Reproducibility and validity of the Toronto-modified Harvard food frequency questionnaire in a multi-ethnic sample of young adults

Abstract

Background/objectives: To assess the reproducibility and validity of a Toronto-modified Harvard food frequency questionnaire (FFQ) among a multi-ethnic sample of young adults.

Subjects/methods: A total of 150 participants recruited from the Toronto Nutrigenomics and Health Study cohort who had existing dietary intakes assessed by FFQ (FFQ1) and reassessment one year later (FFQ2). Of these, 100 participants also completed a three-day food record to evaluate the validity of the FFQ for 38 nutrients (energy, 14 macronutrients, 22 micronutrients, and 1 bioactive). Analyses were also stratified between the two major ethnic groups (Caucasian and East Asian).

Results: Among the full sample, mean intakes of most nutrients (27/38) did not differ significantly between estimates derived from FFQ2 compared to the three-day food record. Energy, sex, and ethnicity adjusted deattenuated Pearson correlation coefficients ranged from 0.20 to 0.92 (mean r = 0.52 ± 0.15), and 34/38 validity coefficients were r ≥ 0.32. Gross misclassification of intakes between FFQ2 and the three-day food record was low (<6%), but energy, polyunsaturated fatty acids (PUFA), and sodium were underestimated by FFQ2. Mean intakes between FFQ1 and FFQ2 did not differ significantly for any nutrient. Between the two major ethnic groups, mean validity coefficients were similar, but varied for individual nutrients with saturated fat, PUFA, and omega 3 being among the most discrepant.

Conclusions: Compared to a three-day food record, the Toronto-modified Harvard FFQ is a reproducible and valid tool to estimate dietary intake among a multi-ethnic sample of young adults. However, incorporation of protocols to improve the assessment of culturally diverse diets should be considered.