Correlations between Fruit, Vegetables, Fish, Vitamins, and Fatty Acids Estimated by Web-Based Nonconsecutive Dietary Records and Respective Biomarkers of Nutritional Status

Abstract Background It is of major importance to measure the validity of self-reported dietary intake using web-based instruments before applying them in large-scale studies. Objective This study aimed to validate self-reported intake of fish, fruit and vegetables, and selected micronutrient intakes assessed by a web-based self-administered dietary record tool used in the NutriNet-Sante prospective cohort study, against the following concentration biomarkers: plasma beta carotene, vitamin C, and n-3 polyunsaturated fatty acids. Participants/setting One hundred ninety-eight adult volunteers (103 men and 95 women, mean age=50.5 years) were included in the protocol: they completed 3 nonconsecutive-day dietary records and two blood samples were drawn 3 weeks apart. The study was conducted in the area of Paris, France, between October 2012 and May 2013. Main outcome measures Reported fish, fruit and vegetables, and selected micronutrient intakes and plasma beta carotene, vitamin C, and n-3 polyunsaturated fatty acid levels were compared. Statistical analyses Simple and adjusted Spearman's rank correlation coefficients were estimated after de-attenuation for intra-individual variation. Results Regarding food groups in men, adjusted correlations ranged from 0.20 for vegetables and plasma vitamin C to 0.49 for fruits and plasma vitamin C, and from 0.40 for fish and plasma c20:5 n-3 (eicosapentaenoic acid [EPA]) to 0.55 for fish and plasma c22:6 n-3 (docosahexaenoic acid). In women, correlations ranged from 0.13 (nonsignificant) for vegetables and plasma vitamin C to 0.41 for fruits and vegetables and plasma beta carotene, and from 0.27 for fatty fish and EPA to 0.54 for fish and EPA+docosahexaenoic acid. Regarding micronutrients, adjusted correlations ranged from 0.36 (EPA) to 0.58 (vitamin C) in men and from 0.32 (vitamin C) to 0.38 (EPA) in women. Conclusions The findings suggest that three nonconsecutive web-based dietary records provide reasonable estimates of true intake of fruits, vegetables, fish, beta carotene, vitamin C, and n-3 fatty acids. Along with other validation studies, this study shows acceptable validity of using such diet-assessment methods in large epidemiologic surveys and broadens new perspectives for epidemiology.