Evaluation of food intake through residual analysis in 90 Basque school canteens.

OBJECTIVE: To measure the consumption of 25 food items (Comstock scale), to test whether school management type (public schools vs. private with subsidy schools) and food origin (on-site preparation vs. catering) affected food consumption and to quantify variability associated with territory, school and individual. METHOD: This cross-sectional study involved 14,717 schoolchildren of ages 2-16, recruited in 90 schools of the Basque Autonomous Community (Spain). Waste (non-consumed food) of 25 analysed food items was visually estimated by trained school monitors, via the Comstock categorical scale thus adapted: 1=0%; 2=25%; 3=50%, 4=75% and 5=100%. To analyse the data, mixed modelling was applied. RESULTS: Vegetables, fish and fruits were the less-accepted food types. While school management did not affect food intake, on-site food preparation had better acceptance for legumes with vegetables, oily and lean fish, meat and pre-cooked meals. The largest source of variability in food intake not accounted for by school management and food preparation was the individual subject, while school and territory had moderate and almost no effects, respectively. CONCLUSIONS: Acceptance of the 25 evaluated foods is adequate, albeit can be improved. We believe that promoting on-site food preparation should improve the acceptance of legumes with vegetables, oily/lean fish, meat and pre-cooked meals. We recommend that future interventions oriented to improve intake should focus on individual subjects.