Abstract
BackgroundTaste of food may be important in food choice and dietary intakes. While taste perception in the laboratory has been widely studied, it has rarely been taken into account when exploring dietary patterns in a population. The aim of this work is to identify consumers’ taste perceptions for a list of commonly consumed foods.MethodsThe National Diet and Nutrition Survey (NDNS) rolling programme year 9 (2016/17) was used to generate a list of foods consumed by adolescents (10–19 years old). 1748 different foods were grouped by food group, and sugar/salt contents into seven main and 23 subgroups. A pilot test was conducted to identify foods with high taste agreement to further limit the list. The shortened food list was included in an online food-taste classification survey which was distributed to Facebook groups and Twitter. Participants were asked to allocate one main taste of the five basic tastes (sweet, salty, sour, bitter or savoury/umami) to each food. Additionally, neutral and never tried options were provided. To minimize participant burden, the food list was divided into three and participants were asked to complete one part with an option to voluntarily complete the rest. Hierarchical cluster analysis was used to allocate foods into taste clusters. Number of clusters was determined based on the elbow method. Cluster linkage method used was based on a dendrogram and cophenetic correlation coefficient (CPCC).ResultsThe pilot questionnaire was completed by 19 participants. 15/19 rated 55 foods as the same taste and these were removed from the final questionnaire. The final food list contained 184 items. The main online questionnaire obtained 209 responses from 162 females, 44 males, (3 not known). The age range was from 18–70+ years with the majority between 40–59 years old. Seven reported having had COVID-19, but only three were tested positive (none experienced a taste loss). Cluster analysis classified foods into 6 taste clusters using the Average weighted linkage (CPCC=71%). Foods were classified as 21% sweet, 9% salty, 9% sour, 13% bitter, 18% savoury, and 29% neutral. These tastes have been applied to all the foods consumed by adolescents reported in the NDNS to generate taste patterns allowing us to explore links with diet and nutritional quality and anthropometry.ConclusionWe have characterised basic tastes for foods reported in the NDNS. This will allow us to create a UK food-taste database to explore dietary-taste patterns in the NDNS.