Eating Your Fruits and Vegetables Won’t Get You Enough Flavanols for Heart Health

Following official dietary guidelines for fruits and vegetables sounds like a solid plan for cardiovascular health. But a new study published in Food & Function suggests those guidelines leave a critical gap: fewer than one in five adults, even those eating plenty of produce, reach the flavanol intake levels that research links to significantly reduced cardiovascular mortality.

Flavanol intake for cardiovascular benefit: what the threshold actually is

Flavanols are a class of plant-derived polyphenols found in apples, berries, grapes, tea, and cocoa products. They are not vitamins or minerals, and no official Dietary Reference Intake value exists for them yet. What does exist is a growing body of evidence pointing to a specific target. The COSMOS trial, the largest randomized controlled study on polyphenols ever conducted, found that 500 mg per day of flavanols reduced cardiovascular disease mortality by 27% and cut total cardiovascular events by 15% in healthy older adults. An expert panel subsequently proposed 400 to 600 mg per day as a target range for cardiometabolic health, covering reductions in blood pressure, cholesterol, and blood sugar. The new study asked a simple but pointed question: do current dietary guidelines actually get people there?

The answer, across two large cohorts totaling more than 30,600 participants, is a clear no.

Only 1 in 5 adults reached 500 mg per day, even following dietary guidelines

Researchers analyzed flavanol intake using two validated urinary biomarkers, gVLMB and SREMB, in the US-based COSMOS cohort (n = 6,509) and the UK-based EPIC-Norfolk cohort (n = 24,154). These biomarkers capture flavanols from fruits, vegetables, and the sources tested in COSMOS, providing a more objective window into actual intake than self-reported food diaries alone. Estimating flavanol consumption from questionnaires alone is notoriously imprecise, given the challenges in measuring flavonoid intake accurately across diverse diets and food composition databases.

Overall, only 19.2% of COSMOS participants and 17.9% of EPIC-Norfolk participants met the biomarker-estimated threshold of at least 500 mg per day. The results were striking even among the most diet-conscious individuals. In COSMOS, participants in the highest quartile of diet quality (measured by the alternative Healthy Eating Index) had just 26% higher odds of meeting the 500 mg threshold compared to those in the lowest quartile. In EPIC-Norfolk, the relationship ran in the opposite direction: those with the best adherence to the UK's Eatwell Guide were actually least likely to hit the target, with a modest but statistically significant inverse association.

A simulation analysis reinforced these findings, showing it is unlikely to reach 500 mg per day of flavanols by eating typical US fruit and vegetable portions, even when deliberately selecting high-flavanol options.

Why fruit and vegetable guidelines miss the flavanol target

The disconnect is partly structural. Current dietary guidelines treat fruits and vegetables as a category, but flavanol content within that category varies enormously depending on cultivar, growing conditions, climate, and harvest timing. An apple from one variety may contain several times more flavanols than a different variety from a neighboring orchard. Breeders have also historically tried to reduce procyanidins, the compounds that contribute astringency to fruits, inadvertently lowering flavanol content in commercially popular varieties.

Tea turns out to be particularly important. In EPIC-Norfolk, where participants consumed an average of 773 grams of tea per day compared to just 134 grams per day in COSMOS, higher tea intake was associated with significantly increased odds of meeting the 500 mg threshold. Yet in American dietary guidance, tea is often associated with concerns about added sugars and caffeine, which may discourage its use as a deliberate flavanol source. Notably, the biomarkers used in this study do not capture black tea-specific compounds like theaflavins and thearubigins, meaning even the estimated tea contribution may be undercounted.

For context, human intervention trials have shown that flavanol-rich foods can improve endothelial function and reduce cardiovascular risk markers, but reaching those intake levels through general dietary advice appears far harder in practice than in theory.

Study design strengths and honest caveats

Using urinary biomarkers rather than relying solely on food frequency questionnaires is a meaningful methodological step forward: it reduces the misclassification bias that plagues self-reported flavanol data. The biomarker thresholds were also deliberately set to overestimate high intake, which means the true proportion of people reaching 500 mg per day could be even lower than the 17 to 19% figures reported.

Several caveats deserve attention. The COSMOS cohort skewed older (average age 71) and had healthier dietary habits than the general US population, so these estimates likely paint a more optimistic picture than would apply nationally. The gut microbiota-derived biomarker gVLMB also has higher inter-individual variability than SREMB, and biomarker validation was conducted in younger participants, leaving open questions about how age affects results. The study also did not evaluate specifically flavanol-forward dietary patterns like DASH or the Mediterranean diet, though simulation data suggest those patterns would still fall short. Further research is needed to fully characterize the organ-level protective effects of flavanols and determine which dietary configurations reliably produce cardioprotective intake levels.

The case for flavanol-specific dietary reference values

The findings carry a practical implication for nutrition policy. Generic guidance to eat more fruits and vegetables may improve many health markers, but it is not a reliable mechanism for optimizing intake of specific bioactives like flavanols. The authors argue that developing Dietary Reference Intake values, or DRI-like targets, specifically for flavanols could reshape agricultural incentives toward higher-flavanol cultivars and help consumers make more targeted food choices.

Men were more likely than women to reach the 500 mg threshold in both cohorts, a difference worth investigating in future work. And the open question of whether black tea's theaflavins and thearubigins contribute independently to the cardiovascular benefits observed in COSMOS remains unresolved. For now, the data suggest that if flavanols matter for heart health at the levels the COSMOS trial indicates, most people, including those conscientiously following dietary guidelines, are not getting there.

Evidence rating

EMERGING — 82/100

Caveats:

• This is an observational analysis — it shows that typical diets fall short of a beneficial intake threshold, but cannot itself prove that reaching 500 mg/day causes the cardiovascular benefit (that evidence comes from the separate COSMOS RCT).
• Flavanol biomarkers used here do not capture tea-specific compounds (theaflavins, thearubigins), so black tea drinkers' intake may be underestimated.
• The biomarker thresholds were deliberately set to overestimate high flavanol intake, meaning the true proportion reaching 500 mg/day is likely even lower than the ~18–19% reported.
• No funding source was disclosed for this analysis.

Cross-sectional biomarker analysis across two large prospective cohorts (COSMOS, n=6,509; EPIC-Norfolk, n=24,154) using validated urinary flavanol biomarkers. The study demonstrates an association between dietary guideline adherence and flavanol intake levels, but cannot establish causation between flavanol consumption and cardiovascular outcomes — that causal evidence derives from the cited COSMOS RCT. Findings have not yet been independently replicated in a third cohort using the same biomarker approach.

References

1. Javier I. Ottaviani, John W. Erdman, Francene M. Steinberg, et al. (2026). Adhering to dietary guidelines does not yield flavanol intake levels associated with beneficial cardiovascular effects. Food & Function. https://doi.org/10.1039/d6fo00867d
2. Kristi M Crowe-White, Levi W Evans, Gunter G C Kuhnle, et al. (2022). Flavan-3-ols and Cardiometabolic Health: First Ever Dietary Bioactive Guideline. Advances in Nutrition. https://doi.org/10.1093/advances/nmac105
3. Fabian Lanuza, Nicola P. Bondonno, Raul Zamora-Ros, et al. (2022). Comparison of Flavonoid Intake Assessment Methods Using USDA and Phenol Explorer Databases: Subcohort Diet, Cancer and Health-Next Generations—MAX Study. Frontiers in Nutrition. https://doi.org/10.3389/fnut.2022.873774
4. Amy Rees, Georgina Dodd, Jeremy Spencer (2018). The Effects of Flavonoids on Cardiovascular Health: A Review of Human Intervention Trials and Implications for Cerebrovascular Function. Nutrients. https://doi.org/10.3390/nu10121852
5. Agnes Jumar, Roland E. Schmieder (2015). Cocoa Flavanol Cardiovascular Effects Beyond Blood Pressure Reduction. The Journal of Clinical Hypertension. https://doi.org/10.1111/jch.12715