— June 27, 2026 · Leafy Greens and Lower Dementia Risk: Three Cohorts and a Meta-Analysis of 222,108 Adults

Table of Contents

1. At a Glance
2. What Was Done
3. What Was Found, Number by Number
4. Leafy Greens and Their Nutrients
5. How It Fits With Earlier Research
6. What It Cannot Show
7. What the Coverage Said
8. What Remains Unknown
9. Since Then
10. Sources and Primary Documents
11. Key Research Papers
12. Connections

At a Glance

• Paper: Huang L, Hailili G, Zhao M, et al. “Association of vegetable and fruit intake with risk of dementia: 3 prospective studies and meta-analysis of cohort studies.” American Journal of Clinical Nutrition 2026;124(3):101421. Online June 27, 2026; September 2026 print issue.
• Who: 13 researchers. The first and last authors are at the Zhejiang University School of Medicine, Hangzhou, and the last author (Changzheng Yuan) also lists the Harvard T.H. Chan School of Public Health. Co-authors are at Rush University Medical Center and the Karolinska Institute.
• Design: three prospective cohort studies analysed with Cox regression, plus a meta-analysis that added 13 earlier cohort studies. The meta-analysis was registered in advance at PROSPERO (CRD42024525765).
• Cohorts: the Health and Retirement Study (6,750 people), the Framingham Heart Study Offspring Cohort (3,068) and the Whitehall II Study (8,521).
• Size: 18,339 adults aged 45 and over, 936 dementia cases, average follow-up 7 to 13 years; 222,108 participants in the meta-analysis.
• Main results: total vegetables and fruit, top versus bottom third, hazard ratio (HR) 0.74 (0.61–0.89). Green leafy vegetables, per daily serving, HR 0.82 (0.70–0.96). Meta-analysis, highest versus lowest vegetable and fruit intake, HR 0.80 (0.71–0.91).
• Funding: the National Natural Science Foundation of China and the Fundamental Research Funds for the Central Universities, both government funders. The authors report no conflicts of interest.
• Evidence type: observational. There was no randomised trial and no intervention.

Back to Table of Contents

What Was Done

The researchers drew on three long-running health studies in the United States and the United Kingdom: the Health and Retirement Study, the Framingham Heart Study Offspring Cohort and the Whitehall II Study. In each study, participants filled in a food frequency questionnaire, a form that asks how often a person usually eats each of a long list of foods. From those answers the team estimated each person’s average daily intake of total vegetables and fruit and of seven subgroups of produce. Green leafy vegetables were one of the seven. The abstract does not name the other six.

New dementia cases were identified using each cohort’s own definition, since the three studies do not diagnose dementia in the same way. The team used Cox regression, a standard method for comparing how fast an event occurs in different groups over time, and adjusted for other differences between people (“multivariable adjustment”). The abstract does not list which factors were adjusted for.

In the second step, the team searched for earlier cohort studies of vegetables, fruit and dementia. They found 13 and combined those results with their own three in a meta-analysis, a method that pools the estimates of several studies into one. The meta-analysis plan was registered at PROSPERO, an international register of systematic reviews, before the results were known.

Back to Table of Contents

What Was Found, Number by Number

A hazard ratio compares how often dementia was diagnosed in one group with how often it was diagnosed in another over the same time. A value of 1.0 means no difference, and a value below 1.0 means a lower rate. The 95% confidence interval shows the range of values compatible with the data. When the whole range sits below 1.0, the result is counted as statistically significant.

In the three cohorts (18,339 adults, 936 dementia cases)

• Total vegetables and fruit, highest third versus lowest third of intake: HR 0.74 (95% CI 0.61 to 0.89). The test for a steady trend across the three thirds was not statistically significant (P-trend = 0.165). The top group differed from the bottom group, but the data do not show risk falling step by step as intake rose.
• Green leafy vegetables, for each additional daily serving, pooled across the three cohorts: HR 0.82 (95% CI 0.70 to 0.96), P-trend = 0.015.
• Other subgroups: “No significant pooled associations were observed for other vegetable or fruit subgroups,” in the abstract’s words.

In the meta-analysis (222,108 participants, 16 cohort studies)

• Vegetables and fruit together, highest versus lowest intake: pooled HR 0.80 (95% CI 0.71 to 0.91).
• Vegetables, highest versus lowest: pooled HR 0.87 (95% CI 0.82 to 0.92).
• Fruit, highest versus lowest: pooled HR 0.90 (95% CI 0.85 to 0.95).

The abstract does not give a meta-analysis estimate for green leafy vegetables on their own, a measure of heterogeneity (how far the 16 studies disagreed with each other), the number of dementia cases in the meta-analysis, or a dose–response curve across the pooled studies. The only figure tied to an amount eaten is the per-serving hazard ratio for green leafy vegetables in the three cohorts. The authors conclude that their study “supports an inverse association of intake of vegetables and fruits with dementia risk” and that “further studies on specific subgroups are warranted.”

Back to Table of Contents

Leafy Greens and Their Nutrients

Green leafy vegetables are a family of whole foods: spinach, kale, collard greens, Swiss chard, arugula, bok choy and leafy lettuces. Coverage of the study listed foods like these. The abstract does not say which foods the questionnaires grouped as “green leafy vegetables,” and the full text was not available to check this.

The 2026 paper did not test how greens might affect the brain. That question comes from earlier research. In 2018 a team at Rush University studied 960 older adults in the Rush Memory and Aging Project. They looked separately at the main nutrients and plant compounds found in green leafy vegetables:

• Vitamin K1 (phylloquinone), the form of vitamin K found in leafy greens. It is best known for its role in blood clotting.
• Lutein, a yellow carotenoid pigment that also gathers in the eye.
• Folate, the natural form of vitamin B9.
• Nitrate, a compound that leafy greens such as arugula and spinach carry in large amounts.
• Kaempferol (a flavonoid), α-tocopherol (vitamin E) and β-carotene.

In that study, a higher intake of each of these, except β-carotene, was on its own linked to slower decline in thinking and memory. People in the top fifth for green leafy vegetables (a median of 1.3 servings a day) declined more slowly. The authors described the difference as “the equivalent of being 11 years younger in age.”

Two later studies from the same Rush cohort looked inside the brain. In 927 participants followed for an average of 7 years, people in the top fifth for total carotenoid intake had a hazard ratio for Alzheimer’s dementia of 0.52 (0.33 to 0.81) compared with the bottom fifth. The results were similar for lutein-zeaxanthin. At autopsy, a higher lutein-zeaxanthin intake went with less severe neuritic plaques and fewer tangles, the protein deposits that mark Alzheimer’s disease. In 325 brains, the main form of vitamin K was menaquinone-4. Higher concentrations were associated with 17% to 20% lower odds of dementia or mild cognitive impairment. These findings are possible mechanisms, not proof. They come from observational studies, and they show which pathways researchers are investigating.

Back to Table of Contents

How It Fits With Earlier Research

Earlier pooled studies. A 2017 meta-analysis of nine studies (five cohort and four cross-sectional; 31,104 participants, 4,583 cases) found that higher fruit and vegetable intake was associated with lower risk of cognitive impairment and dementia (odds ratio 0.80, 95% CI 0.71 to 0.89). Each additional 100 g a day went with an odds ratio of 0.87 (0.77 to 0.99). The association appeared only in groups whose mean age was over 65. The 2026 paper rests on far more people and only on cohort studies, and its pooled figure for total produce (0.80) is close to the 2017 figure.

Same research group, earlier work. In 2019 the 2026 paper’s last author led a study of 27,842 US men. Those who ate the most vegetables had lower odds of poor self-reported thinking and memory when this was assessed in 2008 and 2012 (odds ratio 0.66, 0.55 to 0.80). Intake 18 to 22 years before the memory assessment still mattered when more recent intake was taken into account. In 2024 the same first and last authors studied cohorts in China and the United States. In the US Health and Retirement Study, higher green leafy vegetable intake went with better cognitive scores and slower decline. In the Chinese cohort the confidence interval for green leafy vegetables included zero (β 0.036, −0.001 to 0.073).

The MIND diet. Leafy greens are part of the MIND diet, a hybrid of the Mediterranean and DASH diets built around foods linked to brain health. In a 2015 observational study of 923 older adults, people in the top third of MIND diet scores had a hazard ratio for Alzheimer’s disease of 0.47 (0.26 to 0.76) compared with the bottom third.

The trial that found no difference. The MIND diet was then tested in a randomised trial, published in the New England Journal of Medicine in 2023. The trial assigned 604 older adults without cognitive impairment, with a family history of dementia, a body-mass index above 25 and a suboptimal diet, to three years of either the MIND diet or a control diet, both with mild calorie restriction. The MIND diet “emphasizes consumption of plant-based foods, including green leafy vegetables, nuts and berries, fish, and olive oil.” Cognitive scores rose in both groups (by 0.205 and 0.170 standardised units). The difference of 0.035 (95% CI −0.022 to 0.092; P = 0.23) was not significant, and brain MRI measures were similar in both groups. The authors noted that both groups lost a similar amount of weight (5.0 kg and 4.8 kg), that the control group probably also improved its diet, and that practice from repeated testing could explain the improvement in both groups. Two of the 2026 paper’s co-authors were authors of that trial.

The same week. Two days earlier, on June 25, 2026, JAMA Network Open published a Swedish cohort study of 1,865 adults aged 60 and over. Among people with high blood levels of p-tau217, a marker of Alzheimer’s disease pathology, closer adherence to a less inflammatory dietary pattern went with lower dementia risk (HR 0.71 per standard-deviation increase, 0.58 to 0.88). It is also observational.

Back to Table of Contents

What It Cannot Show

• Cause and effect. The study observed people and did not assign anyone a diet, so it shows an association. People who eat more greens also tend to differ in other ways, such as education, income, exercise, smoking and overall diet. Statistical adjustment reduces this problem but cannot remove it.
• Reverse causation. Dementia develops slowly, and eating habits can change years before a diagnosis. This is a recognised limit of diet-and-dementia cohorts. Follow-up here averaged 7 to 13 years.
• Food questionnaires. Intake was self-reported on food frequency questionnaires. These give rough estimates, not exact measurements, and measurement error of this kind is a recognised limit of such studies.
• Different definitions. Each cohort used its own definition of dementia, and the 16 studies in the meta-analysis did not all define “highest” and “lowest” intake the same way.
• One subgroup among seven. Green leafy vegetables were one of seven subgroups tested. When many comparisons are made, one may reach significance partly by chance. The authors themselves call for “further studies on specific subgroups.”
• No trend for total produce. The top-versus-bottom result for total vegetables and fruit was significant, but the trend test across thirds was not (P-trend = 0.165).
• Unread details. Heterogeneity, sensitivity analyses, the authors’ own list of limitations and the per-cohort results are in the full text, which this report could not access.

Back to Table of Contents

What the Coverage Said

No press release or quoted statement from the authors, a university or a health agency was found for this paper. Its main coverage came days and weeks later.

A wellness news article on July 5, 2026 turned the hazard ratios into percentages: “about a 20% lower risk” for the most vegetables and fruit, “13%” for vegetables, “10%” for fruit and an “18% lower risk of dementia” for each additional daily serving of greens. These are relative differences that match the abstract’s hazard ratios. They do not describe any one person’s chance of developing dementia. The article’s list of greens (spinach, kale, Swiss chard, collard greens, arugula, romaine lettuce, bok choy) is the outlet’s own list, not a definition taken from the paper.

A later article on another website called leafy greens the “single most potent food for dementia prevention.” The paper does not support that wording. It compared seven subgroups of vegetables and fruit, not all foods, and as an observational study it reports a lower risk, not prevention.

Back to Table of Contents

What Remains Unknown

• Whether eating more leafy greens, started in midlife or later, changes the risk of dementia. Only a long randomised trial with dementia diagnoses as its outcome could show this, and none of the studies cited here is such a trial.
• Which part of the food matters, if any: vitamin K1, lutein, folate, nitrate, another compound, or leafy greens as a marker of an overall eating pattern.
• How much is enough. The per-serving hazard ratio comes from three cohorts. No dose–response curve across all 16 studies is reported in the abstract.
• Whether the association differs by type of dementia (Alzheimer’s disease or vascular dementia), by sex, or by genetic risk. The abstract does not report this.
• Whether eating greens raw or cooked, or the foods eaten alongside them, changes the result. The abstract does not address this.

Back to Table of Contents

Since Then

On July 13, 2026, a little over two weeks after this paper, The Lancet published the main results of LatAm-FINGERS, a randomised trial in 1,065 older adults at raised risk of dementia across 11 Latin American countries. A structured two-year programme combining exercise, diet counselling, brain training and monitoring of heart risk factors raised scores on a combined test of thinking and memory faster than flexible health advice did: 0.31 versus 0.20 standard-deviation units a year, a difference of 0.11 (95% CI 0.06 to 0.15). That trial tested a whole lifestyle package, not leafy greens alone, and it measured test scores, not new dementia cases. It is covered in full in the site’s report Lifestyle Program Slows Cognitive Decline in Latin America.

Back to Table of Contents

Sources and Primary Documents

Every figure about the 2026 study was taken from the PubMed abstract, the conflict-of-interest declaration and the Crossref publisher record of the paper, not from news coverage. The full text is subscription-only and was not read.

1. Huang L, Hailili G, Zhao M, Chen H, Shen J, Yan M, Shen T, Wu M, Fei L, Liu X, Xu W, Agarwal P, Yuan C (2026). Association of vegetable and fruit intake with risk of dementia: 3 prospective studies and meta-analysis of cohort studies. American Journal of Clinical Nutrition 124(3):101421. Epub June 27, 2026. — doi:10.1016/j.ajcnut.2026.101421 · PubMed PMID: 42364910
2. PROSPERO international register of systematic reviews, registration CRD42024525765 (the meta-analysis protocol, as cited in the paper’s abstract).
3. Mrhar A, Carballo-Casla A, Grande G, et al. (2026). Diet Quality and Dementia Risk in Older Adults With Alzheimer Pathology. JAMA Network Open 9(6):e2620254. Online June 25, 2026. — doi:10.1001/jamanetworkopen.2026.20254 · PubMed PMID: 42348207
4. Crivelli L, Suemoto CK, Sosa AL, et al.; LatAm-FINGERS Study Group (2026). Multidomain lifestyle intervention for the prevention of cognitive decline in at-risk older adults in Latin America (LatAm-FINGERS): a single-blind, multicentre, randomised controlled trial. The Lancet 408(10553):417-429. Online July 13, 2026. — doi:10.1016/S0140-6736(26)01278-X · PubMed PMID: 42442374

News coverage: a wellness news article, July 5, 2026 (coverage of the study).

Back to Table of Contents

Key Research Papers

Background papers for this report. Each entry was checked against PubMed and Crossref before it was written down.

1. Morris MC, Wang Y, Barnes LL, Bennett DA, Dawson-Hughes B, Booth SL (2018). Nutrients and bioactives in green leafy vegetables and cognitive decline: Prospective study. Neurology 90(3):e214-e222. — PubMed PMID: 29263222
2. Morris MC, Tangney CC, Wang Y, Sacks FM, Bennett DA, Aggarwal NT (2015). MIND diet associated with reduced incidence of Alzheimer’s disease. Alzheimer’s & Dementia 11(9):1007-1014. — PubMed PMID: 25681666
3. Barnes LL, Dhana K, Liu X, Carey VJ, Ventrelle J, Johnson K, et al. (2023). Trial of the MIND Diet for Prevention of Cognitive Decline in Older Persons. New England Journal of Medicine 389(7):602-611. — PubMed PMID: 37466280
4. Yuan C, Fondell E, Bhushan A, Ascherio A, Okereke OI, Grodstein F, Willett WC (2019). Long-term intake of vegetables and fruits and subjective cognitive function in US men. Neurology 92(1):e63-e75. — PubMed PMID: 30464030
5. Jiang X, Huang J, Song D, Deng R, Wei J, Zhang Z (2017). Increased Consumption of Fruit and Vegetables Is Related to a Reduced Risk of Cognitive Impairment and Dementia: Meta-Analysis. Frontiers in Aging Neuroscience 9:18. — PubMed PMID: 28223933
6. Huang L, Zhao C, Gao M, Tao Y, Chen X, Chen H, et al. (2024). Associations of vegetable and fruit intake with cognitive function and its decline: Two longitudinal studies. Journal of Nutrition, Health and Aging 28(6):100223. — PubMed PMID: 38598978
7. Yuan C, Chen H, Wang Y, Schneider JA, Willett WC, Morris MC (2021). Dietary carotenoids related to risk of incident Alzheimer dementia (AD) and brain AD neuropathology: a community-based cohort of older adults. American Journal of Clinical Nutrition 113(1):200-208. — PubMed PMID: 33184623
8. Booth SL, Shea MK, Barger K, Leurgans SE, James BD, Holland TM, et al. (2022). Association of vitamin K with cognitive decline and neuropathology in community-dwelling older persons. Alzheimer’s & Dementia: Translational Research & Clinical Interventions 8(1):e12255. — PubMed PMID: 35475263

PubMed Topic Searches

1. PubMed: Green leafy vegetables and dementia
2. PubMed: Vegetable and fruit intake and dementia, cohort meta-analyses
3. PubMed: Vitamin K1 (phylloquinone) and cognitive decline
4. PubMed: Lutein and Alzheimer’s dementia
5. PubMed: MIND diet randomised trials and cognition

Back to Table of Contents

Connections

• News Archive
• Dementia — types, risk factors and research
• Alzheimer’s Disease — the most common cause of dementia
• Vascular Dementia — dementia from reduced blood flow to the brain
• Spinach — the leafy green, its nutrients and research
• Kale — the cruciferous leafy green
• Collard Greens — a vitamin K1-rich leafy green
• Swiss Chard — the leafy beet relative
• Arugula — a peppery, nitrate-rich salad green
• Vegetables — every vegetable page on the site
• Vitamin K — K1 (phylloquinone) from greens and its roles
• Vitamin B9 (Folate) — the B vitamin and its research
• Lutein and Brain Health — the carotenoid and cognition research
• Mediterranean Diet — one parent of the MIND diet
• Brain Aging — mechanisms and research on the ageing brain
• LatAm-FINGERS (July 13, 2026) — the lifestyle trial in older adults in Latin America

Back to Table of Contents