Comparative validity of the ASSO–Food Frequency Questionnaire for the web-based assessment of food and nutrients intake in adolescents

BACKGROUND: A new web-based food frequency questionnaire (the ASSO–FFQ) was developed within the ASSO Project funded by the Italian Ministry of Health. OBJECTIVE: The aim of the present study is to assess the validity of the ASSO–FFQ at food groups, energy, and nutrients level. DESIGN AND SUBJECTS: The validation study compared the ASSO–FFQ against a weighted food record (WFR) measuring foods, beverages and supplements intake, compiled during the week following the ASSO–FFQ administration. Ninety-two subjects aged 14–17, recruited from secondary schools in Palermo (Italy), completed the ASSO–FFQ and WFR. The intake of 24 food groups, energy, and 52 nutrients were taken as main outcomes. Tests for paired observations, Spearman and Pearson’s correlation coefficients (cc), kappa statistics and classification in quintiles, Bland–Altman plots and multiple regressions, on untransformed and transformed data were used for the statistical analysis. RESULTS: High cc (≥0.40) were found for soft drinks, milk, tea/coffee, vegetables, and lactose; fair energy-adjusted cc (0.25–0.40) for water, alcoholic drinks, breakfast cereals, fishery products, savory food, fruit juice, eggs, and 19 nutrients. The subjects classified in the same or adjacent quintile for food groups ranged from 40% (alcoholic drinks) to 100% (dried fruit); for energy and nutrients from 43% (phosphorus, thiamin, niacin) to 77% (lactose). Mean differences were not significant for water, soft drinks, meat, sweets, animal fats, milk and white bread, and vitamin B(12) and folate. Limits of Agreement were broad for all food groups and nutrients. School, gender, alcohol consumption and between meals mainly affected most food groups’ intake differences. Gender stratification showed females had increased Pearson’s cc for energy and 28 nutrients, such as almost all fats, carbohydrates, vitamins and minerals. CONCLUSIONS: The ASSO–FFQ could be applied in epidemiological studies for the assessment of dietary consumption in adolescents to adequately rank food, energy and nutrient intakes at a group level.