Four distinct food-liking subtypes were identified among the studied participants: starch-free or low-starch pattern (18.09%), vegetarian pattern (5.54%), high protein and low fiber pattern (19.39%), and (4) balanced pattern (56.98%). Quantitative scores confirmed the robustness of the relationship between food liking and actual food consumption patterns among the individuals.
The present study obtained food-liking data from the United Kingdom (UK) Biobank. A total of 181,990 participants who completed a food-liking questionnaire were included.
The mean age of the participants was 70.7 years, and about 57% were female. The data were then analyzed using principal component analysis (PCA) and hierarchical clustering to identify food-liking subtypes. Further, differences in various brain health indicators, including mental health, cognitive function, biomarkers, and brain magnetic resonance imaging (MRI) traits, were assessed among these subtypes using one-way analysis of covariance (ANCOVA). The study included measures of anxiety, depressive symptoms, mental distress, psychotic experiences, self-harm, trauma, and well-being as indicators of brain health.
Longitudinal data on mental disorders were also analyzed using Cox proportional hazards models to examine the differences among the subtypes. Structural equation models (SEMs) were used to examine the relationships between dietary patterns and brain health. Finally, genome-wide association analysis (GWAS) and gene expression analysis were conducted to study the genetic basis of food-liking subtypes and potential biological pathways.
The balanced pattern, subtype 4, showed the lowest measures for mental health issues and the highest scores for overall well-being and cognitive functions, indicating improved brain health and cognition than the other subtypes. On the other hand, subtypes 2 and 3 showed lower scores in well-being and higher scores in mental health issues. Compared to subtype 4, subtype 3 exhibited reduced gray matter volumes in regions like the postcentral gyrus, indicating potential neurological differences.
In contrast, subtype 2 displayed increased volumes in the thalamus and precuneus.
