[Relative validity of food frequency questionnaire for estimating dietary iron intake of students with anemia]

To assess the relative validity of food frequency questionnaire (FFQ) for estimating dietary iron and its related nutrients and food intakes of students with anemia.

FFQ and weighed dietary records combining with 24h dietary recalls (adopted as reference method) were applied respectively to investigate the diet of 156 anemic students aged 10 - 12 years. The average dietary intakes of iron, haem iron, non-haem iron, vitamin C, meat/fish/poultry and eggs were estimated. Wilcoxon matched-pairs signed-rank test, Spearman's rank correlation, cross-classification and actual values for surrogate categories were performed to assess the relative validity of FFQ.

Compared with the reference method, dietary iron and its related nutrients and food intakes were overestimated by FFQ, and bear poor correlation with reference method. The results of cross-classification and actual values for surrogate categories indicated that the low and high intakes of all dietary components could not be clearly differentiated by FFQ.

FFQ presents unsatisfactory accuracy in assessing dietary intakes of iron and related nutrients of anemic students in less developed districts.