Research Foundation Flanders, Department of Public Health, Ghent University, Ghent, Belgium.
J Hum Nutr Diet. 2010 Oct;23(5):502-10. doi: 10.1111/j.1365-277X.2009.01038.x. Epub 2010 Feb 15.
It is important to assess the dietary habits of children as they progress from early childhood to adolescence. However, there are few validated tools available to assess the diet of this group of children. The present study aimed to investigate the relative validity of a food frequency questionnaire (FFQ) developed for use in a biennial longitudinal study.
The parents of 216 Belgian-Flemish preschool children completed a FFQ (including questions on 77 food groups) and an online dietary assessment tool for three nonconsecutive days on their child's diet. Intakes of energy, macronutrients, fibre, calcium and 20 food groups recorded by both methods were compared using Spearman's correlations, Wilcoxon's signed rank test and Bland and Altman analyses.
At the group level, good agreement was found for energy, fat and protein intake, an overestimation was found for carbohydrates (5.6%) and fibre (13.3%), and an underestimation was found for calcium (9%). For the food groups, overestimation as well as underestimation and good agreement were found, although overestimation was most common. Good agreement was found in ranking the children's fruit (r = 0.53), beverages (r = 0.58-0.76), soup (r = 0.51), breakfast cereals (r = 0.55) and calcium (r = 0.59) intake according to both methods.
The results obtained revealed that the FFQ was a useful alternative for estimating energy and macronutrient intake at a group level; however, when used to estimate fibre and calcium intake, respectively, over- and underestimation need to be considered. The short reference period (3 days) may have compromised agreement in ranking and agreement at the individual level.
评估儿童从幼儿期到青春期的饮食习惯非常重要。然而，目前可用的评估这组儿童饮食的工具很少。本研究旨在调查为两年一次的纵向研究开发的食物频率问卷（FFQ）的相对有效性。
216 名比利时佛兰芒学龄前儿童的家长完成了一份 FFQ（包括 77 种食物组的问题）和一份在线饮食评估工具，用于记录他们孩子三天的非连续饮食。使用 Spearman 相关系数、Wilcoxon 符号秩检验和 Bland 和 Altman 分析比较两种方法记录的能量、宏量营养素、纤维、钙和 20 种食物组的摄入量。
在群体水平上，发现能量、脂肪和蛋白质的摄入量具有良好的一致性，碳水化合物（5.6%）和纤维（13.3%）的摄入量被高估，钙（9%）的摄入量被低估。对于食物组，发现存在高估、低估和良好的一致性，尽管高估最常见。两种方法都发现水果（r = 0.53）、饮料（r = 0.58-0.76）、汤（r = 0.51）、早餐谷类（r = 0.55）和钙（r = 0.59）的摄入量的排名具有良好的一致性。
研究结果表明，FFQ 是一种有用的替代方法，可以在群体水平上估计能量和宏量营养素的摄入量；然而，当分别用于估计纤维和钙的摄入量时，需要考虑高估和低估的情况。较短的参考期（3 天）可能会影响个体水平的一致性。
