A validation study comparing energy and nutrient intake between a web-based food frequency questionnaire and a 4-d dietary record

Abstract

ObjectiveReliable assessments of dietary intake is crucial when examining associations between dietary factors and different outcomes. Today, web-based instead of paper-based food frequency questionnaires (FFQs) are common. A web-based FFQ implies a different experience (including design differences, the possibility only to show relevant questions, and automatic checks for completeness) and may capture dietary intake differently. The aim of this study was to validate a well-used paper-based FFQ transferred to a web-based version against a 4-d dietary record. MethodsBaseline data from a dietary intervention study comprising women and men with type 2 diabetes were used. Diet was assessed with the web-based FFQ and a 4-d dietary record. Energy intake and energy-adjusted nutrient intakes were assessed and compared using Spearman's rank correlation and Bland–Altman analysis. The analysis included 101 participants. ResultsThe average energy intake with the FFQ was 2068 kcal, 157 kcal higher than the average from the dietary record. Significant deattenuated correlations were found for intake of energy (r = 0.37; 95% confidence interval [CI],0.19–0.53), carbohydrates (r = 0.85; 95% CI, 0.78–0.89), protein (r = 0.53; 95% CI, 0.38–0.66) and fat (r = 0.59; 95% CI, 0.44–0.70). Significant deattenuated correlations (ranging r = 0.26–0.73) were found for intake of alcohol, whole grain, fiber, saturated fat, monounsaturated fat, polyunsaturated fat, cholesterol, vitamins A, B12, C, D, and E, folate, iron, potassium, magnesium, and calcium. Bland–Altman analyses showed no clear trends but wide limits of agreement. ConclusionOverall, the web-based FFQ performed well and is similar to the paper-based version. It can be a useful tool for dietary assessments.