Introduction We have executed a validity test on our short food frequency questionnaire (FFQ) in “middle-aged” Japanese, but not “young women” yet, and here examined.

Methods One hundred and twenty five young women in Aichi Prefecture, Japan, completed the FFQ and 3 day-weighed diet records (3d-WDRs) in 2009. We compared intakes of energy and 26 nutrients computed with the FFQ against those with 3d-WDRs as reference.

Results Participant's means (SD) of age and BMI were 20.3 (0.5) years and 23.3 (2.3) kg/m², respectively. No significant difference was found between energy intakes of 1579 (221) and 1576 (258) kcal/day according to the FFQ and 3d-WDRs. Pearson's correlation coefficients between each nutrient intake derived from the FFQ and 3d-WDRs were 0.29 (energy), 0.18 (protein), 0.26 (fat), 0.28 (carbohydrate), 0.53 (calcium), 0.20 (iron), 0.21 (sodium), 0.33 (potassium), 0.31 (carotene), 0.30 (Vitamin A), 0.36 (Vitamin D), 0.36 (soluble dietary fibre), 0.32 (insoluble dietary fibre), 0.31 (total dietary fibre) and 0.23 (n-3 polyunsaturated fatty acids), respectively. Median percentages for exact agreement and disagreement according to quartile classification of each nutrient intake measured with both methods were 25.6 and 6.4, respectively.

Conclusions Similarly to our previous study, we showed considerably high relative validity values for consumption of macro- and micro-nutrients estimated with the FFQ vs those assessed with 3d-WDRs. The FFQ therefore appears applicable for categorising individuals according to consumption of energy and the selected nutrients in dietary studies of young Japanese women.