Image-based dietary assessment skills of undergraduate nutrition and dietetics students in a multicultural developing country

Background: Image-based dietary assessment is a digital method relatively new in nutrition education. We aimed to evaluate the nutrient estimation skills of nutrition and dietetics students in Malaysia using image-based dietary assessment, and develop a tool to assess their image-based dietary assessment skills. Methods: The study was performed among undergraduate nutrition and dietetics students in two phases. Phase One (n=75) was conducted in four steps: 1) photography of food/beverage at two angles (45° and 90°) with and without a fiducial marker (ruler/dessert spoon); 2) study tool establishment; 3) pre-testing; and 4) conduct of the study. Incorporating changes from Phase One, 55 students participated in Phase Two. Participants were evaluated on their accuracy in recognizing and estimating portions and nutrients of a food/beverage. Data analysis included response frequency and percentages, and statistical differences (Wilcoxon Signed Rank test) in estimating nutrient content. Results: In both phases, the participants scored above 80% for the correct identification of most food items. Participants scored the lowest when estimating nutrient content in local snacks median (IQR): Phase One, 12.5% (25.0); Phase Two, 16.7% (16.7). Participants in Phase One showed better nutrient estimation skills with both fiducial markers, however, no significant difference was observed in Phase Two without: 21.1% (7.9%); with the dessert spoon: 19.0% (13.8%), p=0.197. Participants’ knowledge, skills, and food images were the main barriers to accurate nutrient estimation. Conclusion: Our study found poor nutrient estimation skills among Malaysian undergraduate nutrition and dietetics students. These key findings warrant urgent attention to address this through innovative dietary assessment training on food and portion size from images.