Comparison of two frequency questionnaires for quantifying fruit and vegetable intake

Abstract

OBJECTIVE: The effect on individual rankings and total intakes of nutrients of correcting total fruit and vegetable frequencies from a long food frequency questionnaire (FFQ) using the responses to two summary questions was examined in a group of women. METHODS: The performance of a self-administered FFQ in ranking individual levels of intake and estimating absolute levels of nutrient and energy intake was compared with the performance of the questionnaire when it was corrected for fruit and vegetable intake reported using the Block summary questions. SUBJECTS: The study population included 123 women, aged between 18 and 54 years, who were recruited from the Family Planning Association Colposcopy Clinic in Sydney. RESULTS: Substantial and significant differences (P < 0.001) were found in fruit and vegetable intakes between the FFQ and the summary questions. Intake frequency by the FFQ was more than double that by the summary questions. When the FFQ was corrected for fruit and vegetable intakes using the summary questions, the intakes of beta-carotene, vitamins A and C, and dietary fibre were more than 20% lower (p < 0.001) than the uncorrected results. However, this had little effect on ranking individuals. This study also examined seasonal differences in vegetable intakes and differences in nutrient intakes when either summer or winter vegetable consumption was substituted for seasonal vegetable intake in the FFQ. Although there were seasonal differences for some foods, the substitution had little effect on intake of nutrients. CONCLUSION: These results indicate that important differences in intakes are observed when two methods, which appear to yield the same results, are used. Further work is needed to determine which, if either, of the two methods yields intakes that can be compared quantitatively with national references for assessing the adequacy of population intakes.