Validation of the web-based dietary assessment tool (RiksmatenFlex) against doubly labelled water and 24-h dietary recalls in Swedish pre-school children
摘要
RiksmatenFlex, a Swedish’ web-based dietary assessment method, has previously been validated in adolescents and pregnant women. The aim of this study was to evaluate RiksmatenFlex in preschool children by comparing energy intake with total energy expenditure (TEE) measured by the doubly labelled water (DLW) method, by comparing reported intakes of energy, foods, macro- and micronutrients with 24-h dietary recall interviews, and by assessing the parents’ perceptions of reporting their child’s dietary intake.
MethodsParticipants were 94 healthy children, 37 ± 15 months old, 51% girls, recruited through advertisements (preschools/social media). Energy, selected nutrient and food group intakes were assessed using RiksmatenFlex as a food diary and 24-h dietary recall interviews. TEE (DLW) was measured in a subsample of 37 children.
ResultsAverage energy intake was overestimated by RiksmatenFlex compared to TEE (796 kJ; 21%; P < 0.001) in the subsample. The Bland-Altman plot showed wide limits of agreement (± 1797 kJ), and bias towards overreporting at higher intake levels (r = 0.34; P = 0.038). Energy intake and absolute intakes of protein, fat, carbohydrates, calcium (also energy standardised) and fruit-vegetables (combined) were significantly higher with RiksmatenFlex than with the 24-h dietary recalls. Other nutrient and food group intakes were fairly similar with the two methods. Spearman rank correlations ranged from 0.73 (wholegrains) to 0.98 (baby foods). Agreement between the two methods was substantial (KW>0.61) for most dietary variables, but not wholegrains, and proportion of energy from macronutrients (KW 0.50–0.60). Parents found reporting food intake in RiksmatenFlex easy, with most preferring on-line reporting over interviews.
ConclusionThe current study showed reasonable validity in young children and the method was well accepted by parents. These finding support its application in large-scale studies of young children, though potential overreporting of energy intake warrants attention.