| Summary: | 碩士 === 輔仁大學 === 食品營養學系 === 93 === The purpose of this study was to test validity and reproducibility of a food frequency questionnaire (FFQ) for preschool children. The study recruited 48 children (33 boys, 15 girls) aged 3 years-old and above whose mothers were the main caretakers from kindergarten full-day program in Taipei, Tainan and Kaoshiung areas. In FFQ, Sixty-one Food items contributing 90% of the main nutrient intakes of preschool children based on the Infant and Preschool Children Nutrition Survey were selected. Itemized frequency selection was used, while the portion size was open-ended assisted with food models. Three-day food record, recorded by mothers at home and investigator at kindergarten, were used as a reference standard for FFQ validity. Reproducibility of FFQ was tested 1-month apart. Results showed that first FFQ overestimated overall nutrient intakes. Reproducibility of 2 FFQ interviews were highly correlated （r=0.55~0.82, P<0.01）, and Kappa were between 0.25～0.63. The correlations of nutrient intakes between food record and FFQ were ranged from 0.02 (vitamin B12) to 0.58 (calcium). Kappa ranged from -0.03 to 0.44. Test of the FFQ validity and reproducibility between Northern and Southern preschool children showed no significant differences indicated that the FFQ could be applied to preschool children living in different area of Taiwan. Validity of FFQ from mother was better than that from kindergarten teacher. There might be learning effects of low educated mothers from food recording which might enhance the validity of FFQ. Adjustment for variation-within-person enhanced the correlation of most nutrient intakes computed from FFQ with those recorded in the food record, especially for micronutrients. While energy-adjusted did not really alter the crude correlation of nutrient intakes from FFQ and food record except for the carbohydrate, protein and fat in first FFQ. SFFQ tended to reduce the overestimate-effect in the original FFQ. However, SFFQ did not show a better result in correlation and agreement with food record. The correlation increased after the energy and variation-within-person adjustment, especially in second FFQ. Both FFQ and SFFQ were suitable in assessing Taiwan preschool children nutrient intakes as a group.
|
|---|
