Development and validation of a food frequency questionnaire.

Regular diet monitoring requires a tool validated in the target population. A 73-item, semiquantitative, self-administered food frequency questionnaire (FFQ), was adapted in French and English from the Block National Cancer Institute Health Habits and History Questionnaire. The FFQ was used to capture usual long-term food consumption among adults living in Quebec. A representative sample of adults aged 18 to 82 (57% female) was recruited by random digit dialling in the Montreal region. Approximately 64% of recruits completed and returned the instrument by mail (n=248). The FFQ was validated in a subsample (n=94, 61% female) using four nonconsecutive food records (FRs). Median energy intakes (in kcal) for men and women, respectively, were FFQ (total sample) 2,112 and 1,823, FFQ (subsample) 2,137 and 1,752, and FR (subsample) 2,510 and 1,830. Spearman correlation analyses between FFQ and FR nutrients were positive (with r ranging from 0.32 for folate to 0.58 for saturated fatty acids) and statistically significant (p<0.001), with better results among women. On average, cross-classification of energy and 24 nutrients from the FFQ and means of four FRs placed 39% into identical quartiles and 78% into identical and contiguous quartiles, with only 4% frankly misclassified. These results suggest that the FFQ is a relatively valid instrument for determining usual diet in Quebec adults.