Background High consumption of UPFs has been proposed as a key driver in the global rise of non-communicable diseases. UK adolescents consume 66% of their daily energy (kcal/d) from UPFs. However, what foods are consumed remains poorly characterised. The current paradigm in nutritional science and policy focuses on nutrients but the effects of the level of processing are not considered. This study examined patterns of UPF consumption in UK adolescents and assessed associations with sociodemographic characteristics.

Methods We analysed 4-day food diary data from 3,270 adolescents (11-18y) in NDNS (2008/09-2018/19). Foods were classified based on level of processing using the NOVA’s food classification system. UPFs were further sub-categorised based on literature and NDNS food categories. Average relative energy intake (%kcal/day) and relative weight (%g/day)from UPFs and each UPF sub-type was calculated. Principal component analysis (PCA) and clustering analysis were used to identify adolescent UPF consumption patterns. Hierarchical clustering analysis using the principal components assigned individuals to a group with similar patterns of UPF sub-type intake. These clusters were interpreted as UPF consumption patterns and labelled according to their main UPF sub-type intakes. Finally, multivariable logistic regression explored associations between sociodemographic characteristics and UPF consumption patterns.

Results UPFs accounted for 66%kcal/day and 45%g/day of adolescents’ diet. Males and adolescents from white ethnic and from and lower socioeconomic status (SES) backgrounds had a higher UPF intake. PCA identified three UPF dietary patterns: ‘Restrictive’,’Traditional’,and ‘Permissive’. The Restrictive pattern, characterized by higher-than-average consumption of UPFs that are often perceived as healthy, such as meat replacements and yogurts, was associated with individuals of a higher SES, females and Asian adolescents. The Permissive pattern was characterized by higher intakes of soft drinks and was associated with lower SES, males, and older adolescents. The Traditional pattern was characterized by higher intakes of industrial breads and was more prevalent in younger and male adolescents.

Conclusion UPF consumption is high in UK adolescents and socioeconomically patterned with higher overall consumption among adolescents from lower SES backgrounds. The types of UPFs consumed also vary by adolescents’ sociodemographic characteristics. Those from higher SES backgrounds consume UPFs perceived as healthy whilst those from lower SES backgrounds consume UPFs commonly high in fat, sugar and salt. The high UPF consumption and the different patterns identified in this study, alongside their links to sociodemographic characteristics, underline the need for further research to inform targeted policies and interventions that focus on the different UPF consumption groups.