Khan Nguyen Cong, Mai Le Bach, Hien Vu Thi Thu, Lam Nguyen Thi, Hoa Vu Quynh, Phuong Tuan Mai, Nhung Bui Thi, Nakamori Masayo, Shimizu Yukue, Yamamoto Shigeru
National Institute of Nutrition, Vietnam.
J Nutr Sci Vitaminol (Tokyo). 2008 Apr;54(2):124-9. doi: 10.3177/jnsv.54.124.
This study was done to develop a Food Frequency Questionnaire (FFQ) for assessing calcium intake in Vietnamese based on data from the National Nutrition Survey in 2000. From the data, a total of 36 calcium-rich food items were selected for the FFQ by ranking food items according to their contribution to the population intake of nutrients. The FFQ was validated in 140 postmenopausal women by comparing estimates of calcium intake from the FFQ with those from multi-pass 24 h recalls. The results showed that the Pearson correlation coefficient between the 2 methods was 0.84 (p<0.001), and that the weighted kappa value was 0.44. Cross-classification analysis indicated that 82.1% of subjects were classified into the same or adjacent quartile by both methods. No subject was grossly misclassified by the FFQ. There is no significant difference of calcium intake between the 2 methods (by paired t test, p>0.05). A reproducibility study also presented good correlation between 2 administrations of the FFQ, with Pearson correlation coefficient being 0.93 (p<0.001) and weighted kappa value being 0.67. In conclusion, this FFQ is useful and reliable for estimating calcium intake in population-based epidemiological studies in postmenopausal Vietnamese women.
本研究旨在根据2000年全国营养调查数据，开发一份用于评估越南人钙摄入量的食物频率问卷（FFQ）。根据食物对人群营养素摄入量的贡献对食物进行排名，从数据中总共选择了36种富含钙的食物纳入FFQ。通过将FFQ得出的钙摄入量估计值与多次24小时回顾法得出的估计值进行比较，在140名绝经后妇女中对该FFQ进行了验证。结果显示，两种方法之间的Pearson相关系数为0.84（p<0.001），加权kappa值为0.44。交叉分类分析表明，82.1%的受试者在两种方法下被归类到相同或相邻的四分位数组。没有受试者被FFQ严重误分类。两种方法的钙摄入量无显著差异（配对t检验，p>0.05）。一项重复性研究还表明，FFQ的两次施测之间具有良好的相关性，Pearson相关系数为0.93（p<0.001），加权kappa值为0.67。总之，该FFQ对于在越南绝经后妇女的基于人群的流行病学研究中估计钙摄入量是有用且可靠的。
