Quick answer: Cocoa flavanols and heart disease are linked by the COSMOS trial, the largest study of its kind: among adults over 60, a daily cocoa flavanol supplement was associated with a 27 percent lower rate of death from cardiovascular disease compared with a placebo. Two caveats matter. The trial tested a concentrated supplement delivering 500 milligrams of flavanols a day, not chocolate, and that 27 percent was a secondary finding. The study's primary goal, cutting total cardiovascular events, did not reach statistical significance.
"Chocolate is good for your heart" is one of the most durable headlines in nutrition, and it usually overstates a small, messy body of evidence. The COSMOS trial changed that conversation, not by proving chocolate is medicine, but by finally being large and rigorous enough to give a real answer about cocoa flavanols, the plant compounds at the center of the claim. The results were encouraging and more nuanced than the headlines suggested. This piece explains what the COcoa Supplement and Multivitamin Outcomes Study (COSMOS) actually tested, what it found, what the United States Food and Drug Administration (FDA) concluded afterward, and why a chocolate bar is not the same as the supplement it used.
What was the COSMOS trial?
COSMOS was a randomized, double-blind, placebo-controlled trial, the gold standard for testing whether a supplement actually causes an effect. Researchers led by Brigham and Women's Hospital and Harvard Medical School enrolled 21,442 older adults, 12,666 women aged 65 and older and 8,776 men aged 60 and older, and followed them for an average of about 3.6 years between 2014 and 2020. Participants took either a cocoa extract supplement providing 500 milligrams of cocoa flavanols per day, a daily multivitamin, both, or a placebo.
That scale is what made the study matter. Earlier research had hinted that cocoa flavanols could support blood vessel function and blood pressure, but those studies were mostly small and short. COSMOS was the first trial large and long enough to ask whether a measured daily dose changes hard outcomes like heart attacks, strokes, and cardiovascular death. Its results were published in the American Journal of Clinical Nutrition in 2022.
What the COSMOS trial found about cocoa flavanols and heart disease
The honest headline is "promising, with an asterisk." The trial's primary goal was to reduce total cardiovascular events, a combined measure of heart attacks, strokes, and related outcomes. Cocoa flavanols lowered that combined measure by about 10 percent, but the result did not reach statistical significance, meaning it could have been due to chance. Where the signal was strongest was a pre-specified secondary outcome: death from cardiovascular disease fell by 27 percent in the cocoa flavanol group, a result that was statistically significant.
When researchers looked only at participants who took the supplement consistently, a per-protocol analysis that adjusts for adherence, the reductions grew larger. The table below summarizes the key numbers and, just as importantly, how much weight each one carries.
| Outcome measured | Result vs. placebo | How much weight it carries |
|---|---|---|
| Total cardiovascular events (the trial's main goal) | ~10% lower | Did not reach statistical significance |
| Death from cardiovascular disease | 27% lower | Statistically significant secondary finding |
| Total cardiovascular events, consistent users | ~15% lower | Per-protocol analysis, adjusted for adherence |
| Cardiovascular death, consistent users | ~39% lower | Per-protocol analysis, adjusted for adherence |
Read carefully, the trial does not say "eat cocoa flavanols and you will not have heart disease." It says a daily supplement was associated with meaningfully fewer cardiovascular deaths in older adults, strongest among those who took it reliably, while the broad primary measure fell short of significance. That is an encouraging signal that warrants more research, not a guarantee.
What the FDA concluded about cocoa flavanols
In February 2023, the FDA announced it would not object to a "qualified" health claim connecting cocoa flavanols in high-flavanol cocoa powder to a reduced risk of cardiovascular disease. A qualified claim is the weakest tier the agency allows: it requires language making clear the evidence is thin. The approved wording is, in effect, that cocoa flavanols in high-flavanol cocoa powder "may reduce the risk of cardiovascular disease, although the FDA has concluded that there is very limited scientific evidence for this claim."
The reference amount the FDA used is 200 milligrams of cocoa flavanols per day, roughly one tablespoon (5 to 6 grams) of high-flavanol cocoa powder. That figure is worth holding onto: it sets a realistic bar for "a meaningful dose," and it is far below the 500 milligrams used in COSMOS.
Why a chocolate bar is not the same as the supplement
This is the part the headlines almost always skip. COSMOS tested a concentrated cocoa extract, not chocolate, and the flavanol content of everyday chocolate is both lower and wildly variable. Flavanols are fragile compounds that are diminished by fermentation, roasting, and especially alkalizing, the processing step that makes cocoa darker and smoother. Two bars can both read "70% cacao" and deliver very different amounts of surviving flavanol, and most will fall well short of 200 milligrams in a sensible portion.
That gap is why processing and sourcing matter more than the percentage on the front. Marmels is made from 62 percent single-origin organic cacao that is minimally processed and not alkalized, so more of the bean's natural flavanols survive than in heavily Dutched cocoa, alongside 12 grams of protein per 60-gram bar. To be clear, a chocolate bar is a treat, not a substitute for the supplement studied in COSMOS, and it makes no claim to reduce heart disease. If you enjoy dark chocolate, minimally processed cacao simply keeps more of what makes the cocoa bean interesting.
Frequently Asked Questions
Do cocoa flavanols really reduce heart disease?
The evidence is promising but not settled. In the COSMOS trial, a daily 500-milligram cocoa flavanol supplement was associated with 27 percent fewer cardiovascular deaths in adults over 60, a statistically significant secondary finding. However, the trial's main goal of reducing total cardiovascular events did not reach significance. The honest summary is an encouraging signal that still needs confirmation, not proof of protection.
What was the COSMOS trial?
COSMOS, the COcoa Supplement and Multivitamin Outcomes Study, was a randomized, double-blind, placebo-controlled trial of 21,442 adults over 60, run by Brigham and Women's Hospital and Harvard Medical School over about 3.6 years. It is the largest and longest trial to test cocoa flavanols against cardiovascular outcomes.
How many cocoa flavanols are in dark chocolate?
It varies enormously, and it is rarely printed on the label. Flavanol content depends on the bean and, more so, on processing: fermentation, roasting, and alkalizing all reduce it. A typical dark chocolate serving delivers far less than the 200 milligrams the Food and Drug Administration uses as a reference, which is part of why COSMOS used a concentrated 500-milligram dose.
Is dark chocolate as good as a cocoa flavanol supplement?
No. The COSMOS results came from a standardized supplement delivering 500 milligrams of flavanols a day, an amount most chocolate cannot match in a reasonable portion. Dark chocolate can be a pleasant source of some flavanols, especially if minimally processed, but it should be enjoyed as a treat, not relied on as a measured dose.
What is the FDA's cocoa flavanol health claim?
In 2023 the Food and Drug Administration allowed a qualified health claim stating that cocoa flavanols in high-flavanol cocoa powder may reduce the risk of cardiovascular disease, while noting that the supporting evidence is very limited. "Qualified" is the agency's weakest claim tier and requires that caveat. The reference dose is 200 milligrams of cocoa flavanols per day, about one tablespoon of high-flavanol cocoa powder.
How much cocoa flavanol should you have a day?
There is no official recommended intake, but the most-cited reference points are 200 milligrams a day, the Food and Drug Administration's figure for its qualified claim, and the 500 milligrams used in COSMOS. Both come from concentrated cocoa, not chocolate bars. If you are considering a supplement for heart reasons, talk to your clinician rather than treating food as medicine.
The COSMOS trial moved cocoa flavanols from "interesting" to "worth taking seriously," while keeping everyone honest about the limits of the evidence. Chocolate remains a pleasure first, with the science as a bonus when the cacao is high quality and minimally processed. You can read how we keep ours close to its natural state on the science behind our bars.
