MétaCan
Menu
← Back to cohort
Record W4415123548 · doi:10.1101/2025.10.09.25337399

Dietary Interventions and Cognitive Function across the Dementia Continuum: A Systematic Review, Meta-Analysis, Meta-Regression and Call to Action for Research Reform

2025· preprint· en· W4415123548 on OpenAlexaff
Megan Kirk, Oliver J. Canfell, Meysam Pirbaglou, Forhad Uddin Hasan Chowdhury, Megan Smith, Rachel O. Reid, Amy D. Hitchcock, Brandon Chang, Heather Knight, Eva Lash, Jason R. Grant, Erin L. Bellamy, Joel Katz, Ivan Koychev, Kamaldeep Bhui

Bibliographic record

VenuemedRxiv · 2025
Typepreprint
Languageen
FieldMedicine
TopicNutritional Studies and Diet
Canadian institutionsYork University
FundersNational Institute for Health and Care Research
KeywordsDementiaPsychological interventionNeurocognitiveCognitionRandomized controlled trialMediterranean dietMeta-analysisAlzheimer's diseaseCognitive decline

Abstract

fetched live from OpenAlex

Abstract Background Dementia cases are projected to rise to 153 million worldwide by 2050. Diet is a modifiable risk factor, and interventions may delay cognitive decline. No review has synthesised the full randomised controlled trial (RCT) evidence across all dietary interventions and dementia stages. Objective To critically appraise and synthesise evidence of dietary interventions on cognitive function across the dementia disease continuum (healthy, at-risk/preclinical, mild cognitive impairment, clinical). Methods Following PRISMA guidelines, seven databases and three trial registries were searched for RCTs from inception to 28 January 2025. Eligible RCTs evaluated multidomain (e.g., diet plus exercise), whole dietary pattern (e.g., Mediterranean diet), or single food (e.g., blueberries) interventions ≥ 2-weeks in adults, 18+ years. Data extraction was triple coded; risk of bias was assessed using the Cochrane Risk of Bias (RoB2) tool. Narrative synthesis was complemented by pooled random-effects meta-analysis. Prespecified meta-regression explored reasons for heterogeneity. Sensitivity analyses were performed. Results Eighty-three RCTs (110 diet intervention comparisons; n = 24 063 participants) met inclusion criteria. Overall, 190 cognitive measures were extracted and mapped onto the six DSM-V neurocognitive domains. A total of 885 cognitive, neuroimaging, and blood biomarker outcomes were assessed. Pooled meta-analysis ( k = 35) showed a small, significant improvement in global cognition (Hedges’ g = 0.25, 95%CI: 0.15 to 0.36). Multidomain ( g = 0.25, 95%CI: 0.10 to 0.41) and total diet ( g = 0.27, 95%CI: 0.10 to 0.44) interventions showed benefit; single food interventions were non-significant. Trial durations ≤ 12 weeks (β = 0.59) and presence of cardiometabolic comorbidities at baseline (β = 0.45) predicted greater cognitive improvements in multidomain interventions; for total diet interventions, lower risk of bias (β = 0.51) moderated effects. Neuroimaging revealed benefits in MCI across all intervention types. Certainty of evidence was rated very low. Discussion Dietary interventions modestly improve cognitive function across dementia stages, with multidomain and total diet interventions showing the most benefit. RCTs during the preclinical phase (e.g., MCI) of dementia in adults with co-occurring cardiometabolic risk are recommended. Given substantial heterogeneity, risk of bias, and measurement inconsistency, international consensus is urgently needed to standardise intervention protocols (e.g., duration, type, dosage) and cognitive outcome measures to strengthen evidence for dementia prevention and care. Funding NIHR Applied Research Collaboration Oxford and Thames Valley, NIHR Oxford Health Biomedical Research Centre (BRC) Registration NIHR PROSPERO database (registration: CRD42023488336)

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.087
metaresearch head score (Gemma)0.135
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesMetaresearch
Consensus categoriesnone
DomainCandidate signal: Evaluation · Consensus signal: none
Study designCandidate signal: Meta-analysis · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.913
Threshold uncertainty score0.460

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0870.135
Meta-epidemiology (narrow)0.0040.002
Meta-epidemiology (broad)0.0260.045
Bibliometrics0.0100.008
Science and technology studies0.0010.001
Scholarly communication0.0060.005
Open science0.0040.003
Research integrity0.0030.003
Insufficient payload (model declined to judge)0.0050.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.380
GPT teacher head0.505
Teacher spread0.125 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

Study designMeta-analysis
DomainEvaluation
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same venuemedRxiv→Same topicNutritional Studies and Diet→French-language works237,207→