ABSTRACT
Background: Food frequency questionnaire (FFQ) is the most frequently used dietary assessment method in estimating dietary intakes in epidemiological studies. Aim: This study aimed to assess the relative validity of a semiquantitative FFQ in evaluating dietary intake among pregnant women in the United Arab Emirates. Methods: Within the Mutaba'ah study, a subsample of 111 pregnant women completed a semiquantitative FFQ and a single 24-hour dietary recall (24-HDR) regarded as the reference method. Absolute and energy-adjusted nutrient and food intakes between the FFQ and 24-HDR were compared using the Wilcoxon signed ranks test, correlations, Bland-Altman analysis, cross-classification, and weighted kappa analysis. Results: There were no significant differences in reported absolute intakes between the FFQ and 24-HDR for carbohydrates, whole grains, white meat, beta-carotene, vitamin K, sodium, and selenium. Spearman's correlation coefficients between the FFQ and 24-HDR ranged from 0.09 (trans fatty acids) to 0.5 (potassium) for absolute intakes. Correlation decreased after energy adjustment. Bland-Altman analysis showed that the FFQ overestimated intakes compared with 24-HDR and that the limits of agreement were wide. The average percentage of pregnant women classified into the same or adjacent quartile of intake by both methods was 73%. Weighted kappa values ranged from -0.02 (white meat) to 0.33 (magnesium). Conclusion: Our findings showed that the semi-quantitative FFQ is a useful tool in ranking pregnant women from the Emirati population according to their dietary intake. However, the validity of some estimated intakes was poor; hence, certain intakes should be interpreted with caution.
ABSTRACT
BACKGROUND: The choline oxidation pathway and metabolites involved have been linked to diseases including cardiovascular disease, type 2 diabetes and cancer. A healthy Nordic diet is a recently defined dietary pattern associated with decreased risk for these diseases. Our aim was to explore associations between adherence to a healthy Nordic diet and plasma concentrations of metabolites of the choline oxidation pathway. METHODS: The Healthy Nordic Food Index (HNFI) and Baltic Sea Diet Score (BSDS) were applied to cross-sectional data (n = 969) from the Västerbotten Intervention Programme in Northern Sweden to score adherence to a healthy Nordic diet. Data included responses to a dietary questionnaire and blood sample analyses (1991-2008). Associations of diet scores with plasma concentrations of metabolites of the choline oxidation pathway and total homocysteine (tHcy), seven metabolites in total, were evaluated with linear regression, adjusting for age, BMI, education and physical activity. RESULTS: HNFI scores showed linear relationships with plasma choline (ß = 0.11), betaine (ß = 0.46), serine (ß = 0.98) and tHcy (ß = - 0.38), and BSDS scores with betaine (ß = 0.13) and tHcy (ß = - 0.13); unstandardized beta coefficients, all significant at P < 0.05. The regression models predicted changes in plasma metabolite concentrations (± 1 SD changes in diet score) in the range of 1-5% for choline, betaine, serine and tHcy. No other statistically significant associations were observed. CONCLUSIONS: A healthy Nordic diet was associated with plasma concentrations of several metabolites of the choline oxidation pathway. Although relationships were statistically significant, effect sizes were moderate. Further research is warranted to explore the underlying mechanisms and associations with health outcomes.
