Objectives: This study was conducted to investigate the systematic error, such as seasonal change or inadequate food items, in a food frequency questionnaire administered to workers in a Nuclear Power Plant, Korea.
Methods: We performed three repeat-tests with 28 subjects on May 13, July 8 and Dec 16, 1992. Our food frequency questionnaire (FFQ) comprised 84 foods organized into 7 food-groups, and was composed of the items of usual intake frequency (8 categories) and the amount per intake (3 or 4 categories) over the previous year. We compared the means of intake frequency and the frequency of the portion-size according to each season using Repeated Measures ANOVA and Pearson's chi-square test with Fisher's exact test.
Results: We found the significant seasonal changes of several food items in intake frequency measurement. These items were typical seasonal foods such as mandarin orange, plum and green vegetables, while the single questions consisted of inadequate food items such as thick beef or similar soup and various kimchi products. Significant seasonal changes in portion-size were found in only two items: cooked rice-brown and fresh/frozen fishes.
Conclusions: The systematic errors observed could caused loss of validity in the FFQ. Consideration should be given for seasonal variation in FFQ survey and methodological concerns are needed to improve the quality for measuring usual diet pattern.