Cocoa has a better case than most "healthy treat" claims, and a new review adds to it. In September 2026, researchers published an updated systematic review and meta-analysis in The Journal of Nutrition, Health & Aging. It pooled 25 randomised controlled trials, with 1,167 participants, that gave people cocoa flavanols and measured their blood pressure or blood vessel function.
What it found
Systolic blood pressure, the top number, fell by an average of 2.7 mmHg. Diastolic pressure fell by 1.8 mmHg. The result held both for people with high blood pressure and for people with normal readings. The authors rated their certainty in the systolic result as moderate.
Two measures of blood vessel health also improved: flow-mediated dilation, which shows how well an artery widens, and pulse wave velocity, a marker of arterial stiffness. The authors say the evidence for these is less certain.
The more flavanols people took, the bigger the fall in blood pressure, but the review found no clear threshold. Doses in the trials ranged from about 11 mg to 1,680 mg of flavanols a day.
How big is 2.7 mmHg?
Small for one person. A single blood pressure medicine usually lowers it by several times as much. A couple of millimetres across a whole population is not trivial, but it will not show up as a difference on your own monitor.
It also matches what was already known. A 2017 Cochrane review of 35 trials found a similar, modest fall, and the large COSMOS trial of cocoa extract capsules did not significantly reduce total cardiovascular events. Our cacao page covers both.
Who paid for it
Five of the seven authors work for Meiji, a Japanese company that makes chocolate and other cocoa foods, and Meiji funded the review. The other two work for a Tokyo consultancy. The authors state that the companies had no role in the study design or results, and the review was registered in advance, which helps. Still, cocoa research is often funded by the cocoa industry, and it is worth knowing who paid when you read headlines about it.
Does chocolate count?
Not reliably. Flavanols are lost during fermentation, roasting and especially "Dutch" alkalising, so two cocoa powders with the same label can contain very different amounts. Most packs do not say how much they contain. Milk chocolate also adds sugar and fat, which defeats the point.
If you want cocoa to do something for you, natural (non-alkalised) cocoa powder or cacao stirred into porridge or a drink is a better choice than a chocolate bar. Sweeten it lightly, for example with coconut sugar or stevia. And if your blood pressure is high, cocoa is an extra, not a replacement for treatment.
