Background and aims: Poor cardiometabolic health is associated with dementia. Considering previous meta-analyses have confirmed associations between ultra-processed foods (UPFs) and cardiometabolic disease, we were interested in the contribution of UPF consumption to the risk of developing dementia.

Methods: We performed a systematic review and meta-analysis of all records registered on Ovid Medline and Web of Science from inception until December 2022 [PROSPERO (CRD42023388363)]. Studies that assessed UPF consumption in adults, determined according to NOVA, and that reported dementia (Alzheimer's disease, vascular dementia and mild cognitive impairment) determined by clearly stated diagnostic criteria (including formal assessment of dementia or use of diagnostic codes) were included. The association between UPF consumption and dementia was assessed using random-effects meta-analysis, controlling for confounding variables. Study quality was assessed using the Newcastle Ottawa Scale and evidence credibility evaluated using the NutriGrade system.

Results: Seven thousand ten records were screened, and 122 records underwent full text review. From these, 10 observational (8 longitudinal) studies, analysing 867,316 individuals, were included. Included studies adjusted for age, socioeconomic status and co-morbidity, alongside other confounders. High (vs. low) intake of UPF was associated with increased risk of dementia (pooled relative risk 1.44 (95% confidence interval 1.09-1.90) (p = 0.02)) (I² = 97.0%), although moderate (vs. low) intake of UPF was not (1.12 (0.96-1.31) (0.13)) (85.0%). Funnel plots demonstrate low risk of publication bias.

Conclusion: High UPF consumption is associated with dementia. Public health measures to reduce overconsumption of UPFs are imperative to reduce the burden of dementia.