News · Nutrition & Diet
A plant-based pattern cut 20-year heart disease risk by 26%
Greek researchers followed 1,988 adults for two decades and found the diet that was best for the planet was also best for the heart. The Western pattern showed nothing.
Based on a peer-reviewed 20-year prospective cohort study in the European Journal of Nutrition
- Researchers enrolled 3,042 adults free of cardiovascular disease in Athens in 2001-2002 and followed them for 20 years, with complete outcome data for 1,988, publishing in the European Journal of Nutrition.
- Three eating patterns emerged from the data itself rather than being defined in advance: plant-based, Western, and high-calorie with little white meat.
- Higher adherence to the plant-based pattern went with 26% lower 20-year cardiovascular risk (HR, 0.74; 95% CI, 0.55-0.99).
- That upper bound of 0.99 sits right at the edge of statistical significance, so the finding is real but not comfortable.
- The high-calorie, low-white-meat pattern was linked to higher lifetime risk and more years of healthy life lost.
- The Western pattern showed no significant association after adjustment, which is the paper's most surprising result.
- Observational, based on a food questionnaire at a single baseline. Diets change over 20 years and this study did not track that.
- The cohort is Greek, so the plant-based pattern here is a Mediterranean one and may not describe plant-based eating elsewhere.
Twenty years is a long time to follow anyone, and it is roughly the timescale on which diet and heart disease actually interact. A study in the European Journal of Nutrition managed it, in Athens, and found the eating pattern with the lightest environmental footprint also had the lowest cardiovascular risk.
The team set out to identify habitual dietary patterns within a Mediterranean population-based cohort, examine their associations with cardiovascular disease (CVD) risk factors and outcomes over 20 years, and interpret them in sustainability terms.
Letting the data pick the diets
Most diet research scores people against a template: how Mediterranean is your diet, how closely does it match DASH. This study inverted that.
A total of 3,042 CVD-free adults from the ATTICA Study were enrolled in 2001-2002, and their food questionnaires were fed through a technique that finds natural groupings without being told what to look for.
Three dietary patterns were identified, explaining 46.5% of variance in consumption: a plant-based, sustainable pattern; a Western pattern rich in animal-sourced and processed foods; and a high-calorie, low-white-meat pattern.
Then everyone was followed. Participants were followed for 20 years, with complete CVD data available for 1,988 individuals - a third of the original group lost along the way, which is ordinary for two decades and still worth knowing.
The plant-based result
The headline finding is the one the title promises.
Higher adherence to the plant-based pattern was associated with a 26% lower 20-year CVD risk, with the estimate running from 0.55 to 0.99 on the study’s risk scale.
That upper number deserves a moment. On this scale, 1.00 means no effect. Landing at 0.99 means the result cleared the conventional bar by the smallest margin available. The direction is consistent with a great deal of other evidence, and the strength of this particular estimate is weaker than 26% makes it sound.
The pattern that did damage
The high-calorie, low-white-meat pattern was linked to increased lifetime risk and to more disability-adjusted life years, a measure that counts years lost to early death alongside years lived in poor health.
Both associations were statistically robust, more so than the plant-based finding. The unhealthy pattern’s harm was easier to detect than the healthy pattern’s benefit, which is a recurring feature of nutrition research and not unique to this study.
The Western pattern did nothing
This is the result that does not fit.
No significant associations were observed for the Western pattern after adjustments.
Animal-sourced and processed foods, and no detectable effect on 20-year cardiovascular outcomes. The authors report it without explaining it, which is the right call, but it is odd enough to sit with.
A few possibilities. Athens in 2002 may not have had a Western pattern extreme enough to separate from the rest. The pattern lumps together foods with genuinely different effects. Or the adjustment for blood pressure, cholesterol and body weight removed the very pathways through which such a diet does its damage, which is a known trap in nutrition analysis.
None of those are established here. The honest summary is that a result many people would have predicted did not appear.
What the design cannot do
Diet was assessed once, at baseline, with a food frequency questionnaire. People were then followed for 20 years without their eating being re-measured. Two decades is ample time to change how you eat, and every such change adds noise.
The patterns are also local. A plant-based pattern derived from Greek eating habits in 2002 means olive oil, legumes, vegetables and bread. That is not what plant-based describes in most other countries, and the finding should not be assumed to travel.
And it remains observational. People who ate this way differed in other respects, and no adjustment fully removes that.
What the authors conclude
Their summary is measured: a plant-based dietary pattern was protective against long-term CVD outcomes, while a high-calorie, low-white-meat pattern was detrimental.
The framing they add is the reason the paper exists. These findings underscore the importance of promoting culturally acceptable, sustainable dietary patterns to reduce CVD risk and support environmental sustainability.
Culturally acceptable is doing real work in that sentence. The pattern that performed best here was not an imported prescription. It was what a large share of this population was already eating.
People also ask
How were the diet patterns chosen?
They were not chosen. The researchers used a statistical technique that finds the natural groupings in what people reported eating, rather than scoring everyone against a predefined ideal like the Mediterranean diet score. Three patterns emerged and together accounted for 46.5% of the variation in consumption. The advantage is that the patterns reflect how this population actually ate; the drawback is that they are specific to this population and do not transfer cleanly.
Is a 26% reduction large?
It is a meaningful size for a dietary finding over two decades, but the uncertainty deserves attention. The range ran from 0.55 to 0.99, and 1.00 would mean no effect at all. The upper edge landing at 0.99 means the result cleared the conventional threshold by the narrowest possible margin. It is evidence, not proof, and a slightly different analysis could have put it on the other side.
Why did the Western pattern show nothing?
The authors report no significant association after adjustment, and they do not claim to know why. Several explanations are possible: the comparison group in a Mediterranean country may still eat relatively well, the pattern may be diluted by including many different foods, or the adjustment for other risk factors may have absorbed the effect. It is a genuinely odd result and worth flagging rather than smoothing over.
What are disability-adjusted life years?
A measure that combines years lost to early death with years lived in poor health, so a condition that disables without killing still counts. Public health researchers use it because heart disease often does both. In this study the high-calorie pattern was linked to more of them, meaning not just more disease but more time spent unwell.
Does this mean plant-based eating is proven to protect the heart?
This is general information rather than medical advice. One 20-year cohort adds to a large body of evidence pointing the same way, but observational studies cannot establish cause on their own. People who eat more plants differ from those who do not in income, exercise, smoking and much else, and adjustment only partly handles that. Anyone making dietary changes for a heart condition should discuss them with a clinician.
References
- Sigala EG, Damigou E, Dalmyras D, et al. Sustainable diets and long-term cardiovascular disease outcomes; insights from the 20-year follow-up ATTICA study (2002-2022). European Journal of Nutrition (2026).
- National Heart, Lung, and Blood Institute. DASH Eating Plan.
- Centers for Disease Control and Prevention. Heart Disease Risk Factors.