Each three-point step up the MIND diet scale was matched, in a Framingham Heart Study cohort of 1,647 adults, by grey matter loss that ran 20% slower. The authors translated that rate into 2.5 years of delayed brain ageing. The figure therefore belongs to a unit of diet score and a rate of tissue loss, which limits how far it can be stretched.
The work comes from a team led by Changzheng Yuan at Zhejiang University and appeared in the Journal of Neurology, Neurosurgery & Psychiatry. It drew a fresh round of coverage at the end of September, though the paper itself dates to March 2026, when the BMJ Group first issued its release on the findings.
Framingham Offspring MRI cohort
The participants come from the Offspring generation of the Framingham Heart Study, which enrolled 5,124 adult children of the original 1948 volunteers and their spouses in 1971. The analysed group, 1,647 middle-aged and older adults with an average age of about 60, completed food frequency questionnaires and at least two brain MRI scans. Dietary data were collected between 1991 and 2001, and scans ran from 1999 to 2019, with an average follow-up near 12 years, as Yuan and colleagues describe in the team’s account of the cohort.
MIND, short for Mediterranean-DASH Diet Intervention for Neurodegenerative Delay, scores a person on how often they eat green leafy vegetables, berries, nuts and fish, and how seldom they eat butter, cheese, red meat, pastries and sweets, and fried fast food. The group’s average score was under 7 of a possible 15 points.
Grey matter, ventricles and the 2.5-year conversion
The BMJ Group release states the central result plainly. Each 3-point increase in MIND score was associated with grey matter loss of 0.279 cubic centimetres per year less, which the authors describe as 20% less age-related decline and the equivalent of 2.5 years of delayed brain ageing. That is a dose-response figure. Participants in the top third of adherence were also compared with the bottom third, and the top group fared better on demographic and health measures, but the 2.5 years attaches to the per-3-point slope and not to a gap between those two groups.
The conversion also is not a measured brain-age gap. Nobody in the study was assigned a biological brain age and found to be 2.5 years younger. The authors expressed a slower rate of shrinkage in the units of ordinary ageing, a standard way to make a rate of tissue loss easier to picture.
Ventricles tell a smaller story. According to News-Medical’s account, each 3-point increase linked to 8% less ventricular enlargement, roughly a year of delayed ageing by the same arithmetic. Ventricles, the fluid-filled spaces that widen as surrounding tissue shrinks, are a second structural marker, so two separate measures leaned the same way, though grey matter carried the larger effect.
Berries, poultry, sweets and fried food
When the team broke the score into its parts, two foods stood out as protective and two as harmful. Berries and poultry were tied to slower ventricular enlargement and less grey matter decline. Sweets were associated with faster ventricular enlargement and greater hippocampal atrophy, and fried fast food with larger drops in hippocampal volume.
The food-level results were not tidy. Higher whole grain intake, a recommended MIND item, tracked with faster grey matter and hippocampal losses, while cheese, which MIND tells people to limit, looked protective. Hui Chen, a co-author, stressed in coverage carried by CP24 that the overall dietary pattern matters more than any single item, which is a sensible reading of results that point in different directions item by item.
The authors offered a mechanism for the berry and poultry findings: antioxidant-rich foods such as berries, and high-quality protein such as poultry, may reduce oxidative stress and limit neuronal damage. The study did not test that mechanism, which stays a hypothesis, and the food-level associations were not adjusted into any claim that eating one food changes a scan.
Observational limits on the brain-ageing claim
The design cannot show that eating the diet slows shrinkage. People who score high on MIND tend to be women, college-educated and non-smokers with less diabetes, hypertension and heart disease, and any of those advantages could account for part of the gap. Food frequency questionnaires depend on memory, diets changed over the years without being re-scored, mild cognitive impairment could not be excluded at baseline, genetic risk was not ruled out and the participants were predominantly White, which limits how far the results travel.
Outside commentators in the CP24 report, including David Katz, said the same: association is not causation, and reverse causality is possible, since people with early brain change may eat differently. Walter Willett of Harvard called the findings consistent with Mediterranean-type patterns while acknowledging the observational limits. The 2.5-year figure, in short, is a rate estimate from 1,647 Framingham Offspring participants, and a randomised trial would be needed to say what a change in diet does to a particular person’s scan.
This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.
More from Morning Overview
- Amazon’s Prime refunds are rising to $200 as millions more customers become eligible
- A handful of car engines are so tough mechanics say they almost never wear out
- The NSA is again telling phone owners to switch off one location setting
- Supplements now rank as the fifth-leading cause of death from liver disease.
