Answers are grounded in this paper and cite their sources.
Cross-sectional analysis finds that high ultra-processed food intake worsens cognitive performance in older adults, suggesting dietary shifts may improve brain health.
Background: Given the increasing consumption of ultra-processed foods (UPFs) and the public health importance of cognitive decline in ageing, understanding how UPFs impact cognitive performance is highly relevant. However, evidence in older adults - particularly in Italy—remains scarce, despite the country’s rapidly ageing population, its comparatively low UPF intake, and its distinct Mediterranean dietary context. Methods: We analysed cross-sectional data from 809 community-dwelling adults aged ≥ 65 years (59.4% women) participating in the NutBrain population-based cohort. Dietary intake was assessed using a 102-item semi-quantitative food frequency questionnaire, and daily grams of foods were classified according to the NOVA system into groups, which were analysed using a compositional data analysis approach. Global cognition and domain-specific performance were measured using standardised neuropsychological tests. Associations between NOVA groups and cognitive outcomes were estimated using multiple linear regression models adjusted for potential confounders. Gender-stratified analyses were also performed. Results: UPFs accounted for 21% of total energy intake, with bakery products as the main contributors. A relative increase of unprocessed or minimally processed foods was associated with better cognitive function (B = 0.36, p = 0.014), whereas a greater contribution of UPFs relative to the overall diet was associated with worse global cognitive function (B = −0.26, p = 0.003). The strongest associations were observed for episodic memory, particularly among women. Conclusions: A higher relative consumption of UPF was associated with worse global and memory-related cognitive performance. Longitudinal and experimental studies are warranted to clarify causality and underlying mechanisms.
No takes yet. Share an insight, caveat, or question.
Prinelli et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: