Murakami Kentaro, Sasaki Satoshi, Takahashi Yoshiko, Okubo Hitomi, Hirota Naoko, Notsu Akiko, Fukui Mitsuru, Date Chigusa
Nutritional Epidemiology Program, National Institute of Health and Nutrition, Toyama 1-23-1, Shinjuku-ku, Tokyo 162-8636, Japan.
Br J Nutr. 2008 Mar;99(3):639-48. doi: 10.1017/S0007114507812086. Epub 2007 Sep 3.
Although many epidemiological studies have examined the association of dietary glycaemic index (GI) and glycaemic load (GL) with health outcomes, information on the reproducibility and relative validity of these variables estimated from dietary questionnaires is extremely limited. We examined the reproducibility and relative validity of dietary GI and GL assessed with a self-administered diet-history questionnaire (DHQ) in adult Japanese. A total of ninety-two Japanese women and ninety-two Japanese men aged 31-76 years completed the DHQ (assessing diet during the preceding month) and 4 d dietary records (DR) in each season over a 1-year period (DHQ1-4 and DR1-4, respectively) and the DHQ at 1 year after completing DHQ1 (DHQ5). We used intraclass correlations between DHQ1 and DHQ5 to assess reproducibility, and Pearson correlations between the mean of DR1-4 and mean of DHQ1-4 and between the mean of DR1-4 and DHQ1 to assess relative validity. Reproducibility correlations for dietary GI and GL were 0.57 and 0.69 among women and 0.65 and 0.58 among men, respectively. Validity correlations for dietary GI and GL assessed by DHQ1-4 were 0.72 and 0.66 among women and 0.65 and 0.71 among men, respectively. Corresponding correlations for DHQ1 were 0.53 and 0.58 among women and 0.57 and 0.60 among men, respectively. White rice was the major contributor to GI and GL in both methods (49-64 %). These data indicate reasonable reproducibility and relative validity of dietary GI and GL assessed by a DHQ for Japanese adults, whose dietary GI and GL are primarily determined by the GI of white rice.
尽管许多流行病学研究已考察了膳食血糖生成指数（GI）和血糖负荷（GL）与健康结局之间的关联，但通过膳食问卷估计的这些变量的可重复性和相对有效性方面的信息极为有限。我们在成年日本人中考察了用一份自填式饮食史问卷（DHQ）评估的膳食GI和GL的可重复性和相对有效性。共有92名31 - 76岁的日本女性和92名日本男性在1年时间里的每个季节都完成了DHQ（评估前一个月的饮食）和4天饮食记录（DR）（分别为DHQ1 - 4和DR1 - 4），并在完成DHQ1后1年完成了DHQ（DHQ5）。我们使用DHQ1和DHQ5之间的组内相关性来评估可重复性，使用DR1 - 4的均值与DHQ1 - 4的均值之间以及DR1 - 4的均值与DHQ1之间的Pearson相关性来评估相对有效性。膳食GI和GL的可重复性相关性在女性中分别为0.57和0.69，在男性中分别为0.65和0.58。通过DHQ1 - 4评估的膳食GI和GL的有效性相关性在女性中分别为0.72和0.66，在男性中分别为0.65和0.71。DHQ1的相应相关性在女性中分别为0.53和0.58，在男性中分别为0.57和0.60。在两种方法中，白米都是GI和GL的主要贡献者（49 - 64%）。这些数据表明，对于日本成年人，通过DHQ评估的膳食GI和GL具有合理的可重复性和相对有效性，其膳食GI和GL主要由白米的GI决定。
