Validity and reproducibility of a food frequency questionnaire (FFQ) for dietary assessment in Malay adolescents in Malaysia.

BACKGROUND Food frequency questionnaire (FFQ) must be tailored to the target populations because dietary habits vary within the populations due to differences in cultural and lifestyles practices. Limited information is available to assess the validity of FFQ used among Malaysian adolescents. OBJECTIVE To construct the validity and reproducibility of a newly developed FFQ in assessing habitual nutrients intake over the past year of 170 Malay adolescent boys and girls in Kelantan, Malaysia. METHODS The FFQ that consisted of 124 food items was assessed, whereas three days of 24-hours dietary recalls (DR) was administered as the standard criteria method. RESULTS Estimated mean intake for most nutrients assessed by the FFQ were higher as compared to the three DRs (p<0.05). Pearson correlation coefficients for energy-adjusted nutrients intake between both methods were ranged from 0.22 (zinc) to 0.68 (calcium) with a median r-value of 0.43. The cross classification of quartile analysis showed that most nutrients were classified into the same or adjacent quartiles (median=52.7%). For the reproducibility of FFQ, the correlation of nutrients ranged from 0.43 for carotene to 0.86 for total fat intake (median=0.67), after adjusting for total energy intake. CONCLUSION The newly developed dietary FFQ is a relatively good and valid tool in assessing habitual nutrients intake for the past year among Malay adolescents in Malaysia.