So you think you know diet?

For every 100,000 inhabitants, Okinawa has 68 centenarians – more than three times the numbers found in US populations of the same size. Even by the standards of Japan, Okinawans are remarkable, with a 40% greater chance of living to 100 than other Japanese people.

Little wonder scientists have spent decades trying to uncover the secrets of the Okinawans’ longevity – in both their genes and their lifestyle. And one of the most exciting factors to have recently caught the scientists’ attention is the peculiarly high ratio of carbohydrates to protein in the Okinawan diet – with a particular abundance of sweet potato as the source of most of their calories.

People, I am not trying to claim this is true!  As is so often the case, I am trying to confuse you and persuade you that maybe you know less than you think.  Here is the full story.  Here are other pieces on why Okinawans live so long, none to me very convincing.


'People, I am not trying to claim this is true!'

Just presenting it, without making claims of whether it has any validity at all.

'As is so often the case, I am trying to confuse you and persuade you that maybe you know less than you think.'

By pointing to something you don't even know is true or not? That is not confusing, that is typical here.

All Tyler is doing is pointing out that uncertainty exists. This is a very important point, because we tend to be very bad at adequately accounting for uncertainty in our decision making.

There is A LOT of uncertainty in what we know about what constitutes a healthy diet. But that doesn't seem to feed into much decision making, whether it's done by policy makers, healthcare professionals, people pushing fad diets, etc.

Strangely, the reply seems to have not met certain standards - guess that the Okinawa diet book from 2005 was too faddish for this comment section.

Or else it would suggest that there is absolutely nothing new about this apparently contrarian posting.

dan1111 is right

It is jumping to a conclusion to claim it is because of diet. It is far more likely to be partly because of good genes and mostly because of the reverence of older people and how they are cared for.

Think about this: IF it were diet and sense they all eat the same diet why then don't they all live to be 100 years old? Obviously there are other factors at work here.

It is true that researchers are looking into diet composition in regards to longevity.

>All Tyler is doing is pointing out that uncertainty exists.

OK, but that's a very dumb point to be making.

>This is a very important point

No, it isn't. See above.

For every glaringly obvious fact, one can find a dissenter on the internet. For every plainly insane conspiracy theory, one can find a proponent on the internet. That Tyler tracks down these people and present their findings without comment says quite a bit about him, but exactly zero about whatever issue is at hand.

LOL, yeah, the Okinawan diet is a "conspiracy theory". Yes humans can exist on a wide variety of diets, but the meat-based high-protein diet is a fad. Real nutrition researchers know that most good diets for longevity are heavy on carbs and low on protein, especially animal protein. Animal proteins are cancer promoting. It's also well known that sweet potatoes are virtually the best food for humans to consume, with incredible nutrient density.

ridiculous comment when global studies seem to point to 15-20% protein intake with higher amounts of protein recommended for elderly. To many focus on what to eat when there seems to be common view of what not to eat: sugar, starch, processed foods, seed oils. Sugar consumption is dramatically up since 1970. we now eat about 57 lbs. of sugar vs then and in 1700 sugar consumption per person was around 2-4 pounds per year. Grains and bread show similar increases in consumption since 1970. Average male today weights in at 200lbs and waist of 40 inches. That is not due to excessive protein intake:

You are a silly fellow indeed. How do you account for the carnivorous animals? The diet of Paleolithic homo sapiens? Our canine teeth? Totally silly, but amusing.

You know the people of Papua New Guinea historically subsisted largely on sweet potatoes and they were heah hunting cannibals. Following your line of "thinking", sweet potato eaters are cannibals.

Thanks for all the amusement your posts provide!!!! 😂🤣

Coincidentally, we were at Costco picking up dog food, and I did notice that they were all advertising their sweet potato content.

For dogs.

Trends and fads. Fads and trends.

Fallon's article might have added some data based "common sense" to this series on the dietary reasons for Okinawa islanders' longevity.

Of course it is completely unacceptable that an Okinawan diet works for Okinawans and a Mediterranean diet works for Mediterraneans. I demand that science comes up with one simple solution that works for me. The world isn't complex; that's scientists trying to make themselves more important than they are.

It would be weird if Okinawans and Mediterraneans, who are virtually identical from a biological perspective, had optimal diets that were completely different. That is almost certainly not what is going on here.

Rather, findings that are supposedly about diet probably aren't about diet.

Kind of weird like lactose intolerance varies between different populations?

Touche! Mediterraneans eat a lot of cheese. I wonder how Okinawans would fare on that diet.

Probably fine, because 'Mediterraneans' are generally on the lactose intolerant side of that divide (and lot of cheese might be a bit of an overstatement), and most cheeses (certainly not all) have minimal to no lactose. And it is also true that the amount eaten plays a role - a bit of feta on a salad is unlikely to cause problems for the lactose intolerant.

"Probably fine, because 'Mediterraneans' are generally on the lactose intolerant side of that divide"

No, we are not.

Guess it depends on how you define 'Mediterraneans' doesn't it? Which in a comment section often obsessed with race, is an absurd formulation to describe various different groups that live around the Mediterranean.

Here is some research regarding Italians - 'A study of regional differences in prevalence of primary adult lactose malabsorption in Italy was conducted on 205 subjects. Their origin was determined by their grandparents' birthplace, 89 from northern, 65 from central, and 51 from southern areas of Italy. Lactose malabsorption was diagnosed with standard oral lactose tolerance test and blood glucose determinations. Lactose malabsorbers showed symptoms more frequently than absorbers after the test load of lactose (p less than 0.01) they also reported milk intolerance more frequently (p less than 0.01). Prevalence of lactose malabsorption is significantly lower in the central sample (19%) than in the northern (52%) and southern (41%) samples (p less than 0.01). This finding contrasts with the hypothesis of a continuous increase in frequency of lactose malabsorption from northern to southern Europe and is probably due to the complex genetic history of the Italian population.'

Prior is a troll who lies simply to muddy the discussion. In this instance, the study he cites flatly contradicts his point (it provides evidence that lactose intolerance does not simply increase add one moves south through Europe).

Prior's data neither supports nor disproves a thesis that lactose intolerance is a strong function of being southern European.

However, a study published in Lancet shows clearly that there is a trend that southern Europeans are more lactose intolerant than norther Europeans, however with Spain at 24% and Greece at 55%, "Mediteranian" is indeed a noisy indicator.

What the study clearly shows, is that the Eastern Europeans have higher incidence of lactose intolerance, to some extent explaining Russian popularity of kefir and yoghurt.

Among the Nordics, Norwegians (12%) and Finns (19%) have higher lactose intolerance than Swedes (7%) and Danes (4%). Lastly, supporting the East/West gradient, also the Irish has only 4% lactose intolerance. The higher lactose intolerance of Norwegians and Finns could be caused by admixture of Laplanders, just as tatar and central asian admixture in Russians explain their higher lactose intolerance.

'Prior's data neither supports nor disproves a thesis that lactose intolerance is a strong function of being southern European'

Actually, it was meant to show that the concept of 'Mediteranians' is laughable, as you then more politely point out.

So, maybe a cache problem again?

'Prior's data neither supports nor disproves a thesis that lactose intolerance is a strong function of being southern European'

Actually, it was meant to show that the concept of 'Mediteranians' is laughable, as you then more politely point out.

The variation is essentially after a certain age, as basically all humans are not lactose intolerant at birth, nor in the first several years of their life.

A truly brilliant observation.

findings that are supposedly about diet probably aren't about diet.

This too. Low-stress lifestyles, which are helped by having closer-knit social connections. I bet they have low rates of depression and anxiety disorders.

They need more capitalism in their lives. Socialism is all fun and games until it's time to pay the bill.

Socialism is about pretending that 300 million other people who you associate with only via a federal government bureaucracy is a "community".

Maybe, but are the genes of Okinawans that different from the genes of mainland Japanese?

~1/4 have v.sig Ryukyuan ancestory. The Japanese government apparently doesn’t consider them a properly seperate group, though I’d guess this a political decision rather than scientific. They’re more similar to the almost extinct Ainu. Different enough to have some sort of impact.

I'm guessing that their genes differ from the mainland more than their diet does.

Really? You think mainland Japanese get most of their calories from sweet potatoes?

a) genetics is important. optimum diet for longevity is unlikely to be constant between populations. plus potential longevity unlikely to be constant.
b) not uncommon for the extreme elderly to turn out to be a pension scam. whereby a child will collect two pensions. As the official age increases the greater the probability of it just being a scam. The numbers should be adjusted Accordiingly

Glazer's Law? There was a story several years back about a discovery that a large percentage of Japanese centenarians had been dead for years, with their families pretending they were alive in order to continue collecting their retirement checks. Might a similar story be behind the fabled Okinawan longevity?

There are major concerns when comparing life expectancy/longevity across an exposure. There are simply too many things that could differ between the two groups which could contribute to mortality between the two groups. Sure they are both Japanese, but I would find it hard to believe that homicide rates, in/out migration, environmental exposures, and rates of smoking and alcohol consumption don't differ between these two groups. At a minimum, one should really explore cause-specific mortality they believed to be associated with diet, and still control for potential confounders.

In addition to the obvious concern over confounding and association/causation, there is an issue of consistency of exposure in epidemiological studies of nutrition. Suppose that there is no confounding between Okinawans and non-Okinawans, and that on average Okinawa's diet is composed of 60% sweet potato calories while non-Okinawa consumes only 20% sweet potato calories. While it could be important that the 40% difference in sweet potato consumption is key, what is missed is what food is present instead of sweet potato among the non-Okinawan.

If we advised the non-Okinawan to eat more sweet potato, most will have to adjust the consumption of non-sweet potato foods accordingly. Some people might eat less red meat, some may eat less yogurt, or less whole milk, fewer eggs, or less fish. Any of those changes could also contribute (or harm) longevity beyond that of the sweet potato.

A clearer, related example might be if we tried to extend the findings to an average American's diet (again assuming no confounding). Given the guidance to eat more sweet potato, suppose American's simply swapped all meat in their diet for sweet potato, but generally ate the same forms of food. They swap the meat on burgers, pizzas, sandwiches, and pasta for sweet potato. However they are still eating these items with the same enriched dough, processed cheese, salt content, and consumption of fruits and vegetables. In this case, I wouldn't necessarily expect American's life expectancy to reach that of Okinawans.

I bring this up because I don't think it is so far from being being practical. When it comes to diets and fads in the (at least in the U.S. in my opinion), people often simply switch components of food rather than change their diets drastically.

It's the feta, the olive oil, the yogurt, the salt, the sweet potato, found in Greek x3, Japanese and Okinawan diets that cause longevity. That and the right genes and good luck.

I will hence make a fad food that's an oily, cheesy, salty kefir-type concoction with a sweet potato base and sell it as health food. "Drink this every day and you'll live to be 100", in quotes, with "No Approved Therapeutic Claims" in bold.

Bonus trivia: the Iranian kefir drink, a sort of fermented milk, is surprisingly good on a hot summer day, the few times I've tried it.

Doogh. Yes, it is.

I'm confused. You are saying the Japanese and the Okinawans eat feta, olive oil, and yogurt in their traditional diets? I think you are joking but I can't tell.

I think most people take a common sense approach that what you eat, as long as you do so in moderation, doesn't make too much of a difference to your life span. A much more important effect is whether or not you are obese - and this is where I think low carb proponents make the argument that actually it is easier to stick in a low carb diet and lose weight than a high carb diet, especially in western society. Not that the actually foods themselves are healthy or not. So this observation it entirely compatible with that - I believe the Okinawan's are low on the obesity scale, probably because although their traditional diet is high in carbs, it is not particularly dense calorie wise and not tailored like western style carbs to be "moreish".

Another thought - isn't this just like p-hacking? Of course, even if absolutely no differences in genes (unlikely of course because of different selection pressures) and no dietary impact some group is randomly going to be highest longevity. It is a bit like isolating successful lottery winners and noting something about that population (probably they eat more carbs than average as well). Doesn't prove anything really.

True, but the flip side of that coin is that study that found a lot of US scratch card lottery winners came from the same geographic region of the USA, then it was discovered that perhaps the computer had made a mistake in generating the pseudo-random winning numbers so as to favor a certain geographic region... So the correlation/causation jury is still out.

Well, it's possible that there's something like exercising and maintaining a healthy weight will help you live to be 80, but if you want to live to 100 that's going to take some extra special refinement of your diet and lifestyle.

A high rate of pension fraud I would say

So what happens to Okinawans who move away and adopt a different diet? Immigrants to the US tend to gain weight...

One can now eat a steady diet of McDonald's (and other fast food places high in protein and fat) in Okinawa, which no doubt will expand the waistlines while reducing longevity. Today's centenarians weren't so unfortunate.

One thing that characterizes our species is our awareness of our own mortality. By our mid-50's, most of us have noted there are clouds on the horizon, and we're being swept ever closer to the edge of the world. Frightening. We don't want to hear it's in our genes. There must be something we can do! Well, sure calorie restriction is more and more established to work, but we want something less difficult, less unpleasant. Exercise? Social Network? Calorie Restriction? Attitude? who knew! And yet, compared to our genes, not all that important. Is there a there there? The population fraction we're talking about is a rounding error. But I suppose desperate times call for desperation...

When I was a kid in the 1970s, the Guiness Book of World Records made much of the high rate of centenarians in Georgia (then in the USSR). But that was disputed as a possible draft evasion move from many years earlier that had carried forward. Does anyone still talk about Georgian longevity?

They are among the longest living groups but the more extreme cases of longevity were just legends.

They eat a plant based diet but also eat "well under 2,000 calories a day."

I will go with a diet that is more pleasurable and shorter lived.

Good nutrition has been more or less settled science for decades. Whole grains, fruits and vegetables, lean proteins, and plant-based fats. Everything else in moderation. Anyone who tells you anything else has at least a book to sell, and usually some celeb-endorsed supplements and possibly a personalized diet plan to sell you. You gotta monetize, after all, and who could monetize two sentences of common sense?

I agree with you completely. But in defense of fad diets, there is decent science behind why they work. Pretty much just picking a food group and cutting it out, regardless of the food group, will lead to a reduction in weight. Satiety is highly context dependent, and it's well known that variety leads to higher food consumption. This is documented as the "buffet effect" in nutrition. Let people eat food A until they're completely full can can't have one more bite. Then put food B down in front of them and they'll eat a surprisingly large amount more.

Therefore it's pretty effective just to pick some arbitrary group of foods and just declare that it's off-limits. Doesn't matter if it's low-carb, low-fat, vegetarian, raw food, low glycemic, gluten-free, local food, whatever. They all pretty much do the same thing. Eliminate a large proportion of the dietary options and people will eat remarkably less.

Tying this back to the Okinawans, their traditional diet is 80% sweet potato. Does their remarkable health have anything to do with the super-nutritional properties of the sweet potato? Probably not. Rather, if you eat nothing but sweet potato day in and day out, you're really not going to be that excited about going back for seconds. Okinawans eat a lot less calories, with all the health benefits that caloric restriction entails, pretty much because their diets are boring and monotonous.

Fair points, especially about the sweet potato. :)

"Good nutrition has been more or less settled science for decades". [SNIP]

40% of the USA population is obese, 5% morbidly so.

One of the above sentences is false.

That makes no sense. It is quite possible for both sentences to be true, and in fact both sentences are true. Unless you're arguing that the same 40% of the US population that is obese is also accurately and consistently following the settled dietary guidelines. I've seen what people eat.

Some nutrition that isn't settled science: Is a calorie a calorie? Do calories of carbohydrates have different metabolic effects than those of protein and fat? What are those metabolic effects? I'd argue that we know very little about nutrition and its effects on chronic inflammation, heart disease, diabetes, and dementia.

"Different metabolic effects?" Of course. What an incredibly general statement. There is far less there than you think there is.

Fair enough. Does eating red meat, full fat dairy, and eggs increase cardiovascular risk? Are there unhealthy synergies between saturated fats , refined grains, and seed oils that are behind our obesity and type 2 diabetes epidemic? Hormonal effects...etc.

I would say, better yet, you persuade us that maybe there is more to know than we think!

The key to longevity: lead a peaceful, temperate life with adequate opportunity for physical labor and social connection, but not so much that you injure yourself or stress out about drama.

If that sounds horrendously boring, it is. But that's how you survive.

I'm going to guess that seafood is really the common factor. It seems like all of the long-lived people live on islands where there is an abundance of fresh fish and the people eat lots of kinds of seafood - fish, shellfish and sea vegetables.

Of course islands are also where predators are less likely - less likely preditation results in evolution tending to select for longer lives and lower reproduction rates (if the soma is not at great risk, worth investing more in the survival of the soma). So maybe just a genetic effect.

It has literally nothing to do with their diet. It is 100% genetics, the same way it is genetics in that individual you know you is in his 50s, looks in his 30s, and never gains weight, no matter what they eat or how much.

Appropriate username

I got my 23andme back. Zero percent native American but more Neanderthal genes than 70 of samples. So I guess I should try paleo!

"70%" of course

Why would you have your dna published where the cops could read it?

Also when do you think one-night-stand children stop looking for their birth fathers? Asking for a friend...

Not sure if matters anymore. If one of your relatives has had it done, they can find you.

That's interesting, as it relates to the whole privacy and control thing in the more recent blog post.

Guess what, I said yes to all questions because I don't really care.

This is very misleading. Their diet is composed primarily of carbohydrates, but carbohydrates that have a glycemic index in the 30s. This is on par with beans, which, fun fact, are a staple in every other region with a high concentration of centenarians found so far. Contrast this with staple carbohydrates in the West, which have glycemic indices in the 60s or 70s

Seafood, beans, and greens (seaweed included).

People, I am not trying to claim this is true! As is so often the case, I am trying to confuse you and persuade you that maybe you know less than you think.

If it were false, why would it persuade us that we know less than we think? So of course, you ARE trying to claim it is true; you just don't care if the things you claim are true actually are true. Because, as you said, you are trying to confuse us. Confused people are easier for you to con.

Does anyone think they know diet? Perhaps in the narrow sense. There are any number of diets that work in a controlled setting. But when diet goes from there to the real world it meets people with varied commitment and skill. People who drink (or give their kids) apple juice, because its healthy. Or the weird curse of diet soda.

I'm not sure I care what's going on in Okinawa, because it is so far from actionable advice you can give to an average American.

For what it's worth, when I am being high commitment I use a calorie counter (+food, -exercise) app. It helps me eat more intentionally, less junk. But we now know from RCTs that giving those things to people doesn't in itself create the commitment.

A better question for the 21st is how to create the commitment to fitness.

I havn't seen much discussion here about intermittent fasting yet.

Supposedly fasting initiates a process called autophagy which results in the body disposing of damaged cells, then activating stem cells which replace them with healthy ones. It supposedly mimicks some of the effects of calorie restriction.

Some of the intermittent fasting diets out there strike me as foolish though. I doubt that skipping meals does much other than screw up your blood glucose. You probably need to fast for a few days in a row to actually initiate autophagy.

I have no idea what a "fasting diet" should look like, but it seems natural enough. The body reserves energy for when food is scarce. Though possibly fasting reinforces the cycle, the body saving for the next fast.

I have never been that motivated. And I probably consider it unnecessary. If you burn 500-1000 calories a day in exercise it's really easy to stay within a reasonable "net calories" for the day. Lazy days are harder and maybe that is the key.

Intermittent fasting doesn't seem to work other than as a side-effect of reducing overall caloric consumption. Extended (3-5 day fasting) does have benefits.

1.) For Intermittent Fasting, "fasting" doesn't mean don't eat anything at all, it means a day where you are limited to 25% of your normal daily caloric intake
2.) While there currently appears to be health benefits to IFD, it's not clear if its due to the fasting per se, or whether it's just a way of reducing total caloric intake that's easier to stick to over time.

"As is so often the case, I am trying to confuse you and persuade you that maybe you know less than you think. "

Tyler., I'm always somewhat confused and don't know much. So you're going to have to lower the bar. ;)

Read Doug's and ChrisA's comments out to husband. "I don't know how much it helps when the comments are smart if the articles are dumb."

TC is enjoying laying his little traps for us, as evidently the one yesterday with the NPR poll (well, since I didn't click the link how the hell was I to know it was a poll of NPR listeners - what kind of poll is that? And why would NPR-listening college-town Latinos approve Trump so markedly?) - but I didn't need this lesson. When you're a woman of a certain age, you've become used to most of your friends announcing they're on an elimination diet, that seems finally aimed principally at making sure no calories are ingested: no bread, no rice, no grains at all, no tomatoes, no nuts, no dairy, no eggs.

My one little trick for having gained no more than 10 pounds since I was a teenager: the Prepare-All-Your-Own Food-and-Spend-No-More-Than-$4-on-a-Taco-When-You-Don't. Not practical for most of you, probably, but the thing is, when you cook your own meals, you converge upon a balance between calories and effort. In tandem with the sweet-potato-eaters of the article, it has much less to do with what the food is, than the amount. You can, if you so choose, eat waffles for dinner.

I wish you would comment more. You lead the way here in common sense.

I don't know if I agree about eating waffles for dinner, but I wholeheartedly endorse cooking your own food, with the clarification that by "cooking your own food" that means actually cooking it more or less from scratch, not popping some toaster waffles in the toaster oven. If you have too actually cooking the amount of effort involved tends to vary linearly with how "processed" the food is.

On weeknights, for me, it's easier to pop a salmon fillet in the oven and make some frozen vegetable mix (brocolli, cauliflour and carrots) and a salad than it would be to make (say) hamburgers. French fries are pretty much impossible to even make at home if you don't have a deep fryer. Anything with breading is a giant pain in the ass.

The danger of this approach is of course that you'll cave in and indulge in a frozen pizza once a week or so, because that is easier than cooking, and end up eating toaster waffles for breakfast.

actually cooking = "actually do your own cooking" or "actually cook". Take you pick.

@Peri - Thanks for the shoutout. Of course Tyler is just providing a theme for us to discuss around so we shouldn't be too hard on him. Remember it's all for free!

On the cooking (from scratch) for yourself as a way to control weight - I wholeheartedly agree. You use much less sugar and fats when you do, plus the food you make yourself will generally contain more fibre. You have to like cooking though (which I do) or have spouse who does. TBH I can't avoid thinking negative thoughts about people who are obese but eat out or processed food a lot and/or don't exercise. It is hard to think these kind of people have good qualities as workmates or friends. Very snobbish of me of course.

There's a big push by the bien pensants for low meat diets and high plant based at the moment:

I'd guess it's all pretty much motivated reasoning by poseurs. Just as the "China Study" always was.

People who tell us to eat less meat, and that it should be astronomically more efficient and cheaper to produce (far fewer energy inputs to generate the product), but somehow this never seems to show up in your bag of "vegetarian substitute" products from the supermarket.

More interesting is the general resistance of our bodies and robusticity to variation in dietary sources -

Calorie restriction is probably a better bet than specific macronutrient shifts, if there is anything to dietary restriction at all. Or anything to dietary advice at all, beyond the avoidance of sugar and staying within your 2500 calories (male) a day.

This is the least Straussian post in quite some time.

I am working on a browser plug-in that replaces all posts like this with "Lower Your Priors. - end of message"

Interesting and probably untrue; the slogan of the intellectual venture capitalist, bis. Who really needs the personal journal of a self-confessed fog-generator any more?

This BBC article seems to be not only out of date but contradicts slightly the well referenced article on Wikipedia of the same subject:

This only: The Fifth Law of Behavioral Genetics.

Otherwise called Finding 4 from Plomin's "Top 10 Replicated Findings from Behavioral Genetics."

Living to 100 years on half the toilet paper, living most of their lives constipated from all those sweet potatoes. Win-win.

Comments for this post are closed