Objective: To assess the validity and reproducibility of a food frequency questionnaire with 82-food items (FFQW82) for nutrition education to modify eating habits.
Methods: The validity of nutrient intake assessed by FFQW82 was evaluated with seven-day diet records (7d-DR) as the gold standard, with questionnaire being administered twice with a one month interval to assess reproducibility. Twenty nine men (ages 42-63) and sixty women (ages 35-53) completed both surveys. Daily consumption of energy for 12 food groups and 9 nutrients was calculated from the FFQW82 and the 7d-DR with respect to breakfast, lunch, dinner, and whole-day meals. A Pearson correlation coefficient was obtained for each using log-transformed data.
Results: The estimated energy intakes from FFQW82 were higher than those from 7d-DR for men and women (relative differences of the median were 7% and 15%, respectively). As for validity, Pearson correlation coefficients of total energy intakes were 0.61 for men and 0.47 for women. Those for breakfast and lunch ranged from 0.66 to 0.89, while those for dinner were 0.19 and 0.26 for men and women. The daily intakes of nine nutrients ranged from 0.28 (potassium) to 0.65 (carbohydrate) in men and from 0.39 (fat) to 0.59 (calcium) in women. Regarding reproducibility, Pearson correlation coefficients for whole-day total energy intake were 0.65 for men and 0.69 for women. Those for the other nutrients ranged from 0.46 (salt) to 0.70 (carbohydrate) in men and from 0.59 (fat) to 0.70 (salt) in women.
Conclusion: The results suggest that FFQW82 could be used as a tool to obtain information concerning unbalance energy intake with reference to individual meals and to understand appropriate intake of the various food groups for nutrition education of middle-aged men and women.