Validity and reproducibility of a food frequency questionnaire among Lebanese adults for the assessment of food groups.

Abstract

Background: Diet is a major modifiable risk factor in the etiology of chronic diseases. Dietary assessment is a critical step in evaluating diet-disease association and in formulating recommendation for dietary interventions. To date, there exists no valid dietary assessment tool for adults in Lebanon. The purpose of this study was to develop and examine the validity and reliability of a Food Frequency Questionnaire (FFQ) for the assessment of dietary intake in relation to food groups among Lebanese adults. Methods: Healthy Lebanese adults of both genders, aged between 18 and 65 were recruited from various faculties and offices at the American University of Beirut (AUB), Lebanon (n=120) by convenient sampling. Participants completed the FFQ twice, during a one year interval. Within each of the four seasons of this one year, three 24-hour recalls (24-HRs) representing two weekdays and one weekend were collected. Sociodemographic, lifestyle questionnaire and anthropometric measurements were also collected. For the development of the FFQ, a list of 94 culturally appropriate food items were included and they were categorized into 23 food groups with 4 frequency options (daily, weekly, monthly, never). Food items were analyzed into food groups based on usage or similarities in composition. Similar grouping was conducted for the 24-HR. Calorie estimation of the various food groups was conducted using Nutritionist Pro software. For validity, dietary intake data collected via the FFQ was compared to the mean of repeated 24-HR recalls (MPRs). Percent difference in means, spearman rank correlations and Bland-Altman plots were used to assess the validity of the developed FFQ. Intra-class Correlation Coefficient (ICC), percent agreement, and weighted kappa (κ w) were used to evaluate the FFQ’s reproducibility. Statistical Package for Social Sciences (SPSS) was used in the analysis and a p0.05 indicated significance. Results: Out of the 120 adults, 110 completed the study (dropout rate: 8.3percent). The mean (