Open Access Research

Development of a food frequency questionnaire to estimate habitual dietary intake in Japanese children

Tomomi Kobayashi1*, Sanae Tanaka2, Chihiro Toji1, Hideko Shinohara1, Miharu Kamimura1, Naoko Okamoto3, Shino Imai1, Mitsuru Fukui4 and Chigusa Date3

Author affiliations

1 Department of Food and Nutritional Sciences, School of Natural Science and Ecological Awareness, Graduate School of Humanities and Sciences, Nara Women's University, Kitauoya-nishimachi, Nara, 630-8506, Japan

2 Department of Food Science and Nutrition, Human Environmental Sciences, Mukogawa Women's University, 6-46 Ikebiraki-cho, Nishinomiya, Hyogo 663-8558, Japan

3 Department of Food Sciences and Nutrition, Faculty of Human Life and Environment, Nara Women's University, Kitauoya-nishimachi, Nara, 630-8506, Japan

4 Laboratory of Statistics, School of Medicine, Osaka City University, 1-4-3 Asahi-machi, Abeno-ku, Osaka, 545-8585, Japan

For all author emails, please log on.

Citation and License

Nutrition Journal 2010, 9:17  doi:10.1186/1475-2891-9-17

Published: 10 April 2010

Abstract

Background

Food frequency questionnaires (FFQ) are used for epidemiological studies. Because of the wide variations in dietary habits within different populations, a FFQ must be developed to suit the specific group. To date, no FFQ has been developed for Japanese children. In this study, we developed a FFQ to assess the regular dietary intake of Japanese children. The FFQ included questions regarding both individual food items and mixed dishes.

Methods

Children (3-11 years of age, n = 621) were recruited as subjects. Their parents or guardians completed a weighed dietary record (WDR) for each subject in one day. We defined FOOD to be not only as a single food item but also as a mixed dish. The dieticians conceptually grouped similar FOODs as FOOD types. We used a contribution analysis and a multiple regression analysis to select FOOD types.

Results

We obtained a total of 586 children's dietary data (297 boys and 289 girls). In addition, we obtained 1,043 FOODs. Dieticians grouped into similar FOODs, yielding 275 FOOD types. A total of 115 FOOD types were chosen using a contribution analysis and a multiple regression analysis, then we excluded overlapping items. FOOD types that were eaten by fewer than 15 subjects were excluded; 74 FOOD types remained. We also added liver-based dishes that provided a high amount of retinol. A total of 75 FOOD types were finally determined for the FFQ. The frequency response formats were classified into four type categories: seven, eight, nine and eleven, according to the general intake frequency of each FOOD type. Information on portion size was obtained from the photographs of each listed FOOD type in real scale size, which was the average amount of the children's portion sizes.

Conclusions

Using both a contribution analysis and a multiple regression analysis, we developed a 75-food item questionnaire from the study involving 586 children. The next step will involve the verification of FFQ reproducibility and validity.