Objective: To describe the lunchtime choices and nutritional intake of primary-school-aged children in England 4 months after the introduction of interim food-based standards for school lunches.
Design: Cross-sectional 2 d weighed food records collected in January and February 2007.
Setting: Six primary schools in Sheffield, England.
Subjects: One hundred and twenty-three pupils aged 8-10 years.
Results: Vegetables (81 % v. 8 %) and cakes and biscuits (43 % v. 23 %) were chosen more frequently by pupils consuming a school lunch, while fruit (40 % v. 36 %), meat products (18 % v. 14 %), confectionery (72 % v. 0 %), savoury snacks (69 % v. 0 %) and drinks not meeting the school food standards (40 % v. 0 %) were chosen more often by pupils eating a packed lunch. Mean energy intake was lower in the school lunch group compared with the packed lunch group (1402 (sd 573) v. 2192 (sd 619), P = 0·005). Nutrient density (per MJ energy) was significantly better in school meals for key nutrients including protein (9·8 (sd 2·7) v. 6·3 (sd 1·9) g), fat (7·4 (sd 2·7) v. 10·6 (sd 2·8) g), NSP (2·8 (sd 1·3) v. 1·1 (sd 0·4) g), vitamin A (151·3 (sd 192·8) v. 69·1 (sd 55·6) μg), folate (29·6 (sd 11·6) v. 17·0 (sd 7·0) μg), iron (1·3 (sd 0·3) v. 0·9 (sd 0·3) mg) and zinc (1·1 (sd 0·4) v. 0·7 (sd 0·3) mg).
Conclusions: Schools were largely compliant with the interim food-based standards for school meals 4 months after their introduction. Within the context of the new standards, children taking a school lunch are more likely to eat a more nutritious lunch, in terms of less high-fat/salt/sugar foods and nutrient density. The introduction of nutrient-based standards is warranted. Efforts to improve the lunchtime intake of children taking a packed lunch are also required.