ABSTRACT
Introduction: A culturally valid tool is lacking in order to assess vitamin D intake among Malaysians. This study aimed to develop and determine the validity and reliability of a semi-quantitative food frequency questionnaire (FFQ) for assessment of vitamin D intake among pregnant women living in urban areas of Malaysia. Method: A total of 79 volunteer mothers (aged 18 to 40 years) in their first trimester of pregnancy participated in this study voluntarily. They completed both the 3-day 24-hour dietary recalls (3DR) and semi-quantitative food frequency questionnaires (FFQ1). The FFQ1 was validated against vitamin D intake calculated from the 3DR. After one week, 41 (51.89%) subjects completed the same FFQ1 (now called FFQ2) for reliability assessment. Results: The median vitamin D intake from 3DR was 117.6 IU/ day with an interquartile range (IQR) of 54.8 to 193.7 IU/ day. The median and IQR values from FFQ1 were 147.7 and 103.7 to 233.9 IU/ day, respectively. The Spearman correlation coefficient was strong and positive, r=0.810 (p=0.01); cross-classification analyses revealed no misclassification and the Bland-Altman plot showed good agreement between the two dietary methods. The intra-class correlation (ICC) for reliability of vitamin D intake was 0.98; the cross-classification analysis showed 95.12% and was classified into the same quartile which supported a consistent habitual eating pattern. Conclusion: The new FFQ developed to assess vitamin D intake showed high correlation and high agreement with the majority of the sample population classified into the same quartile. It is concluded that the new FFQ is a valid and reliable screening tool for vitamin D intake for pregnant women in the urban areas of Malaysia.