women) into three regions, from South to North: Götaland (N=308/404), Svealand (N=257/355) and Norrland (N=76/114).  Mixed linear models were used to account for correlations between days and adjust seasonal and regional associations for: age (categorical), day of the week (week vs weekend), education and DEI.

Results: Mean DEI and EDO frequency were not significantly different between seasons, regions or their interaction; however, in weekend vs week days, DEI was significantly higher (P<0.001) by ~0.5 MJ/d, whereas EDO frequency -driven by small EDO- was significantly lower (P<0.001) by ~0.3, thereby increasing meal energy density.  Up to age 55 years, time of first EDO was ~75-90 min later at the weekend; however, 15 min by 65 years.  Time of first EDO was not associated with region or season; however, in women (P=0.021), time of last large EDO was earlier in Norrland, except in summer time when meals were ~60 min later, exceeding other regions.  These differences were driven by weekend days.

Discussion: Habitual patterns overrule seasonal and regional variation, potentially helping with daylight adaptation; however, week and weekend days were associated with different meal patterns, delaying breakfast by 1 hour and consuming higher energy dense meals.  Short-term dietary assessment instruments require balanced week days to study the potential role of meal pattern irregularity for health consequences.