卫生研究

目的 比较食物频率问卷(food frequency questionnaire, FFQ)与24小时膳食回顾法和称重法在评估孕妇食物、能量和营养素摄入量之间的异同，评估其效度。方法 2017年6—9月在河北省武强县全县范围内共招募82名孕妇作为调查对象，对调查对象分别开展最近一周FFQ调查，连续3天24小时膳食回顾调查，连续3天入户称重调查，计算并分别比较孕妇FFQ和其他两种方法获得的食物、能量和营养素日均摄入量的差异；通过Spearman相关系数和Bland-Altman图分别评估FFQ与其他两种方法的相关性和一致性。结果 FFQ与24小时膳食回顾法评估大豆及其制品、水果类、乳类及其制品和坚果类食物平均每日摄入量的差异无统计学意义，其中两种方法评估的水果类和乳类及其制品摄入量均呈正相关(r=0.34～0.56,P<0.01)。FFQ与称重法评估谷类、大豆及其制品、蔬菜类、水果类、水产类、蛋类、乳类及其制品及坚果类食物平均每日摄入量的差异无统计学意义，其中两种方法评估的谷类、蔬菜类和坚果摄入量均呈正相关(r=0.53～0.58,P<0.05)。FFQ与24小时膳食回顾法评价能量、蛋白质、碳水化合物、碳水化合物供能比、膳食纤维、总维生素A、硫胺素、核黄素、维生素C、钙、磷、钾、镁、铁及锌平均每日摄入量的差异无统计学意义，其中能量、蛋白质、脂肪和碳水化合物摄入量均呈正相关(r=0.41～0.58,P<0.05)。FFQ与称重法评价能量、脂肪、脂肪供能比、碳水化合物、碳水化合物供能比、总维生素A、硫胺素、核黄素、维生素C、钙、钾、钠及镁平均每日摄入量的差异无统计学意义，其中能量、蛋白质和碳水化合物两种方法所得摄入量均呈正相关(r=0.49～0.52,P<0.05)。Bland-Altman分析结果显示，无论是24小时膳食回顾法(91.1%～94.9%)还是称重法(92.9%～100%),大部分调查对象能量、蛋白质、脂肪和碳水化合物摄入量均在95%一致性界限范围内。结论 FFQ能够较准确评价孕妇食物与营养素摄入量。

OBJECTIVE This study aimed to compare the similarities and differences among Food Frequency Questionnaire(FFQ), 24-hour dietary recall method, and weighing method in assessing the intakes of foods, energy, and nutrients in pregnant women, and to evaluate the validity of the FFQ.METHODS From June to September in 2017, a total of 82 pregnant women were recruited across the entire Wuqiang County in Hebei Province. The subjects were surveyed using the FFQ for the most recent week, 24-hour dietary recall for 3 consecutive days, and in-home weighing for 3 consecutive days. The differences in the average daily intakes of food, energy, and nutrients obtained by the FFQ and the other two method were calculated and compared. The correlation and consistency between the FFQ and the other two method were evaluated through the Spearman correlation coefficient and the Bland-Altman plot, respectively.RESULTSThere was no statistically significant difference between the FFQ and 24-hour dietary recall method in assessing the average daily intakes of soybeans and their products, fruits, dairy products and nuts. The intakes of fruits and dairy products evaluated by the two method were all positively correlated(r=0.34-0.56,P<0.05). There was no statistically significant difference between the FFQ and the weighing method in assessing the average daily intakes of cereals, soybeans and their products, vegetables, fruits, aquatic products, eggs, dairy products and nuts. Among them, the intakes obtained by the two method for cereals, vegetables and nuts were all positively correlated(r=0.53-0.58,P<0.05). There was no statistically significant difference between the FFQ and 24-hour dietary recall method in evaluating average daily intakes of energy, protein, carbohydrates, the ratio of carbohydrates to energy supply, dietary fiber, total vitamin A, thiamine, riboflavin, vitamin C, calcium, phosphorus, potassium, magnesium, iron and zinc. Among them, intakes of energy, protein, fat and carbohydrates were all positively correlated(r=0.41-0.58,P<0.05). There was no statistically significant difference between FFQ and weighing method in evaluating the average daily intakes of energy, fat, the ratio of fat to energy supply, carbohydrates, the ratio of carbohydrates to energy supply, total vitamin A, thiamine, riboflavin, vitamin C, calcium, potassium, sodium and magnesium. Among them, the intakes of energy, protein and carbohydrates obtained by the two method were all positively correlated(r=0.49-0.52,P<0.05). Bland-Altman analysis showed that for most subjects, the intakes of energy, protein, fat and carbohydrates were within the 95% consistency limits between FFQ and 24-hour recall method(91.1%-94.9%)or weighing method(92.9%-100%).CONCLUSION FFQ can relatively accurately evaluate the intakes of food and nutrients among pregnant women.