cocoa
Processed cocoa (Theobroma cacao) supplies flavanols (especially epicatechin and catechin), theobromine, magnesium, and small amounts of caffeine. The mechanism — flavanols → endothelial nitric oxide → vasodilation → small BP reduction — is well-supported. But the large COSMOS trial in 21,442 older adults was NULL for the primary cardiovascular endpoint over 3.6 years (a secondary signal for CV mortality is interesting but post-hoc). Pop-health 'cocoa is heart-protective' framing oversells the trial reality.
Quick decision guide
May help most
Adults who already enjoy dark chocolate and want a small BP / endothelial-function benefit. Not a justified reason to add cocoa solely for cardiovascular prevention given the COSMOS primary-endpoint null.
Common dosing range
200–500 mg cocoa flavanols/day (~1 oz of 70%+ dark chocolate, or a flavanol-standardised cocoa supplement).
When to expect effects
Endothelial-function biomarkers improve in days–weeks; BP changes in 2–8 weeks; hard CVD outcomes were not improved in 3.6 years of COSMOS.
Watch out for
Heavy-metal contamination (cadmium, lead) in dark chocolate and cocoa products is a real problem — Consumer Reports 2022 testing found most products exceeded California Prop 65 daily thresholds at 1 oz/day. Caffeine/theobromine can disturb sleep and amplify reflux.
Evidence snapshot
What is it
Cocoa is the processed seed product of Theobroma cacao, supplying flavanols (especially epicatechin and catechin), theobromine, magnesium, and small amounts of caffeine.
Is it worth it for you?
Use this as a quick fit check, not a diagnosis.
Worth considering if…
Probably skip if…
Evidence at a glance
| Goal | Effect | Best fit | Time |
|---|---|---|---|
Blood pressure reduction Good Evidence | Mean systolic BP reduction of 1.8 mmHg, diastolic 1.8 mmHg over 2–18 weeks (Cochrane 2017) | Adults with prehypertension or mild hypertension looking for dietary BP levers alongside salt reduction and exercise | 2–8 weeks |
Endothelial function (flow-mediated dilation) Good Evidence | +1.34% FMD improvement vs control (42-trial meta-analysis); biomarker-level improvement consistent | Adults with risk-factor-burden endothelial dysfunction looking for adjunctive measures | Hours to days for acute FMD response; weeks for sustained improvement |
Cardiovascular event prevention (primary CV prevention) Mixed Evidence | COSMOS: HR 0.97 (95% CI 0.88–1.07) for primary composite CV endpoint — NULL. Secondary CV mortality signal HR 0.73 (95% CI 0.54–0.98) — exploratory. | None established for primary CV prevention — COSMOS was null on the primary endpoint | 3.6 years of COSMOS follow-up showed no primary-endpoint benefit |
Cognitive function (older adults) Mixed Evidence | Small short-term cognitive-score improvements; longer-term cognitive outcomes in COSMOS-Mind were null overall | Possibly older adults with poor diet quality looking for a modest dietary lever | Weeks for acute biomarker / cognitive-test changes; long-term benefit unclear |
Blood pressure reduction
- Effect
- Mean systolic BP reduction of 1.8 mmHg, diastolic 1.8 mmHg over 2–18 weeks (Cochrane 2017)
- Best fit
- Adults with prehypertension or mild hypertension looking for dietary BP levers alongside salt reduction and exercise
- Time
- 2–8 weeks
Endothelial function (flow-mediated dilation)
- Effect
- +1.34% FMD improvement vs control (42-trial meta-analysis); biomarker-level improvement consistent
- Best fit
- Adults with risk-factor-burden endothelial dysfunction looking for adjunctive measures
- Time
- Hours to days for acute FMD response; weeks for sustained improvement
Cardiovascular event prevention (primary CV prevention)
- Effect
- COSMOS: HR 0.97 (95% CI 0.88–1.07) for primary composite CV endpoint — NULL. Secondary CV mortality signal HR 0.73 (95% CI 0.54–0.98) — exploratory.
- Best fit
- None established for primary CV prevention — COSMOS was null on the primary endpoint
- Time
- 3.6 years of COSMOS follow-up showed no primary-endpoint benefit
Cognitive function (older adults)
- Effect
- Small short-term cognitive-score improvements; longer-term cognitive outcomes in COSMOS-Mind were null overall
- Best fit
- Possibly older adults with poor diet quality looking for a modest dietary lever
- Time
- Weeks for acute biomarker / cognitive-test changes; long-term benefit unclear
Evidence for 4 uses
AI-assisted evidence assessment — talk to your doctor before relying on any single supplement.
Blood pressure reduction
Supplement benefitRied's 2017 Cochrane review pooled 35 RCTs (n=1,804) and found cocoa flavanols modestly reduced systolic BP by 1.8 mmHg (95% CI 0.7–2.9) and diastolic by 1.8 mmHg (95% CI 0.5–3.0) over typical 2–18 week trials. This is a real but small effect — comparable to a moderate reduction in dietary salt. Useful as one lever in a comprehensive lifestyle approach to BP control, not as a substitute for first-line antihypertensives in people who need them.
Bottom line: Modest, real BP-lowering effect. Use as one piece of a lifestyle approach, not a substitute for prescribed therapy.
Endothelial function (flow-mediated dilation)
Supplement benefitHooper's 2012 meta-analysis of 42 RCTs showed cocoa significantly improves flow-mediated dilation (FMD) by ~1.3 percentage points — a clinically meaningful improvement in arterial endothelial responsiveness. Mechanism is well-characterised: epicatechin → endothelial nitric oxide synthase activation → improved vasodilation. This is the strongest biomarker-level case for cocoa flavanols' cardiovascular relevance. The challenge is that biomarker improvement hasn't translated to a primary CV endpoint benefit in COSMOS.
Bottom line: Real biomarker improvement. The biomarker-to-outcome translation is where the optimism breaks down.
Cardiovascular event prevention (primary CV prevention)
Supplement benefitThe COSMOS trial (Sesso 2022 in Am J Clin Nutr) is the definitive primary-CV-prevention trial of cocoa flavanols: 21,442 older US adults randomised to 500 mg flavanols/day (with 80 mg epicatechin) or placebo, followed 3.6 years. The PRIMARY composite cardiovascular endpoint was NOT significantly reduced (HR 0.97, 95% CI 0.88–1.07). A secondary cardiovascular mortality signal (HR 0.73) is interesting but post-hoc — the kind of result that needs replication before becoming a treatment indication. Pop-health framing of 'cocoa flavanols protect your heart' as established fact misrepresents the trial reality.
Bottom line: Don't take cocoa supplements expecting heart-attack prevention. The big definitive trial said no.
Evidence is mixed
Biomarker meta-analyses (BP, FMD) are consistent and modestly positive. The much larger and longer COSMOS trial in a primary-prevention population was null for the composite cardiovascular endpoint, with only a secondary cardiovascular-mortality signal that needs replication. Pop-health 'cocoa is heart-protective' framing oversells what's actually been demonstrated.
Cognitive function (older adults)
Supplement benefitSeveral short-term trials of cocoa flavanols in older adults have reported modest improvements in cognitive scores (working memory, processing speed), and FMD-improvement-driven cerebral blood flow is a plausible mechanism. COSMOS reported on cognitive outcomes (the COSMOS-Mind sub-study) and showed no significant overall cognitive benefit at 2–3 years vs placebo, though a subgroup analysis suggested possible benefit in adults with poorer baseline diet quality. Overall the cognitive case for cocoa is weak and not currently a justifying indication.
Bottom line: Weak cognitive signal. Not a meaningful 'brain supplement.'
How it works
How to take it
What to track
Bottom line: 200–500 mg flavanols/day in the morning, preferably from a flavanol-standardised supplement or a tested low-heavy-metal dark chocolate.
5 commercial forms
Compare the main delivery options and what they’re best suited for.
Flavanol-standardised cocoa supplement (CocoaVia / COSMOS form)
Trial-equivalentStandardised cocoa extract that delivers a specified flavanol dose (e.g. 500 mg/day with 80 mg epicatechin in COSMOS). Clean dose, controlled processing, typically tested for heavy metals. The form that has actual trial-grade evidence behind it.
Standardised flavanol content; reference for clinical trials.
Dark chocolate (70%+ cocoa)
Dietary sourceWhole-food source of cocoa flavanols + sugar + fat. Flavanol content varies widely between brands and processing methods. Provides ~50–500 mg flavanols per ounce depending on the product. Tasty but heavy-metal-contamination risk is real — choose tested brands.
Variable flavanol content; Dutch processing dramatically reduces it.
Cocoa powder (unsweetened)
Cooking ingredientDefatted cocoa powder used in baking and beverages. Non-alkalised ('natural process') retains more flavanols than alkalised ('Dutch process'). Heavy-metal content varies by brand and source region.
Non-alkalised retains more flavanols than Dutch-process.
Raw cacao / cacao nibs
Whole-food artisanalUnprocessed or minimally-processed cacao beans / nibs marketed as a 'superfood.' Higher flavanol retention than heavily processed cocoa, but also the form most commonly flagged for high cadmium and lead contamination (Consumer Reports 2022). Choose tested brands or avoid.
Higher flavanol content but higher heavy-metal contamination risk.
Dutch-process / alkalised cocoa
Low flavanolCocoa treated with alkali to reduce acidity and bitterness. The alkalisation process destroys most of the flavanols. Common in commercial chocolate products and hot chocolate mixes. Not a meaningful source of cocoa flavanols.
Alkalisation destroys most flavanols.
Safety
Know the common side effects, key cautions, and who should avoid it.
Common side effects
Serious risks
Heavy metal contamination — Consumer Reports 2022 testing of 28 dark chocolate bars found 23 exceeded California Prop 65 maximum allowable daily levels for lead, cadmium, or both at 1 oz/day intake. Some products exceeded the safe daily level by 4–5×. Cumulative cadmium exposure is a long-term renal and bone concern; lead exposure is a developmental concern especially in children and pregnancy. Choose brands with published below-Prop-65 third-party testing.
Theobromine toxicity in dogs and cats — theobromine is metabolised slowly in dogs/cats and is acutely toxic at small doses for body weight. Even small amounts of dark chocolate can kill small dogs. Keep chocolate and cocoa supplements away from pets and call a veterinary emergency line for accidental ingestion.
Atrial fibrillation episodes in susceptible individuals — high cocoa/chocolate intake has been linked to AF triggering in some case reports, likely via caffeine and theobromine.
Migraine triggering — cocoa is a known dietary migraine trigger for a subset of migraine patients via tyramine and other vasoactive compounds.
Who should avoid it
- Pregnant women and children at multi-ounce-per-day intakes of dark chocolate from untested brands — heavy-metal exposure is the main concern.
- People with severe reflux (GERD) — cocoa is a documented reflux trigger.
- People sensitive to caffeine or theobromine, or with anxiety / palpitations triggered by stimulants.
- Migraine sufferers who have identified cocoa/chocolate as a personal trigger.
- Anyone with active atrial fibrillation or paroxysmal AF triggered by stimulants.
- Dogs, cats, parrots — theobromine is toxic.
Pregnancy & breastfeeding
Moderate cocoa intake (a square of dark chocolate / day) is generally considered acceptable in pregnancy. Heavy intakes of dark chocolate or cocoa supplements expose the fetus to caffeine, theobromine, AND cadmium/lead — multiple per-day servings of untested brands are best avoided. If you want the cocoa-flavanol BP / endothelial benefits in pregnancy, choose a brand with published below-Prop-65 heavy-metal testing and discuss with your obstetrician.
Bottom line: The biggest practical safety concern is heavy-metal contamination — pick a tested brand. Caffeine/theobromine sleep effects and reflux are the next-most-common issues.
Interactions
Cocoa contains modest amounts of tyramine. Combined with MAOIs, large intakes could theoretically trigger hypertensive crisis. Practical risk at typical cocoa intakes is low but the combination should be discussed with the prescriber.
Cocoa flavanols modestly lower BP, potentially compounding antihypertensive effect. Usually small enough not to require dose adjustment; monitor BP if adding regular cocoa intake.
Caffeine and theobromine in cocoa compound stimulant effects. Track total caffeine equivalents to avoid jitter, palpitations, sleep disturbance.
Cocoa flavanols have mild antiplatelet activity. Clinically significant warfarin interaction is rare at typical intakes; monitor INR if substantially increasing cocoa intake.
Cocoa contains modest oxalates and calcium that can reduce absorption of bisphosphonates and tetracyclines if taken together. Separate by 2 hours.
Documented interactions
Evidence-graded pair pages with sources, dosing notes, and timing guidance — a complement to the narrative section above.
See all 1 cocoa interaction →Food sources
| Food | Amount | %DV |
|---|---|---|
| Dark chocolate, 85% cocoa | 1 oz (~200–500 mg cocoa flavanols, brand-dependent) | — |
| Dark chocolate, 70% cocoa | 1 oz (~150–300 mg cocoa flavanols, brand-dependent) | — |
| Cocoa powder, non-alkalised | 1 tbsp (~150–250 mg cocoa flavanols) | — |
| Cocoa powder, Dutch-processed | 1 tbsp (~30–80 mg cocoa flavanols, reduced by alkalisation) | — |
| Milk chocolate | 1 oz (~30–80 mg cocoa flavanols) | — |
| Hot chocolate mix, commercial | 1 packet (typically <50 mg cocoa flavanols) | — |
| Cacao nibs, raw | 1 tbsp (~150–300 mg cocoa flavanols; high heavy-metal risk) | — |
Dark chocolate, 85% cocoa
- Amount
- 1 oz (~200–500 mg cocoa flavanols, brand-dependent)
- %DV
- —
Dark chocolate, 70% cocoa
- Amount
- 1 oz (~150–300 mg cocoa flavanols, brand-dependent)
- %DV
- —
Cocoa powder, non-alkalised
- Amount
- 1 tbsp (~150–250 mg cocoa flavanols)
- %DV
- —
Cocoa powder, Dutch-processed
- Amount
- 1 tbsp (~30–80 mg cocoa flavanols, reduced by alkalisation)
- %DV
- —
Milk chocolate
- Amount
- 1 oz (~30–80 mg cocoa flavanols)
- %DV
- —
Hot chocolate mix, commercial
- Amount
- 1 packet (typically <50 mg cocoa flavanols)
- %DV
- —
Cacao nibs, raw
- Amount
- 1 tbsp (~150–300 mg cocoa flavanols; high heavy-metal risk)
- %DV
- —
Choosing a product
What to look for on the label — and what to be skeptical of.
Look for…
Be skeptical of…
Frequently asked questions
Is dark chocolate the same as cocoa flavanols?⌄
Dark chocolate contains flavanols but amounts vary widely by brand and processing; standardized supplements provide a consistent dose.
How much caffeine is in cocoa?⌄
About 12 mg per tablespoon of cocoa powder, plus 150-250 mg of theobromine, which has milder stimulant effects.
References by claim
Track cocoa with Pilora
Set up dose reminders, check interactions, and join the community in the Pilora iPhone app.
Coming to App StoreDisclaimer: These statements have not been evaluated by the FDA. This page is educational, not a substitute for personalized medical advice. Evidence grades are AI-assisted assessments — talk to your doctor before starting any new supplement, especially if you’re pregnant, breastfeeding, on medications, or managing a chronic condition.