Hypothesis: parents may selectively restrict sugar-containing foods for pre-school children with a high BMI

Abstract

Sugar-containing foods are popular among children, but often perceived as contributing to obesity. This study attempted to identify any associations between dietary intake of sugar-containing foods, or NME sugars' and body mass index (BMI) in young children participating in the National Diet and Nutrition Survey of Children aged 1 1/2 to 4 1/2 years. Dietary data were obtained from 4-day weighed dietary diaries completed by the child's caregiver. Associations between BMI and the dietary intake of energy, macronutrients and sugar-containing foods were examined in 1586 children. The sugar-containing foods investigated were: biscuits/cakes, confectionery, soft drinks, table sugar/preserves and breakfast cereals, all expressed as a percentage of dietary energy. Associations between BMI and macronutrients were generally weak, although statistically significant because of the large sample. In age-adjusted partial correlations, BMI was positively correlated with energy (r = 0.12; P < 0.0001) and percentage energy from protein (r = 0.09; P = 0.001), and weakly but inversely associated with the percentage of energy from non-milk extrinsic sugars (r = -0.05; P = 0.04). Thus, heavier children tended to have the lowest concentration of sugars in their diets. Similarly, children whose diets contained proportionally less biscuits and cakes, or sugar and preserves, had the highest average BMI (P < 0.05). Children with a high BMI (top quartile) consumed diets proportionately lower in sugar-containing foods than other children, despite similar or higher total energy intakes. These results are consistent with previous data that fail to support a positive link between BMI and intake of sugars or sugar-containing foods. It is hypothesised that, even in this young age group, some selective avoidance of sugar, biscuits and cakes may be practised, and this may be more evident among middle to upper social class households.