Thursday, August 21, 2008

Kitava: Wrapping it Up

There's a lot to be learned from the Kitava study. Kitavans eat a diet of root vegetables, coconut, fruit, vegetables and fish and have undetectable levels of cardiovascular disease (CVD), stroke and overweight. Despite light smoking. 69% of their calories come from carbohydrate, 21% from fat and 10% from protein. This is essentially a carbohydrate-heavy version of what our paleolithic ancestors ate. They also get lots of sunshine and have a moderately high activity level.

The first thing we can say is that a high intake of carbohydrate is not enough, by itself, to cause overweight or the diseases of civilization. It's also not enough to cause insulin resistance. I sent an e-mail to Dr. Lindeberg asking if his group had measured Kitavans' glucose tolerance. He told me they had not. However, I can only guess they had good glucose control since they suffered from none of the complications of unmanaged diabetes.

The Kitavan diet is low in fat, and most of the fat they eat is saturated because it comes from coconuts. Compared to Americans and Swedes, they have a high intake of saturated fat. So much for the theory that saturated fat causes CVD... They also have a relatively high intake of fish fat, at 4g per day. This gives them a high ratio of omega-3 to omega-6 fatty acids, with plenty of DHA and EPA.

Their blood lipid profile is not what a mainstream cardiologist would expect. In fact, it's "worse" than the Swedish profile in many ways, despite the fact that Swedes are highly prone to CVD. This raises the possibility that blood lipids are not causing CVD, but are simply markers of diet and lifestyle factors. That's very easy for me to swallow because it never made sense to me that our livers would try to kill us by secreting triglycerides and withholding HDL. The blood lipid profile that associates best with CVD and metabolic syndrome in the West (but has no relation to them on Kitava) is one that's consistent with a high carbohydrate intake. Where does carbohydrate come from in the West? White flour and sugar maybe?

Kitavans also have very low serum leptin. This may be a keystone to their leanness and health. It suggests that their diet is not interfering with the body's metabolic feedback loops that maintain leanness.

The Kitavan diet is one path to vibrant health. Like many other non-industrial groups, Kitavans eat whole, natural foods that are broadly consistent with what our hunter-gatherer ancestors would have eaten. It amazes me that as humans, we can live well on diets that range from near-complete carnivory to plant-rich omnivory. We are possibly the most adaptable species on the planet.

The ideal diet for humans includes a lot of possibilities. I believe the focus on macronutrients is misguided. There are examples of cultures that were/are healthy eating high-fat diets, high-carbohydrate diets and everything in between. What they do not eat is processed grains, particularly wheat, refined sugar, industrially processed seed oils and other modern foods. I believe these are unhealthy, and this is visible in the trail of destruction they have left around the globe. Its traces can be found in the Pacific islands, where close genetic relatives of the Kitavans have become morbidly obese and unhealthy on a processed-food diet.


Methuselah said...

I am definately not an expert in this field, but am wondering about your statement

The first thing we can say is that a high intake of carbohydrate is not enough, by itself, to cause overweight or the diseases of civilization.

Is it possible that there is a further factor that was not measured, possibly non-dietary, which had a sufficiently positive effect that it was able to counter-balance the effects carbohydrates would normally have on CVD?

I suppose I feel a more accurate statement would be "...a high carbohydrate diet does guarantee CVD..."

This way we do not imply that an additional factor is required to cause the CVD when it could in fact be that an additional factor in this case mitigated the CVD-inducing effects of the carbohydrate.

What do you think?

Pay Now Live Later

Stephan Guyenet said...

Hi Methuselah,

Yes, that's always a possibility. And the way you re-worded the statement is more logically rigorous of course.

However, I don't think the scenario that something is protecting them against carbohydrate toxicity is the most likely one. Why would we need a protective factor to keep us from being damaged by something we've evolved to tolerate over millions of years?

I think these data are consistent with data from other non-industrial cultures in other parts of the world, as well as with studies of modern-day "paleolithic diets" in Western cohorts (although those have not been very good studies to date).

There probably are some protective factors actually, like their high vitamin D levels, activity level and maybe other unidentified factors. But I think the most parsimonious explanation of the data as a whole is that this kind of diet is healthy on a societal level.

Sue said...

Perhaps their diet is more nutrient dense particularly with the fat soluble vitamins and that is protective somehow.

Gyan said...

Processed grains mean exactly what?
Is traditional milling included in "processed grains"?
Poor Indians with very high grain intake (generally fresh) do not develop metabolic syndrome.

Stephan Guyenet said...


Perhaps. Lindeberg didn't measure the fat-soluble vitamin content of their diet. I suspect they have a good vitamin status because of the straight teeth I've seen on them in pictures. They're definitely getting plenty of vitamin D.


What I mean by 'processed grains' is any grain that has been refined. The issue is actually more complex than that however. The most important thing is whether or not a grain has been treated in a way that minimizes anti-nutrients and other toxins that it naturally contains. Traditional societies (including the US and Europe) used soaking, sprouting and fermentation to make grains more digestible.

The modern habit of grinding wheat and baking quick-rise bread out of it guarantees that you will bear the full brunt of the anti-nutrients. Ironically, whole wheat is worse than refined wheat in this regard. Traditional sourdough is better because it contains lactobacilli that chew up the anti-nutrients and it's fermented for longer. However, I think wheat is unique in that you can't completely detoxify it even by this method.

I don't know which poor Indians you're referring to, but certainly not N American Indians. As a whole, they have some of the highest rates of obesity and diabetes of any populations in the world. They also eat a poor version of the American diet, rich in refined wheat and sugar.

You might be referring to S American Indians, who are probably eating corn. In that case, they are almost certainly nixtamalizing the corn, a traditional process that removes certain anti-nutrients. I agree that certain properly-treated grains can be part of a healthy diet.

Unknown said...

Coincidentally, just after reading that blog post I saw this article on a news site:

Unfotunately it didn't indicate what type of carbs were studied.

Stephan Guyenet said...

Hey Reid,

Yeah I read that study. I believe the media reports are overinterpreting the finding, as usual.

Thackray said...

Carbohydrates fall into two categories: 1) those which are digested in the small intestine and 2) those which are digested in the large intestine. Cat #1 (my nomenclature) causes insulin to rise, Cat #2 enters the blood stream as volatile fatty acids which have no effect on blood sugar or insulin. If these people were eating primitive root vegetables and primitive fruits (that have not been selectively breed for sweetness like the ones available to us today) then most of the carbohydrate in their diet would fall into the Cat #2 category which would reduce the effective (ala Dr Eades) carbohydrates to perhaps as little as 20% of the diet.

The distinction between Cat #1 and Cat #2 carbohydrates is not the same as simple versus complex carbohydrate. Many so called complex carbohydrates are digested in the small intestine and affect insulin levels. The only real way to determine effective carbohydrate is to test insulin response of the actual food on actual people. In this case, the actual food in what the Kitavans ate and the actual people are the Kitavans whose insulin response to Cat #1 carbohydrate may differ from more studied populations.

In my opinion, the Katavin diet might easily be viewed as low carbohydrate given the high percentage of Cat #2 carbohydrates and a possible limited insulin response to Cat #1 carbs inherent in the people themselves.

Thackray said...

As a follow on to my post above, I note that many tubers store their carbohydrate as inulin and oligosaccharides which are indigestible in the small intestine (zero effective carbohydrates) but are digested in the large intestine. So insulin is unaffected by consuming the carbohydrates in such tubers. Here is more information about inulin:

Tubers that store their carbohydrates as inulin and oligosaccharides typically do not store carbohydrates as starch (like a potato does).

So the effective carbohydrate percentage of the Kitavan diet my be dramatically lower than the total carbohydrate input.

Since we don’t know the effective carbohydrate percentage of the Kitavan diet, that makes the carbohydrate information we have much less meaningful.

Philip Thackray

Stephan Guyenet said...

Hi Thackeray,

Thanks for the idea. I don't believe that's the case however. The root vegetables the Kitavans are eating fall into category #1. They have already been characterized for glycemic index and glycemic load, and they aren't that different from grains and potatoes.

Kitavan root vegetables are not primitive; they have been selected by pacific islanders and others for thousands of years precisely for their abundance of digestible carbohydrate.

I believe the health of their diet is not a matter of grams of carbohydrate, or even how that carbohydrate digests. It has to do with avoiding (what I believe are) the long-term metabolic effects of wheat and possibly also refined sugar and industrial foods.

Thackray said...


I looked back at your first post about the Kitavans and saw that the tuber portion of their diet consisted of “yam, sweet potato, taro and cassava”. If, as you say, these are similar to our modern tubers then this is mostly digestible (small intestine) rather than fermentable (large intestine). As Roseanna Roseannadanna would say “never mind” (you may be too young to remember that!).



Thackray said...

Sorry, "never mind" was not Gilda's Roseanna Roseannadanna it was her Emily Litella.


Unknown said...

Kitavans are neither dying of CVD nor cancer. What is the main cause of death among Kitavans? Infectious diseases?

Stephan Guyenet said...


That's OK, it was a good idea.


If I remember correctly, they are mostly dying of infectious disease and "senescence" ("old age"). That's a vague term that I believe implies a general decline of organ systems due to age.

There's no mention of homicide in Lindeberg's papers, but many other Pacific island cultures were fond of spearing one another as well.

Debs said...

I wonder if even root vegetables could be damaging carbohydrates if they weren't in a diet that included saturated fat. Including natural saturated fat seems to be another typical pattern in healthy diets.

To Phil: Speaking of root vegetables... "If you make Puerto Rico a steak, the next thing they'll want is a baked potato!"
(I'm too young to remember Gilda Radner on SNL, but there are videos, thankfully.)

Food Is Love

JohnN said...

@ Thackray:
Good point about the two categories of carb. I think the tubers have both components starch and fiber. Due to the symbiotic relationship between man and its microbiota, it's possible that the tubers (both starch and fiber) are first consumed by the gut flora and short chain fatty acid is then released into the blood stream for nourishment. This can explain the low insulin level (without high level of physical exertion.)

Would you be able to list the types of fruits available on the island? Tropical fruits without intensive human selection may not contain high level of fructose (as in the West) and could be another source of fiber to be broken down by the gut flora to be used as fat.

Stephan Guyenet said...


I don't know what types of fruit they were eating, but they were a significant source of calories. Given that virtually every fruit has a similar ratio of fructose to glucose (some of it is bound up in sucrose), we can assume that they were getting their fructose.

JohnN said...

Here's the quote from Staffan Lindeberg web site:
"The residents of Kitava lived exclusively on root vegetables (yam, sweet potato, taro, tapioca), fruit (banana, papaya, pineapple, mango, guava, water melon, pumpkin), vegetables, fish and coconuts".
The level of fructose may vary with banana (but not plantain), pineapple (not always sweet) and mango (also eaten green like vegetable) rank near the top. I know these items very well coming from the tropic.
All of the tubers and fruits available on the island contain very high level of fiber and a huge amount of potassium whose contribution to vascular and overall health has been under-reported.

Stephan Guyenet said...

Bananas are one of the sweetest and starchiest fruits out there. Fiber doesn't stop that sugar from being absorbed rapidly. I think the contribution of fiber to health has been overstated. There are a number of carnivorous hunter-gatherer groups that ate virtually no plant foods and were healthy. Intervention trials using high-fiber diets to control disease have been a dismal failure.

The fiber hypothesis was a way to explain why indigenous cultures became so sick after adopting white flour and sugar. No one wanted to admit the obvious: wheat flour and sugar are toxic.

JohnN said...

It sure looks like golf when it should have been tennis we playing.
In the last two post I made 4 points:
1. It's worth exploring the fiber to short chain fatty acid conversion. Such conversion is possible in nature and could explain the low basal insulin and lacks of CHD.
2. The concomitant high consumption of fiber.
3. The amount of fructose in fruits varies greatly unlike what we find in the US which are bred for high sugar.
4. The Kitavan food pyramid reveals a high level of potassium intake. In conjunction with low sodium consumption can be an important factor. Potassium intake from natural foods is also a marker
for other minerals such as magnesium, selenium, etc.
I sort of agree with the first poster (Methuselah) - the 4th paragraph.
I regret that it's my first post in your blog and we have gotten off on the wrong foot.

Stephan Guyenet said...


I'm sorry if I came across as brusque. I just meant to explain that I don't think the fiber theory is the most likely explanation. I welcome your ideas even if I don't agree with them. One of the things I value about the blog is having a forum for discussion. I do change my mind sometimes after a good discussion.

I just looked it up and Kitavans get about 10% of their calories from fruit. I think it's safe to assume that most of that is from sugar, and about half the sugar is fructose.

I don't know enough about the Na/K balance to assess that, but I'm skeptical of the danger of salt, based on the Kuna who eat a high-salt diet and don't develop hypertension or CVD.

Anonymous said...


Just to let you know, I've blogged your whole Kitavan series, integrated with a bit of Inuit, spiced with Kuna.

Scott Kustes and Loren Corian also show up.

Stephan Guyenet said...

Thanks Richard, I like the post. You summed it up well. Maybe I should try summing things up every now and then too instead of just piling on more information all the time!

Heather Twist said...

I highly agree that the carbs are not "enough" to cause glucose issues. There are a lot of healthy Asians eating a whole lot of rice! One thing though: their rice is not iron-enriched, and US rice is. The Kitavans are eating a low-iron diet, and likely a lot of them have parasites too. They don't get iron overload, so they handle glucose and fructose OK. In the US, high ferritin levels go hand in hand with T2 diabetes, and donating blood helps insulin resistance. When ferritin levels are artificially made high because of blood donations (thalassemia) heard disease and blood sugar problems follow.

Also, the fish those islanders eat is not generally the fish we eat. They are little dinky fish, commonly fried very crispy and the bones are eaten too (nice calcium source!). They are not top-of-the-food chain fish that concentrate heavy metals.

I've been experimenting with this myself, eating a Pacific island diet. Lost weight, blood pressure down, glucose doing just fine thank you. All while eating lots of rice, vegies, fruits, and little dinky fish.

Heather Twist said...

I should also mention that the Pacific diet doesn't include wheat or dairy. Both of those foods can do interesting things to gut permeability, I think, which might have something to do with messing up the normal iron-regulation system (which has to do with gut cells storing the iron and then releasing in only when signaled to do so).

Robert said...


Does your island diet include much coconut? Also, what fish could one find in the northest that would have a similar nutritional profile to these "dinky" fish you mentioned? I consume a lot of salmon and sardines but both are fatty, cold-water species.



Heather Twist said...

@auggiedoggy: I do eat a fair amount of coconut oil ... it's our standard oil now. I get the expeller-pressed in 5-gallon containers from Wilderness Family Naturals, once or twice a year. It costs about the same as olive oil, that way, and is wonderful for cooking.

For fish: we have a number of "Asian" stores in our area, and I just get whatever they have. Some may be cold water fish, but the labels aren't always very clear. Since the store I go to is mainly Vietnamese, probably a lot of them are not cold-water.

Incidentally, these stores seem to usually have a *awesome* vegetable department ... something that gets left out of some of these studies. I don't know what half the things are ... they are just labelled "vegetable" ... but they make our usual grocery stores look rather pathetic. I get leechees, papayas, mangoes, taro, and a zillion kinds of greens. Which I stir-fry with my fish. Garlic is big too: in huge bags, and ginger. Those garlic/ginger/fish sauce dips are amazing, and I think have their own set of health properties.

The rice I get is actually grown in California, but made for "export", so it doesn't have the additives. Anyway, it tastes a lot different.

Robert said...

@Heather: Thank you for the reply. An large Asian food store opened fairly recently in my area. I'll have to check it out.


Kaigabu Kamnanaya said...

The common food that the Kitava islanders eat to stay alive for the next say are tuberous such as:
Yams (teitu, kuvi, kasiwena, kwanadu, gorina(sweet potato), bisia, taro (uni and viya), including wild and domesticated greens like unonu,gayasu, yakumalasi,beans,nuweta, yakirodi, including fruit from pawpaw, pumpkin, water melon, lemon (kamokuku), natu, yaku, mango, breadfruit, gwadira, sasana, seida (okari-nuts(, coconuts, plus with protein not always (daily basis) such as fish (main one), pork (pig -- rarely), chicken --rarely), crabs and shell fish.

Thank you

Kaigabu Kamnanaya
(Kitava Island)

miche said...

Hi Stephan,

I enjoyed reading your post. Very interesting. I had a question regarding wheat. Do you think all wheat should be avoided including sprouted wheat? I've been eating more healthy and sprouted wheat toast with coconut oil in the morning is my breakfast. Are the gluten-free breads any better? I know most have added sugar. Thoughts? Thanks!

Unknown said...

It could very well be that fact that their food is natural... no GMO and their food is not heavily processed. The soil is most likely natural... with out chemicals. And the air they breath is not polluted.