Cocoa flavanols
Cocoa extract standardised to total cocoa flavanols with a declared (−)-epicatechin content — 500 mg flavanols including 80 mg epicatechin per day (the COSMOS formulation). Never Dutch-processed/alkalised cocoa, which destroys the flavanols; chocolate bars are a poor delivery vehicle.
Moderate
Some good studies, not many repeats
The surrogate endpoints (FMD, BP) are replicated across dozens of trials with moderate-certainty Cochrane evidence. But COSMOS — 21,442 people, 3.6 years — missed its primary cardiovascular endpoint (HR 0.90, 95% CI 0.78-1.02); the CVD-death reduction (HR 0.73, 95% CI 0.54-0.98) was secondary. Nothing here maps to a wearable metric: FMD and blood pressure are not in Umbral's metric set, and there is no reliable VO2max or step-count effect. Ship it as a cardio card, not as a fatigue fix.
Across dozens of studies it helped blood vessels relax a little and lowered blood pressure by about 2 points. Small effects.
Across 15 trials, cocoa flavanols improved flow-mediated dilation by 1.17 percentage points (95% CI 0.76 to 1.57) and, in 35 Cochrane trials, lowered blood pressure by 1.76 mmHg systolic (95% CI -3.09 to -0.43) and 1.76 mmHg diastolic (95% CI -2.57 to -0.94).
(−)-Epicatechin increases nitric oxide bioavailability in the vascular endothelium, improving endothelium-dependent vasodilation and modestly lowering peripheral resistance.
Bigger blood-pressure effects in people with elevated baseline BP. Effects are largest in the 150-1000 mg flavanol range; below 150 mg nothing happens.
What to buy
- ✓Cocoa extract standardised to total cocoa flavanols with a declared (−)-epicatechin content — 500 mg flavanols including 80 mg epicatechin per day (the COSMOS formulation). Never Dutch-processed/alkalised cocoa, which destroys the flavanols; chocolate bars are a poor delivery vehicle.
- ✓500-750 mg total cocoa flavanols/day. The dose-response is an inverted U with the peak near 710 mg — more is not better.
- ✓Third-party tested (USP, NSF, Informed Sport)
- ✕Proprietary blends that hide the dose per ingredient
- ✕Unstandardised extracts with no active-compound percentage
- · Cocoa extract carries caffeine and theobromine and can disturb sleep if taken late in the day.
- · Nausea was reported slightly more often than placebo in COSMOS; GI complaints in ~1% vs 0.4% in the Cochrane review.
- · Cocoa products can carry cadmium — prefer brands with third-party heavy-metal testing. Cocoa is also a recognised migraine trigger for some people.
As an Amazon Associate, Umbral earns from qualifying purchases. Product data is provided by Amazon and may change; ratings and prices shown are Amazon’s, not Umbral’s. Recommendations are based on the evidence grade and the form used in trials, never on commission.
Non-alkalised, which is the whole point: Dutch processing destroys the flavanols you are buying it for. The best-selling cocoa flavanol product here.
Buy on Amazon
This is the COSMOS formulation — 500 mg of flavanols with the epicatechin content declared. It is the material the large cardiovascular trial actually used.
Buy on Amazon
The higher-dose version of the same extract, for the upper end of the 500–750 mg range. There is no evidence that going past 750 mg adds anything.
Buy on AmazonPrefer to compare yourself? This opens Amazon’s full selection for this supplement, filtered to the form we recommend.
See all on AmazonEvery result, next to the line that means no effect
Each diamond is a study's best estimate. The thin line shows how sure the study was: the shorter the line, the more certain the result.
- Systolic blood pressuremmHg · Ried 2017, Cochrane CD008893−1.76−3.09 … −0.43
- Diastolic blood pressuremmHg · Ried 2017, Cochrane CD008893−1.76−2.57 … −0.94
- Flow-mediated dilationpercentage points · Sun 2019, Food & Function (15 articles, 18 arms, n=730)+1.17+0.76 … +1.57
- Cardiovascular deathhazard ratio (secondary endpoint) · COSMOS 2022, Am J Clin Nutr+0.73+0.54 … +0.98
Each unit gets its own scale, so numbers are never mixed on one line. When a study gave no range, we leave it blank instead of inventing one.
3 studies over 5 years
The studies
- 1.Dose-response relationship between cocoa flavanols and human endothelial function: a systematic review and meta-analysis of randomized trialsFood & Function · 2019 · meta-analysis · 15 experiments · n = 730
Flavanol intake of 80-1248 mg/day for ≥7 days improved FMD by 1.17% (95% CI 0.76-1.57), with an inverted-U dose-response peaking around 710 mg total flavanols / 95 mg epicatechin.
- 2.Effect of cocoa flavanol supplementation for the prevention of cardiovascular disease events: the COcoa Supplement and Multivitamin Outcomes Study (COSMOS) randomized clinical trialThe American Journal of Clinical Nutrition · 2022 · rct · 1 experiments · n = 21,442
500 mg flavanols/80 mg epicatechin daily over 3.6 years did not significantly reduce total CVD events (HR 0.90, 95% CI 0.78-1.02, P=0.11) but did reduce cardiovascular death (HR 0.73, 95% CI 0.54-0.98).
- 3.Effect of cocoa on blood pressureCochrane Database of Systematic Reviews · 2017 · meta-analysis · 35 experiments · n = 1,804
Flavanol doses of 30-1218 mg/day (mean 670 mg) over 2-18 weeks lowered systolic BP 1.76 mmHg (95% CI -3.09 to -0.43) and diastolic BP 1.76 mmHg (95% CI -2.57 to -0.94); moderate-quality evidence, larger effect in those with elevated baseline BP.
Umbral is a wellness product, not a medical device. It does not diagnose, treat, cure or prevent any disease. Supplement information is educational; talk to a clinician before starting anything, especially if you are pregnant, under 18, managing a condition or taking medication.
These statements have not been evaluated by the Food and Drug Administration.