The Kuna Study: What The Panama Cacao Research Measured
The "Kuna study" is not a study. It is a body of observational research carried out in Panama, reported in several papers by Norman Hollenberg, Naomi Fisher, Marjorie McCullough, Vicente Bayard and colleagues, and none of it randomised anyone to anything. It is the most quoted cacao research in the supplement trade, and its own authors spend a good deal of their text explaining what it cannot show.
This page is about the methodology: who was looked at, what was recorded, what was estimated rather than measured, and which caveats the researchers wrote down themselves.
What the Kuna research actually is
The Kuna, also written Guna, are an indigenous people of Panama. Some live on the San Blas islands, others have moved to the mainland and to Panama City. That split is the whole design: one ancestry, two environments, compared with each other rather than a control group.
A 2009 paper by Hollenberg, Fisher and McCullough in the Journal of the American Society of Hypertension sets out what was recorded. Raised blood pressure was uncommon among Kuna living on the islands, including in older adults; the authors describe "several hundred" island volunteers, and several hundred islanders over 60 with mean systolic pressure below 110 mm Hg. Kuna who had moved to urban areas showed considerably more hypertension, with pressure rising with age. Sodium and chloride intake, assessed by 24-hour recalls and urine collections, exceeded that of most Western populations, so a low-salt diet is not the explanation. Island Kuna were reported to drink several cups of cocoa a day, and their urinary flavanol and nitrate/nitrite levels were higher than those of mainland Kuna.
Notice what that paragraph lacks: a sample size. "Several hundred" is the figure the authors give. For work cited this heavily, that absence is itself worth recording.
The mortality comparison, and what a death certificate cannot tell you
The arm most often reduced to a statistic is Bayard, Chamorro, Motta and Hollenberg in International Journal of Medical Sciences, 2007. It compared causes of death recorded on death certificates between mainland Panama and the San Blas islands over 2000 to 2004, with denominators from the 2004 census. Mainland: 77,375 deaths. San Blas: 558 according to the abstract, 548 according to the methods, a discrepancy the paper does not reconcile.
Age-adjusted rates per 100,000 came out as roughly 83.4 against 9.2 for cardiovascular disease, 68.4 against 4.4 for cancer and 24.1 against 6.6 for diabetes. Table 3 of the same paper reports ischaemic heart disease at 43.9 against 8.7 and stroke at 41.3 against 0.55, so the abstract groups conditions differently from the tables. Infectious disease was the leading cause of death in San Blas.
Two things matter more than the ratios. Cocoa intake was not measured in this study at all: the figure of more than 900 mg of flavanols a day is an estimate carried in from earlier work. And blood pressure was not an outcome here; the introduction mentions it citing prior papers. The authors then name the flaw in their own data source. Diagnoses on death certificates reflect local diagnostic capacity, and they record that no myocardial infarction was logged on Ailigandi in fifteen years despite the island having an ECG machine.
The confounders the authors list themselves
- Death certificates recorded only age. Ethnicity was not recorded, so Kuna identity is inferred from where the deaths occurred.
- Diet, stress, pollution, tobacco use and genetics are discussed as possible explanations, with no data to separate them.
- Migration is not quantified, though the discussion notes that some Kuna work on the mainland and retire to the islands.
- The San Blas population is older than the mainland population; the authors argue this strengthens the contrast, which is an argument rather than an adjustment.
- The 2009 paper states that the authors cannot prove flavanols caused the lower cardiovascular and cancer mortality, calls for a better data source than death certificates, and notes that cocoa trials vary in dose, flavanol content and control design, most without double-blinding.
- The 2009 work was funded by M&M Mars, Inc. The 2007 paper closes by calling for a randomised controlled trial.
That last point is the honest summary of the literature: the people who did the observational work asked for a trial, because they knew what observation can settle. An ecological comparison generates a hypothesis. It cannot isolate one dietary component from everything else that differs between an island and a capital city.
What a label is allowed to say, and what is in a Chaski tube
None of the above is usable as a claim by a brand in Great Britain, and the reason is specific. Commission Regulation (EU) 2015/539 of 31 March 2015 authorises a single entry relating to cocoa flavanols. Its conditions of use require that the product provide a daily intake of at least 200 mg of cocoa flavanols with a degree of polymerisation of 1 to 10, that the consumer be told that intake figure, and that use be limited to cocoa beverages made with cocoa powder, dark chocolate, or capsules and tablets of high-flavanol cocoa extract. Without a measured flavanol figure, the entry is unavailable.
Chaski has no measured cocoa flavanol figure, we have never had one determined, and we will not estimate one from a food table. Nothing on this page is offered as a claim of any kind. For background to cacao drinking, see the origin and history of ceremonial cacao and what "ceremonial grade" actually means; the rest sits in our guides index.
On composition: a Chaski tube holds 600 g of 77% dark chocolate buttons made from single-origin Peruvian and Ecuadorian cacao and coconut sugar, with Lion's Mane as a fruiting body extract, Cordyceps sinensis as mycelium, and Ginkgo Biloba, which is in the process of being replaced by 200 mg of Tremella fuciformis (10:1 fruiting body extract) per daily dose. No emulsifiers, no seed oils, no palm oil, no added cane sugar. The cacao is low-temperature ground. Every batch is tested for lead, cadmium and pesticides. Daily dose on the pack is four buttons. The component list and pack detail are on the Chaski Cacao page.
Frequently Asked Questions
Was the Kuna research a clinical trial?
No. It is observational epidemiology. The work compared Kuna people living on the San Blas islands with Kuna who had moved to mainland Panama and with the wider mainland population, and nobody was randomised to any intervention. The 2007 paper by Bayard and colleagues in International Journal of Medical Sciences compared causes of death recorded on death certificates from 2000 to 2004 and closes by calling for a randomised controlled trial. The 2009 paper in the Journal of the American Society of Hypertension states that the authors cannot prove flavanols caused the differences they recorded.
How many people were in the Kuna studies?
It depends which arm you mean, and the numbers are less precise than the citation count suggests. The 2009 review describes "several hundred" island volunteers and several hundred islanders over 60, without giving an exact figure. The 2007 mortality paper counted 77,375 deaths on the mainland against 558 in San Blas according to its abstract, or 548 according to its methods section, a discrepancy the paper does not resolve. Population denominators came from the 2004 Panamanian census rather than from a cohort the researchers enrolled and followed.
Was cocoa intake measured in the mortality study?
No. That is one of the most commonly missed details. The 2007 paper by Bayard and colleagues worked from death certificates and census denominators, and recorded no dietary data at all. The figure of more than 900 mg of flavanols a day, widely quoted alongside it, is an estimate carried in from earlier work rather than a measurement taken in that study. The 2009 paper does report that urinary flavanol and nitrate/nitrite levels were higher in island Kuna than in mainland Kuna, which is a biochemical comparison rather than an intake record.
What did the researchers say about other explanations?
They listed several and could not separate them. Death certificates recorded only age, and not ethnicity. Diet, stress, pollution, tobacco use and genetics are discussed without data to distinguish them. Migration between island and mainland was not quantified, and the discussion notes that some Kuna work on the mainland and retire to the islands. Infectious disease was the leading cause of death in San Blas. Diagnostic capacity differed between settings: the authors record that no myocardial infarction was diagnosed on Ailigandi in fifteen years despite an ECG machine being present.
Sources
- Hollenberg NK et al., Journal of the American Society of Hypertension, 2009 — review of the Panama observational work.
- Bayard V et al., International Journal of Medical Sciences, 2007 — death-certificate comparison, 2000–2004.
- European Commission, Official Journal of the European Union, 2015 — Regulation (EU) 2015/539.
See what is in the tube
Chaski Cacao — 600 g tube, 77% dark, single-origin Peruvian and Ecuadorian cacao