Polyphenols and Aging: Beyond the Headlines

There is a reliable gap in the polyphenol literature. Diets rich in these plant compounds are associated with better outcomes across large populations. Trials that isolate a polyphenol and give it as a supplement mostly fail to reproduce that. Both findings are real, and the interesting question is what sits between them.

What the observational data says

The epidemiology is consistent. A 2017 dose-response meta-analysis in the American Journal of Epidemiology pooled 22 prospective cohort studies and found that high total flavonoid consumption was associated with a lower risk of all-cause mortality (RR 0.74, 95% CI 0.55–0.99). Each additional 100 mg/day of flavonoid intake tracked with roughly 6% lower all-cause mortality and 4% lower cardiovascular mortality, and the associations held across several subclasses — flavonols, flavones, flavanones, anthocyanidins and proanthocyanidins.

That is a genuine signal. It is also entirely observational: people who eat more flavonoids eat more plants, and differ in a hundred other ways that no statistical adjustment fully removes.

What happens when you test it

The most decisive test to date is COSMOS, published in the American Journal of Clinical Nutrition in 2022. It randomized 21,442 older US adults — 12,666 women aged 65 or over and 8,776 men aged 60 or over — to a cocoa extract supplying 500 mg of cocoa flavanols daily (including 80 mg of epicatechin) or placebo, and followed them for a median of 3.6 years. It is the largest trial of a polyphenol supplement ever conducted. The cocoa extract and infrastructure support came from Mars Edge, with NIH grant funding.

Endpoint	Hazard ratio (95% CI)	Interpretation
Total cardiovascular events (primary)	0.90 (0.78–1.02), P = 0.11	Not significant — primary endpoint missed
Cardiovascular death (secondary)	0.73 (0.54–0.98)	Significant, but one of several secondary endpoints
Myocardial infarction	0.87 (0.66–1.16)	No significant effect
Stroke	0.91 (0.70–1.17)	No significant effect
All-cause mortality	0.89 (0.77–1.03)	No significant effect
Total CVD, adherence-adjusted	0.85 (0.72–0.99)	Per-protocol analysis, not the pre-specified primary result

The largest polyphenol supplement trial ever run did not meet its primary endpoint. The 27% reduction in cardiovascular death that made the headlines was a secondary outcome in a trial whose main comparison was null.

Smaller compounds tell a similar story with even less to work with. A Cochrane review of green and black tea for primary cardiovascular prevention identified 11 randomized trials lasting at least three months, covering 821 people in total. Not one of them reported a cardiovascular event. The review could only assess risk-factor surrogates — black tea reduced LDL cholesterol and blood pressure, green tea reduced total and LDL cholesterol and blood pressure, though the green tea results were unstable in sensitivity analysis. The authors’ conclusion was that the evidence is limited and should be treated with caution.

Why the supplement version underdelivers

Bioavailability is the recognised bottleneck, and it is not a small one.

• Only small polyphenols are absorbed directly in the small intestine. Complex polyphenols pass through undigested and reach the colon, where gut bacteria transform them into the metabolites that actually enter circulation.
• Because that conversion depends on an individual’s microbiota, the same dose produces very different exposures in different people. A high polyphenol content in a food does not imply high bioavailability of that polyphenol.
• The scale of that variation is measurable. In one crossover study of 100 healthy adults, only 12% had detectable urolithin A — a gut-derived metabolite of pomegranate and walnut ellagitannins — in plasma on their normal diet, and roughly 40% converted the precursors efficiently after pomegranate juice. The rest converted poorly or not at all.

So a capsule of an isolated polyphenol is making two bets at once: that the compound was the active ingredient in the food, and that your gut will turn it into something that reaches your tissues. Neither is guaranteed.

The takeaway

The dietary evidence for polyphenol-rich foods is good, consistent, and observational. The supplement evidence is thinner than the marketing implies: the biggest trial ever run missed its primary endpoint, the tea literature has never recorded a cardiovascular event, and bioavailability varies so much between people that a labelled dose is a poor guide to what actually circulates. Eating the plants is well supported. Buying the isolated compound is not — not because it has been disproven, but because the trials that would justify it have either not been done or have come back null. We look at two specific cases in more detail in urolithin A and mitophagy and sirtuins and resveratrol.

Sources

• Grosso et al., “Dietary Flavonoid and Lignan Intake and Mortality in Prospective Cohort Studies: Systematic Review and Dose-Response Meta-Analysis” (American Journal of Epidemiology, 2017)
• Sesso et al., “Effect of cocoa flavanol supplementation for the prevention of cardiovascular disease events: the COSMOS randomized clinical trial” (American Journal of Clinical Nutrition, 2022)
• Hartley et al., “Green and black tea for the primary prevention of cardiovascular disease” (Cochrane Database of Systematic Reviews, 2013)
• Lippolis et al., “Bioaccessibility and Bioavailability of Diet Polyphenols and Their Modulation of Gut Microbiota” (International Journal of Molecular Sciences, 2023)
• Singh et al., “Direct supplementation with Urolithin A overcomes limitations of dietary exposure and gut microbiome variability” (European Journal of Clinical Nutrition, 2021)

This is sample content created during site scaffolding. Sources have been attached but each claim still needs verification against them before launch.