Validity and reproducibility of a food frequency Questionnaire among Chinese women in Guangdong province.

OBJECTIVE: To evaluate the validity and reproducibility of a food frequency questionnaire (FFQ) developed for assessing the association between dietary factors and breast cancer risk among Chinese women in Guangdong. METHODS: 61 women (24-64 years) were recruited from the community in Guangzhou city. An 81-item FFQ was administered twice, one year apart (FFQ1, FFQ2). In the mean time, six 3-day dietary records (DRs) were collected at two month intervals within the year. Daily consumption of nutrients and foods from the FFQs and DRs, correlation coefficients between the two FFQs and the FFQ with DRs were calculated. RESULTS: Median intakes of nutrients and food group items are higher in FFQ1 than FFQ2. The energy-adjusted Pearson correlation coefficients between the FFQ1 and FFQ2 ranged from 0.46 to 0.71 for nutrients and 0.36 to 0.66 for food group items, respectively. In the validation study, energy-adjusted correlation coefficients were 0.25 to 0.65 for nutrients and 0.30 to 0.68 for food groups. Mean proportion of subjects being classified into the same quartile of nutrients and foods intake from the FFQ and DRs was 36% and 43%, respectively. Mean misclassification of subjects into opposite quartiles was 5% for nutrients and 3% for foods. Bland-Altman analysis showed that no linear trend existed between the differences and means for nutrients. CONCLUSIONS: The 81-item FFQ has satisfactory reproducibility and reasonable validity, and is useful in assessing the usual consumption of major nutrients and food groups among Chinese women in Guangdong.