Dietary carbohydrate and nocturnal sleep duration in relation to children's BMI : findings from the IDEFICS study in eight European countries

Previous research has found an association between being overweight and short sleep duration. We hypothesized that this association could be modified by a high carbohydrate (HC) diet and that the timing and type (starch or sugar) of intake may be an important factor in this context. Participants in the prospective, eight-country European study IDEFICS were recruited from September 2007 to June 2008, when they were aged two to nine years. Data on lifestyle, dietary intake and anthropometry were collected on two occasions. This study included 5944 children at baseline and 4301 at two-year follow-up. For each meal occasion (morning, midday, and evening), starch in grams and sugar in grams were divided by total energy intake (EI), and quartiles calculated. HC-starch and HC-sugar intake categories were defined as the highest quartile for each meal occasion. In a mutually adjusted linear regression model, short sleep duration as well as HC-starch in the morning were positively associated with body mass index (BMI) z-scores at baseline. HC-starch at midday was positively associated with body mass index (BMI) z-scores in children with short sleep duration, and negatively associated with BMI z-scores in those with normal sleep. After adjustment for baseline BMI z-scores, associations between total HC from starch or sugar and high BMI z-scores at two-year follow-up did not persist. Our observations offer a perspective on optimal timing for macronutrient consumption, which is known to be influenced by circadian rhythms. Reduced carbohydrate intake, especially during morning and midday meals, and following nocturnal sleep duration recommendations are two modifiable factors that may protect children from being overweight in the future.