CHAMGAP
APPROVEDReviewed and approved by the Chamgap Editorial Team (2026-07-07). The draft was written by AI, all 12 cited sources were opened and checked for existence, and the verdict passed blind grading and adversarial audit. Methodology v0.6.
Verdict No. 118 · Search date 2026-07-07 · Methodology v0.6

Cocoa flavanols,
does it really help with Cognitive function and vascular/blood-flow health?

30-Second Summary
C
Evidence Grade C · 48 · Safety caution
The evidence is conflicting or limited.
What the
research shows
Cocoa flavanols show repeated small improvement signals for vascular endothelial function (FMD) and blood pressure in human RCTs and meta-analyses. However, cognitive improvement is mainly supported by small short-term studies and acute physiologic-response studies, while large long-term COSMOS cognitive studies did not confirm an overall cognitive effect. Flavanol content, sugars, fat, caffeine, and theobromine in commercial chocolate may differ from standardized research extracts.
What the
ads claim
In the Korean market/articles, Lotte 'Dream Cacao Flavanol' was launched as a health functional food in 2015, and reports repeatedly mentioned 200.2 mg flavanols in 27 g/day, antioxidant and blood-circulation help, and EFSA recognition of improved blood flow. Some articles and informational content broadly connect cacao polyphenols to antioxidant effects, vascular health, blood pressure, prevention of atherosclerosis, improved cognitive ability, and dementia prevention. Recent informational articles also introduce a COSMOS subgroup analysis suggesting cocoa extract may help cognition in older adults with low diet quality, while also mentioning that the overall effect was absent and that chocolate products differ in sugar/saturated fat and flavanol content.
*

Useful facts when choosing a product

  • Lotte 'Dream Cacao Flavanol,' reported in 2015, was said to contain 200.2 mg flavanols in the recommended daily intake of 27 g.
  • The product was reported as being sold in health functional food sections of department stores and discount stores and in pharmacies, with a recommended consumer price of 15,000 won.
  • Domestic reports mixed expressions such as 'blood circulation,' 'improved blood flow,' 'antioxidant,' and 'known to be effective for dementia prevention.' Regulatory recognition was not used as a criterion to raise or lower this evidence grade.
  • Cocoa extract/cocoa drinks used in trials are often research products standardized for flavanol/epicatechin content, making direct extrapolation to ordinary chocolate products difficult.
Gap Measurement · Verdict 118 · C 48
What advertising claims
What independent, higher-quality research supports
△ GAP
01

What the research actually shows

The vascular evidence is relatively consistent. A meta-analysis of 42 acute/short-term RCTs reported improved FMD (chronic +1.34%, acute +3.19%) and diastolic blood pressure by about -1.60 mmHg, and the Cochrane blood-pressure review also judged there was moderate evidence for a small blood-pressure reduction of about 2 mmHg, mainly in healthy adults. However, FMD is a surrogate marker, and the large COSMOS cardiovascular RCT (21,442 people, 500 mg/day cocoa flavanols) was not statistically significant for the primary composite cardiovascular event endpoint (HR 0.90, 95% CI 0.78-1.02; P=0.11). Cognitive evidence is weaker. Two CoCoA RCTs of 90 people and a 37-person dentate gyrus study were positive for TMT/VFT or specific memory tasks, but Mars funding/test-product supply was repeated, and primary/key results depended on multiple cognitive tests or neuroimaging surrogate markers. Conversely, COSMOS-Mind in 2,262 people for 3 years and COSMOS-Clinic in 573 people for 2 years found no significant benefit of cocoa extract 500 mg/day on overall global cognition.

02

Why this is classified as C (48)

Separated by outcome, vascular evidence includes many human RCTs and meta-analyses, but the main effects focus on surrogate markers such as FMD and small blood-pressure changes, so the boundary rule caps it at C. Cognitive evidence includes small positive RCTs but depends heavily on Mars funding/product supply, and large independent long-term COSMOS-Mind and COSMOS-Clinic studies found null overall global cognition effects. Therefore, the combined 'cognition and vascular' claim is C with a score of 48.

Counterpoint. It is difficult to say there is no effect at all. Standardized cocoa flavanols repeatedly show small improvements in endothelial-function and blood-pressure markers, and signals were reported in some cognitive subgroups with lower diet quality or lower flavanol exposure. But these signals have not been established as overall-population cognitive improvement or clinical cardiovascular event prevention.

Rejudgment record. Draft=blinded convergent — Vascular outcomes show modest consistent signals in surrogate markers such as FMD and blood pressure but miss the primary clinical CVD endpoint, and cognition lacks an overall effect in large COSMOS after small/Mars-positive studies, so the compound claim is C

Cross-check — Codex and Claude

This verdict was drafted by Codex through literature review and source-existence checks, cross-checked through blind grading and adversarial audit, and settled by reapplying the methodology boundary rules. Cases with split grades were resolved through rejudgment.
03

Evidence Table

StudyDesignSampleFundingEndpointResultWeight
Study 1gut/gastrointestinal/antioxidantDomestic product report: 27 g/day, flavanols 200.2 mg, blood-circulation and antioxidant claims.Core
Study 2Domestic report used expressions for improved blood circulation and dementia prevention together.Core
Mastroiacovo et al. 2015RCT90Possible manufacturer/industry involvementcognition90 cognitively normal older adults, 8 weeks, 993/520/48 mg flavanols; MMSE null, TMT/VFT positive; Mars support.Core
Desideri et al. 2012RCT90Possible manufacturer/industry involvement90 older adults with MCI, 8 weeks, 990/520/45 mg; MMSE null, TMT/VFT and BP/insulin-resistance positive; Mars support.Core
Baker et al. 20222,2622,262 people, 3 years, cocoa extract 500 mg flavanols/day; no effect on global cognition (z=0.03, P=0.28).Supporting
Vyas et al. 2024Cohort study573573-person COSMOS clinic subcohort, 2 years; primary global cognition null (MD -0.01 SU).Supporting
Brickman et al. 2014RCT37Possible manufacturer/industry involvementmemory37 people aged 50-69, 3 months, 900 vs 10 mg flavanols; DG fMRI and a specific memory task positive; Mars product/support.Supporting
Rendeiro et al. 202017Possible manufacturer/industry involvement17 healthy young men, acute crossover; positive for brain oxygenation response and high-difficulty Stroop, but centered on a small surrogate-marker study.Supporting
Hooper et al. 2012Meta-analysis of RCTs1,29742 acute/short-term RCTs, 1,297 people; improved FMD and DBP -1.60 mmHg, evidence low-moderate.Supporting
Ried et al. 2017blood pressureModerate evidence that flavanol-rich chocolate/cocoa lowers short-term blood pressure by about 2 mmHg, with heterogeneity.Supporting
Sesso et al. 202221,442Possible manufacturer/industry involvement21,442 people, 3.6 years; primary total CVD events HR 0.90 (P=0.11) null, secondary CVD death HR 0.73.Supporting
Socci et al. 2017cognitionReview summarizing cognitive studies: acute/chronic RCTs are partly positive but few and heterogeneous.Supporting
§

Receipt — 12 References

Every cited source was opened and checked against the live page on 2026-07-07.

Reference 1
checked
Reference 2
checked
Mastroiacovo et al., 2015, Am J Clin Nutr, CoCoA RCT
checked
Desideri et al., 2012, Hypertension, CoCoA MCI RCT
checked
Baker et al., 2022, Alzheimer's & Dementia, COSMOS-Mind
checked
Vyas et al., 2024, Am J Clin Nutr, COSMOS-Clinic
checked
Brickman et al., 2014, Nat Neurosci, dentate gyrus RCT
checked
Rendeiro et al., 2020, Scientific Reports
checked
Hooper et al., 2012, Am J Clin Nutr meta-analysis
checked
Ried et al., 2017, Cochrane Review, Effect of cocoa on blood pressure
checked
Sesso et al., 2022, Am J Clin Nutr, COSMOS cardiovascular trial
checked
Socci et al., 2017, Front Nutr, Enhancing Human Cognition with Cocoa Flavonoids
checked
Draft and rewrite: Codex (AI) · Verification: Codex blind grading and adversarial audit · Final adjudication: Claude
Reviewed and approved: Chamgap Editorial Team · Approval date: 2026-07-07 · Corrections: none

Cite this verdict

Cocoa flavanols x cognitive function and vascular/blood-flow health Evidence Grade C card
[Chamgap] Cocoa flavanols x cognitive function and vascular/blood-flow health — Evidence Grade C·48. 12 cited sources checked. Source: https://health-receipt.pages.dev/en/verdicts/cognition/cacao-flavanol-cognition/ · CC BY 4.0

CC BY 4.0 — free to use with attribution; do not distort grades, numbers, or verdict meaning.

!

What this document does and does not do

Chamgap is an information source. It reports what research has and has not confirmed; it does not tell readers what to take or buy. That decision belongs to readers and, when needed, medical or legal professionals. This verdict reflects literature available up to the search date and may change as new research appears. Nothing here is medical advice.