Diets
mediterranean diet decreases cognitive decline and dementia
In plain terms: Does a Mediterranean diet protect against dementia and age-related cognitive decline?
Leans support Diets 🔬 Includes disconfirming
Part of: 🥗 mediterranean diet
RefutedContestedStrong support
consensus score 0.55
Well-supported at the population level — multiple meta-analyses and large prospective cohorts (including UK Biobank and MIND/DASH work), plus a supportive PREDIMED-NAVARRA trial, consistently link Mediterranean adherence to lower dementia and Alzheimer's risk. The dedicated RCT evidence is still the weaker leg (some diet-cognition trials are null), so the effect looks real but modest.
📅 Last reviewed: 2026-07-15 ⓘ
Evidence ladder
How far up the ladder this claim has climbed. A high consensus on a low rung means "consistent so far," not "proven in people."
Top evidence so far: All trials, pooled (Meta-analysis)
MechanismIn-vitroAnimalObservationalRCTMeta-analysis
How the studies fall
12 support 2 contradict 2 tested null 0 mixed · 16 sources, 14 independent groups
What the evidence shows
The Mediterranean (and MIND) dietary pattern is associated with slower cognitive decline and lower dementia risk.
The evidence (22)
| Source | Grade | Stance | Quality | Finding |
|---|---|---|---|---|
| Barnes LL, Dhana K, Liu X, Carey VJ, Ventrelle J, Johnson K, Hollings CS, Bishop L, Laranjo N, Stubbs BJ, Reilly X, Agarwal P, Zhang S, Grodstein F, Tangney CC, Holland TM, Aggarwal NT, Arfanakis K, Morris MC, Sacks FM 2023 · The New England journal of medicine | RCT | contradicts | high | Primary MIND-diet RCT (Barnes/Morris et al., NEJM 2023, NCT02817074). n=604 (301 MIND, 303 control), 3-year, 2-site. Global cognition change: MIND +0.205 SU vs control +0.170 SU; mean difference 0.035 SU (95% CI -0.022 to 0.092, P=0.23). No |
| Zhao M, Yan M, Fei L, Chen H, Wu M, Li Y, Huang L, Shen J, Liu G, Guasch-Ferré M, Salas-Salvadó J, Yuan C 2026 · Nutrients | observational | tested-null | moderate | Pooled analysis of 3 prospective cohorts (HRS, Framingham Offspring, Whitehall II), n=17,349, >190,000 person-years, 992 incident dementia cases. Higher nut consumption (a MIND/MedDiet component) associated with lower dementia risk: pooled |
| Fekete M, Varga P, Ungvari Z, Fekete JT, Buda A, Szappanos Á, Lehoczki A, Mózes N, Grosso G, Godos J, Menyhart O, Munkácsy G, Tarantini S, Yabluchanskiy A, Ungvari A, Győrffy B 2025 · GeroScience | meta-analysis | supports | moderate | Meta-analysis of 23 studies: MedDiet adherence linked to lower cognitive impairment (HR 0.82), dementia (HR 0.89), and Alzheimer's (HR 0.70). |
| Martínez-Lapiscina EH, Clavero P, Toledo E, Estruch R, Salas-Salvadó J, San Julián B, Sanchez-Tainta A, Ros E, Valls-Pedret C, Martinez-Gonzalez MÁ 2013 · J Neurol Neurosurg Psychiatry | RCT | supports | moderate | PREDIMED-NAVARRA RCT: MedDiet supplemented with olive oil or nuts improved cognitive performance versus low-fat control. |
| Gholami Z, Derakhshan Y, Maracy MR, Paknahad Z 2026 · BMC neurology | RCT | supports | low | RCT, n=56 metabolic syndrome patients. MIND diet vs control: total cognitive function +60% (model 1) / +52% (model 2); executive/visuoconstruction +42%; attention +2.0-2.05x vs control. |
| Kim S, Jang EH, Lee S 2025 · Clinical nutrition research | observational | supports | moderate | Systematic review, 11 studies (7 prospective cohorts, 2 cross-sectional, 1 RCT, 1 case-control), 17,201 participants aged 57-91. Six cohort and two cross-sectional studies found significant positive MIND-cognition associations; one cohort s |
| Chen H, Dhana K, Huang Y, Huang L, Tao Y, Liu X, Melo van Lent D, Zheng Y, Ascherio A, Willett W, Yuan C 2023 · JAMA Psychiatry | observational | supports | high | Large prospective cohort found MIND/Mediterranean-DASH diet adherence associated with lower dementia risk. |
| Radd-Vagenas S, Duffy SL, Naismith SL, Brew BJ, Flood VM, Fiatarone Singh MA 2018 · Am J Clin Nutr | RCT | mixed | moderate | Systematic review of RCTs found limited and inconsistent evidence that MedDiet improves cognition or brain morphology, with few adequately powered trials. |
| Sager R, Gaengler S, Willett WC, Orav EJ, Mattle M, Habermann J, Geiling K, Schimmer RC, Vellas B, Kressig RW, Egli A, Dawson-Hughes B, Bischoff-Ferrari HA 2024 · The journal of nutrition, health & aging | observational | contradicts | high | DO-HEALTH prospective cohort, n=2028 of 2157 participants, mean age 74.9, 3-year follow-up. MIND diet adherence at baseline NOT associated with odds of incident MCI (OR 0.99, 95% CI 0.94-1.04 at MoCA<26; OR 1.03, 95% CI 0.96-1.1 at MoCA<24) |
| McBean L, O'Reilly S 2022 · Eur J Clin Nutr | meta-analysis | tested-null | low | SR/MA of 12 diet-quality RCTs in older adults - Mediterranean, DASH and Nordic diets POOLED (no Mediterranean-only estimate in the abstract; cannot-tell whether the full text has one). Memory SMD 0.98 (95% CI 0.39-1.57) and executive function SMD 2.02 (1.04-3.01) favoured intervention, but I2=99% and 'neither outcome was significant in sensitivity analyses' (I2 then 0%; those estimates not given in the abstract); global cognition, incident dementia and other domains narratively 'inconsistent with no clear direction'. Authors: effect 'unclear given the considerable heterogeneity and bias present across studies'. Estimates lean toward the claim but do not survive sensitivity analysis - underpowered/non-robust trend, tested-null; nothing points against, so not mixed. Abstract-grade. |
| van Soest AP, Beers S, van de Rest O, de Groot LC 2024 · Advances in nutrition (Bethesda, Md.) | observational | mixed | moderate | Systematic review of 40 articles (32 unique cohorts), human studies only, MIND diet and brain aging (no meta-analytic pooling). Higher MIND adherence protective for dementia in 7/10 cohorts, for global cognition in 3/4 cohorts, for episodic |
| Shannon OM, Ranson JM, Gregory S, Macpherson H, Milte C, Lentjes M, Mulligan A, McEvoy C, Griffiths A, Matu J, Hill TR, Adamson A, Siervo M, Minihane AM, Muniz-Tererra G, Ritchie C, Mathers JC, Llewellyn DJ, Stevenson E 2023 · BMC Medicine | observational | supports | high | UK Biobank cohort: higher MedDiet adherence associated with ~23% lower dementia risk independent of polygenic genetic risk. |
| Abasız S, Arslan M 2026 · Nutrients | observational | supports | low | Cross-sectional Turkish adults. Higher MEDAS scores associated with better verbal memory/recall subcomponents (Oktem-SBST) and lower error scores (p<0.05); no HR/OR reported. |
| Zuliani G, Gianfredi V, Veronese N, Volpe M, Maggi S, Onder G, Silano M, Nucci D, Zanetti M, Benussi A, Burgio MI, Fadda M, Guindani P, Sollai S, Cereda E, Caffarra P 2026 · Nutrition | meta-analysis | supports | moderate | Systematic review/meta-analysis (Italian National Guidelines): higher MedDiet adherence associated with reduced risk of neurological disease including dementia; evidence base mostly observational. |
| Tse JHW, Law QPS, Tsang JTY, Suen LKP, Tyrovolas S, Kwan RYC 2025 · J Nutr Health Aging | observational | mixed | moderate | Systematic review of MIND (Mediterranean-DASH hybrid) diet: population studies favorable but RCTs show mixed effects on clinical cognition. |
| Liu Y, Li Y, Li Y, Zhang Y, Wang Y, Gu X, Yue Y, Kang JH, Eliassen AH, Rimm EB, Wang M, Willett WC, Hu FB, Stampfer MJ, Wang DD 2026 · Alzheimer's & dementia : the journal of the Alzheimer's Association | observational | supports | high | NHS+HPFS prospective cohort, n=130,240. Highest vs lowest MedDiet: HR=0.79 (95% CI 0.73-0.84) dementia; MIND HR=0.86 (0.81-0.91); 0.75-1.59 fewer years cognitive aging; 40-42% lower subjective cognitive decline risk. |
| Barbaresko J, Lellmann AW, Schmidt A, Lehmann A, Amini AM, Egert S, Schlesinger S, Nöthlings U 2020 · Advances in Nutrition | meta-analysis | supports | moderate | Umbrella review whose units are meta-analyses: 'In total, 20 meta-analyses including 98 SRRs were identified. All original meta-analyses were rated as being at high risk of bias.' 'Summary risk ratios (SRRs) and 95% CIs were recalculated using a random effects model' — a re-pass of each MA's primaries. MedDiet: 'inversely associated with the incidence of Alzheimer disease (SRR: 0.63; 95% CI: 0.48, 0.82; ... 4 primary studies)' (AD only). Review — zero weight; the pooled cohorts vote via the meta-analyses on this claim. |
| Amanat S, Dordevic AL, Brodtmann A, Cardoso BR 2025 · Adv Nutr | observational | tested-null | low | Systematic review in stroke survivors; meta-analyses were run only for B-vitamin and energy-protein RCTs. The Mediterranean evidence is one cohort re-reported narratively: Cherian 2019, 106 stroke survivors, 5.9 y follow-up - MIND adherence associated with better global cognition (beta 0.083, 0.007 to 0.158) but 'no significant associations between adherence to Mediterranean or DASH diets and cognitive outcomes'; no Mediterranean estimate given. A narrative report of one primary study, not independent evidence. |
| Coelho-Júnior HJ, Trichopoulou A, Panza F 2021 · Ageing Res Rev | meta-analysis | supports | moderate | Meta-analysis found MedDiet adherence associated with better cognitive function and physical performance in older adults. |
| Grasso M, L'Episcopo F, Fidilio A, Olvera-Moreira MA, Toscano G, Muratore S, Drago M, Musso S, Bentivegna V, Costanzo L, Toral-Noristz M, Zambrano-Villacres R, León Brizuela L, Lanza G, Ferri R, Caraci F 2026 · Frontiers in nutrition | observational | supports | low | Cross-sectional, n=92 Italian older adults (37 MCI). Highest vs lowest MedDietScore quartile: OR=0.07 (95% CI 0.01-0.60) for MCI, dose-response; attenuated after adjusting for inflammatory biomarkers (mediation). |
| Liu Y, Gu X, Li Y, Wang F, Vyas CM, Peng C, Dong D, Li Y, Zhang Y, Zhang Y, Zeleznik OA, Kang JH, Wang M, Hu FB, Willett WC, Okereke OI, Eliassen AH, Kraft P, Stampfer MJ, Wang DD 2025 · Nat Med | observational | supports | moderate | Prospective US cohorts (Nurses Health Study/HPFS): MedDiet adherence modulated dementia-related metabolites and was inversely linked to dementia risk, most strongly in APOE4 homozygotes. |
| Agarwal P, Leurgans SE, Agrawal S, Aggarwal NT, Cherian LJ, James BD, Dhana K, Barnes LL, Bennett DA, Schneider JA 2023 · Neurology | observational | supports | moderate | Rush Memory and Aging Project autopsy cohort, n=581 decedents, mean 6.8yr dietary follow-up before death. MIND and Mediterranean diet patterns associated with lower global AD pathology (MIND beta=-0.022 p=0.034; Mediterranean beta=-0.007 p= |
Disagree, or know a study we missed?
We grade by evidence, not opinions. The way to weigh in is to point us to a study we haven't cited (check the evidence table above first), or to flag a problem with one we have. Every submission is reviewed; if it holds up, the grade updates and shows in Science Changes Its Mind.
Opens a short form. You'll sign in with Google so submissions are tied to a real account — we don't display your identity, and we only accept a link we can verify (PubMed, DOI, ClinicalTrials.gov).
Educational only, not medical advice. Grades and scores reflect published evidence weighted by study design and quality; see the methodology.