Highlights
– In patients with established cardiovascular disease (CVD), higher adherence to the EAT‑Lancet Healthy Reference Diet (HRD) was associated with lower risk of non‑fatal vascular events, driven primarily by reduced stroke incidence.
– Each 10‑point increase in the EAT‑Lancet HRD score (0–140) corresponded to a 13% lower risk of non‑fatal vascular events and a 24% lower risk of stroke in the Utrecht cohort.
– Longitudinal data from ATTICA indicate that education predicts EAT‑Lancet adherence and that diet mediates some of the education‑CVD relationship, with sex‑specific patterns: the protective association was more evident among women at lower educational levels.
Background: clinical context and unmet need
Secondary prevention after a cardiovascular event (myocardial infarction, ischemic stroke, peripheral arterial disease) aims to reduce recurrent events and mortality using evidence‑based therapies (antiplatelet agents, lipid‑lowering therapy, blood pressure control) alongside lifestyle modification. Diet is a key modifiable risk factor: randomized and observational data established cardioprotective effects of dietary patterns such as the Mediterranean diet, and clinical guidelines recommend plant‑rich diets for primary and secondary prevention (Visseren et al., Eur Heart J 2021).
The EAT‑Lancet Commission published a Healthy Reference Diet (HRD) in 2019 designed to optimize human health and environmental sustainability (Willett et al., Lancet 2019). The HRD emphasizes vegetables, fruits, whole grains, legumes, nuts, unsaturated plant oils, and limits red meat, added sugars, and refined grains. Observational studies in generally healthy populations suggest higher EAT‑Lancet adherence associates with lower CVD incidence, but evidence in secondary prevention (patients with established CVD) has been limited. Understanding the relationship in secondary prevention populations is clinically important because diet modification could provide additive benefits to pharmacotherapy and risk factor control, and sustainable dietary recommendations may have public health and environmental co‑benefits.
Study designs and populations
This article synthesizes two recent cohort analyses:
Utrecht Cardiovascular Cohort — SMART‑MART (Second Manifestations of ARTerial disease)
Hoes et al. analyzed patients with established CVD enrolled in the Utrecht Cardiovascular Cohort (SMART‑MART). Dietary intake was quantified using a food frequency questionnaire and scored against the EAT‑Lancet HRD to yield a continuous score from 0 to 140 (higher = closer concordance). The primary endpoint was non‑fatal vascular events (non‑fatal myocardial infarction and non‑fatal stroke). Cox proportional hazards regression models adjusted for age, sex, education, lifestyle factors (smoking, physical activity), and energy intake were used to estimate hazard ratios (HRs) per 10‑point increase in HRD score. Total follow‑up amounted to 24,212 person‑years during which 209 non‑fatal vascular events occurred.
ATTICA Study — 20‑year prospective cohort
Sigala et al. reported on 3,042 community‑dwelling CVD‑free adults enrolled 2001–2002 and followed for 20 years (complete outcome data for 1,988). Baseline adherence to the EAT‑Lancet diet pattern (EAT‑LDP) and educational attainment were the main exposures. The investigators estimated direct and indirect effects of education on 20‑year CVD incidence using generalized structural equation models and nested Cox models, and performed sex‑stratified and interaction analyses to explore effect modification. Behavioral mediators examined included diet and physical activity.
Key findings
Primary results — SMART‑MART (patients with established CVD)
Median EAT‑Lancet HRD score in the cohort was 57 (IQR 41–68) of 140, indicating modest overall adherence. After multivariable adjustment, each 10‑point increase in the HRD score was associated with:
- Non‑fatal vascular events: HR 0.87 (95% CI 0.79–0.96) — a 13% relative risk reduction per 10 points.
- Non‑fatal stroke: HR 0.76 (95% CI 0.63–0.91) — a 24% relative risk reduction per 10 points.
- Non‑fatal myocardial infarction: HR 0.90 (95% CI 0.81–1.02) — a trend toward lower risk that did not reach conventional statistical significance.
These associations were robust to adjustment for major confounders. Absolute risk reductions were not reported in the summary data provided here; however, the consistency across endpoints implies a meaningful clinical signal, particularly for stroke.
Education, sex and long‑term risk — ATTICA
Over 20 years, an inverse educational gradient in CVD incidence and burden was observed — higher education associated with lower long‑term CVD risk. Each additional year of education correlated with greater adherence to the EAT‑Lancet pattern (β = 0.45, 95% CI 0.40–0.50) and higher odds of physical activity. Dietary adherence and activity together mediated a portion of the relationship between education and long‑term CVD incidence. Notably, among women, the cardioprotective association of EAT‑LDP adherence was more pronounced at lower educational levels, suggesting effect modification by sex and socioeconomic position.
Table 4.Results from the GSEM investigating the multifactorial interplay between education and the 20-year CVD incidence among participants of the ATTICA Study (n = 1988).
*** p-value < 0.001, ** p-value < 0.01, * p-value < 0.05. Abbreviations: BMI: body mass index, EAT-LDP: EAT-Lancet diet pattern, OR: odds ratios, 95% CI: 95% confidence interval.
Interpretation and clinical relevance
These studies collectively provide convergent evidence that greater alignment with the EAT‑Lancet HRD or EAT‑LDP is associated with lower risk of CVD outcomes across different populations and stages of disease. The Utrecht cohort extends prior primary prevention findings into a secondary prevention population: in patients with established CVD, a more plant‑forward, less red‑meat and refined carbohydrate‑heavy diet was linked to fewer recurrent non‑fatal events, particularly ischemic stroke.
Biological plausibility is strong. Plant‑forward diets reduce several pathophysiologic drivers of recurrent events: they improve lipid profiles (lower LDL cholesterol with reduced saturated fat and increased soluble fiber), reduce blood pressure, provide anti‑inflammatory and antioxidant nutrients, improve endothelial function, and favorably affect glucose metabolism and body weight. Some mechanisms may be more relevant for stroke (e.g., blood pressure lowering) than for coronary plaque rupture leading to myocardial infarction, which could explain the stronger association with stroke observed in SMART‑MART.
The ATTICA findings emphasize socio‑behavioral determinants: education predicted healthier dietary patterns and activity, which mediated part of the long‑term CVD risk gradient. The stronger diet effect among less‑educated women highlights the need for equity‑oriented, sex‑sensitive interventions. In practice, recommending the EAT‑Lancet pattern in secondary prevention may yield both individual and societal benefits (cardiometabolic and environmental), but implementation must address affordability, cultural acceptability, and health literacy.
Strengths and limitations
Strengths:
- SMART‑MART: investigation in a well‑characterized secondary prevention cohort with adjudicated non‑fatal events and adjustment for major confounders.
- ATTICA: long (20‑year) follow‑up with advanced mediation and interaction analyses, enabling assessment of social determinants and sex differences.
- Both studies examined a dietary index grounded in a widely discussed, sustainability‑minded dietary framework (EAT‑Lancet), linking health and planetary considerations.
Limitations:
- Observational design: residual confounding and reverse causation cannot be excluded. Healthier individuals may adhere to the HRD and also engage in other protective behaviors not fully accounted for.
- Dietary assessment relied on food frequency questionnaires at baseline; changes over time were not captured, and measurement error is inherent.
- The EAT‑Lancet scoring system and cutoffs may differ between studies, limiting direct comparability and clinical translation of a numeric score.
- SMART‑MART reported non‑fatal events; effects on fatal events and all‑cause mortality require further study.
- Generalisability: cohorts were European; applicability to other regions, food systems, and socioeconomic contexts should be tested.
Clinical implications and practice integration
For clinicians caring for patients with established CVD, these data support recommending plant‑forward dietary patterns consistent with the EAT‑Lancet HRD as part of comprehensive secondary prevention. Practical messages can echo guideline recommendations: increase vegetables, fruit, whole grains, legumes, nuts, and oily fish where appropriate; limit processed and red meats, refined grains and added sugars; and favor unsaturated over saturated fats.
Implementing these changes requires patient‑centred counseling, feasible meal plans, and linkage with dietetic services. Clinicians should recognize socioeconomic and sex‑based barriers: lower educational attainment and limited resources may impede adherence. Referral pathways, community programs, and policy interventions (subsidies, food environment improvements) could facilitate equitable uptake.
Research gaps and next steps
Key priorities:
- Randomized trials of EAT‑Lancet–style interventions in secondary prevention to establish causality and quantify absolute risk reductions and safety.
- Longer‑term studies with repeated dietary assessment to understand dose–response, timing, and the effects on fatal events and quality of life.
- Implementation science research to test pragmatic strategies that increase adherence among diverse socioeconomic and cultural groups and to evaluate cost‑effectiveness and environmental co‑benefits.
- Mechanistic studies to delineate pathway‑specific effects (e.g., blood pressure vs lipid mediation) and to understand differential effects by sex and baseline risk.
Expert commentary
Current secondary prevention guidelines already emphasize heart‑healthy eating patterns (e.g., Mediterranean or DASH diets). The emerging data on the EAT‑Lancet HRD add an explicit sustainability lens while maintaining cardiometabolic benefit. Clinicians should weigh the evidence: while randomized data are lacking for EAT‑Lancet specifically in secondary prevention, the observed associations, biological plausibility, and alignment with existing recommendations justify incorporating EAT‑Lancet principles into dietary counselling, with attention to patient preferences and barriers.
Conclusion
Higher adherence to the EAT‑Lancet Healthy Reference Diet is associated with fewer recurrent non‑fatal vascular events in patients with established CVD, with a particularly robust inverse association for stroke. Educational attainment and sex modulate adherence and long‑term benefit, underlining the importance of equity‑oriented, sex‑sensitive dietary interventions. While observational, these findings support recommending plant‑forward, minimally processed dietary patterns as part of comprehensive secondary prevention and motivate randomized and implementation trials to confirm benefits and optimize delivery.
Funding and trial registration
Funding sources and trial registration were not specified in the summary for the SMART‑MART analysis provided here. The ATTICA Study is a long‑standing cohort with multiple funding streams; investigators should be consulted for specific grant numbers. No clinicaltrials.gov identifiers are reported for the observational analyses summarized.
References
1. Hoes LLF, Colizzi C, van der Schouw YT, Geleijnse JM, Dorresteijn JAN, van der Meer MG, Teraa M, Visseren FLJ, Koopal C. The relationship between the EAT‑Lancet dietary pattern and risk of cardiovascular events in patients with established cardiovascular disease. Eur J Nutr. 2025 Dec 1;64(8):324. doi: 10.1007/s00394-025-03754-2 IF: 4.3 Q1 . PMID: 41324709 IF: 4.3 Q1 ; PMCID: PMC12669304 IF: 4.3 Q1 .
2. Sigala EG, Pitsavos C, Barkas F, Liberopoulos E, Sfikakis PP, Tsioufis C, Panagiotakos D. Sex‑Based Associations Between Education Level, EAT‑Lancet Diet, and 20‑Year Cardiovascular Risk: The ATTICA Study (2002‑2022). Nutrients. 2025 Aug 30;17(17):2827. doi: 10.3390/nu17172827 IF: 5.0 Q1 . PMID: 40944215 IF: 5.0 Q1 ; PMCID: PMC12429912 IF: 5.0 Q1 .
3. Willett W, Rockström J, Loken B, et al.; EAT‑Lancet Commission. Food in the Anthropocene: the EAT‑Lancet Commission on healthy diets from sustainable food systems. Lancet. 2019;393(10170):447‑492.
4. Estruch R, Ros E, Salas‑Salvadó J, et al.; PREDIMED Study Investigators. Primary prevention of cardiovascular disease with a Mediterranean diet. N Engl J Med. 2013;368:1279‑1290.
5. Visseren FLJ, Mach F, Smulders YM, et al.; ESC Scientific Document Group. 2021 ESC Guidelines on cardiovascular disease prevention in clinical practice. Eur Heart J. 2021;42(34):3227‑3337.
