Search This Site!
Nutrition Past and Future

9 The Journalist Gary Taubes 9: Anomaly Hunter 3

Gary Taubes hatched another anomaly-packed sentence on page 25 of Good Calories, Bad Calories, which you see in the lower half of your screen. With these examples, he argues that Ancel Keys and the other diet-heart pioneers demonstrated a lack of familiarity with basic concepts of logic and fairness. They only saw what they wanted to see in the science of their day. Anything that didn’t conform to their views they just “explained away or rejected,” as Taubes put it at the very bottom of this slide. Question, please: What does it mean here to explain away something? It seems like the word “away” is an extra and useless word. Why isn’t it enough to say simply that Keys explained the findings from these studies? Does Taubes mean to say that he didn’t like those explanations? Maybe that’s it. I wonder why. He certainly doesn’t explain his specific objections to what Keys had to say. If Keys didn’t explain an old paper or two adequately, we can certainly do that today without his help. I’m about to show you why this sentence in the second paragraph is just another batch of superficial anomalies that don’t touch diet-heart. I’ll tell you all you need to know about them. Taubes will probably say I am just explaining them away though.

First, I would like to point out that one of his listed groups here is the African nomads. I have already made two videos about the Masai. Please watch those for more information than what I will offer here. I have also created two videos in this playlist called Ancestral Cholesterol that are quite pertinent to the African nomads. I have so much information on this topic that I needed to create separate videos to cover it all properly, and I avoid recycling my research out of respect for my viewers.

The first exceptional case Taubes mentions here is the Navajo.

Here is the study we are to consider. It tells us that the Navajo had a low occurrence of coronary disease. They also had low cholesterol. That’s all we need to know about the Navajo in this discussion as far as I’m concerned. Taubes probably finds this one interesting because the diets of these Navajo were not said to be very different from the diets of non-Navajo control subjects in Cleveland. The implication is made that their fatty diets didn’t give them the high cholesterol that the researchers would have expected to see. They chalked that up to heredity. Maybe the Navajo naturally tended to have lower cholesterol regardless of diet.

Look at this table and you can see we are dealing with only a very few Navajo here. It is clear that their cholesterol levels were indeed on the low side. Once again, I must ask: Mr Taubes, are you saying that saturated fat does not raise cholesterol? Is that your point here?

If you think it does not, then why did you write that it does?

Now I would love to show you a weakness in the methodology of this study such as how they only collected diet information through a 24-hour recall, or something along those lines. But this paper had an even bigger flaw than that. There is absolutely no methodology listed whatsoever regarding the ascertainment of diet information. We are only given this vague assertion of their dietary practices here, which you see has no references to support it. I would think fake anomalies generally propagate better in the absence of data, and there was no diet data here at all. How does Taubes find studies that are this pointless? In that way he actually is a thorough researcher, isn’t he? No one scrapes the bottom of the barrel like Gary Taubes.

Recall that we are talking about a very few people in this study. We have no assurance that these 36 individuals are representative of the larger Navajo population, and we have no information about what they actually ate. All we are left with is a chart showing us 36 people with low cholesterol. That’s it. Mr Taubes, what are we to make of this? Would you care to tell us why it is appropriate to look at this study as a damaging blow to the legacy of Ancel Keys? Think about it.

TAUBES: You do not over-interpret your data. You do not say your studies say something they do not say. And to do that is bad science.

Actually this is bad journalism, too. It is unbelievable to me that this man is going around accusing others of being captive to confirmation bias.

I looked around for other papers on the Navajo to get closer to the real story. Here is another reference reporting low cholesterol and low rates of heart disease.

You see this paper referenced the study which Taubes found only to say that the Navajo had low cholesterol. There isn’t very solid information here, either.

I looked into Navajo dietary practices at that time. There were no consistent findings on this. Their intake of meat was irregular even for those families raising their own animals. This paper makes reference to Many Farms, a small reservation in Arizona. I’ll come back to Many Farms in a moment.

This paper found that the Navajo had generally good risk factors beyond cholesterol. They had low blood pressure. They were less obese. They got plenty of sleep. Nevertheless, at autopsy, they were still found to have some arterial plaque.

Although the study that Taubes likes was run at Fort Defiance, AZ, Many Farms is of interest to me because this paper about the medical research conducted there explains a lot about the circumstances of the Navajo back then. I offer this excerpt to show you that the Navajo were receiving inadequate medical care and they were suffering from terrible poverty before and after World War II. Frankly, they probably had a low prevalence of heart disease back then because they were so poor. Animal foods would have been very expensive for them back then.

Ancel Keys was aware of the study of the Navajo that Taubes referenced and he did dissect it, or explain it away if you prefer. He was incredulous at the authors’ easy assumption that the Navajo ate the same way as most other Americans. How could they have said this? Remember, they had no reference for that claim.

According to the best information available, Keys said their diet was not likely to have been as full of saturated fat as was depicted in that paper. They consumed much less meat and fat than was common in the US back then. They also did not seem to be especially unusual in their rates of heart disease. Keys also mentioned their poverty as a complicating factor. Now ask yourself, is this Keys explaining that paper or explaining it away? Would Taubes have preferred that Keys use no real data on diet just like the authors of that paper? Would that have been better science in his learned opinion? Once again, we look at a Taubes reference and find nothing there.

Next on this list is a comparison of Irish brothers separated by an ocean. Some Irishmen were in Boston. Other Irishmen were in Ireland.

This is the paper Taubes selected. It is a very short paper. The brothers in Boston were found to have a much higher death rate from coronary artery disease than their brothers in Ireland. The authors tried to understand why.

Taubes wants us to take note of this passage, I’m sure. The Boston men had a lower proportion of animal fat in their diets than did the Irish, yet they suffered more heart disease. But they also ate more animal protein. They also had similar ratios of saturated to polyunsaturated fats in their diets. Can you tell that we are once again dealing with a flimsy piece of evidence which Taubes has dug up just to confuse us? How does he find this stuff?

The Boston Irish were significantly fatter than their counterparts. They had higher blood pressure.

Notice, they consumed the same amount of sugar. Let’s pause and think about how the journalist Gary Taubes must have seen this, a study showing a large difference in heart disease with no differentiation in sugar consumption. How would John Yudkin have explained this away? Once again, the low carbers are very selective in how they use the old research. We learned that the Bostonians ate more beef, margarine (which was probably partially hydrogenated), and oil. Meanwhile, the brothers in Ireland consumed more starches like cereals and potatoes. Did you catch that, Mr Taubes? More starches, less heart disease? The men in Ireland were more physically active. You’d better believe that made a difference in their health. Here we see that men who ate more starches and were more physically active were lower in weight and had less heart disease. What do you make of that, Mr Taubes?

The Boston brothers had much higher cholesterol. Mr Taubes, do you think this may have been a factor in the differing rates of heart disease between the two groups? How would this study have posed a challenge to Ancel Keys? I’m pretty sure he thought high blood cholesterol was a risk factor for heart disease.

Taubes innocently forgot to mention that that initial little paper was followed up by a much larger project that followed both sets of brothers over 20 years. A 20-year longitudinal study should greatly overshadow the paper Taubes liked, which was nothing more than a little snapshot in time. Here you see that those who died from heart disease were more likely to have consumed less carbohydrate, less starch, and less vegetable protein. Those who died of heart disease tended to consume more dietary cholesterol. This study was designed to surpass the other one, which was no more than a little preliminary statement before the real study got going. Taubes decided you did not need to know about this one. He doesn’t want you to know what dietary habits killed these guys. If you knew, you might stop caring about what he says.

We are now up to the African nomads. I said a few moments ago that I would like you to look at my Masai videos as well as my Ancestral Cholesterol videos for my complete discussion of the African nomads referenced in this sentence. Right now, I will at least touch on the study Taubes had in mind for this statement, but I won’t bring to your attention the most interesting aspect of it. That would steal the thunder of my Ancestral Cholesterol video about the Masai. All in good time.

This is the summary for this study. It was written by George Mann. Mann declared that despite the fatty diet the Masai consumed, which was composed almost entirely of animal products, they had low cholesterol and a lack of heart disease. Consider this for a moment and immediately it is apparent that Taubes is yet again trying to suggest that if you consume a lot of saturated fat your cholesterol will not become elevated…

Even though he doesn’t believe this himself.

Once we read ahead after this sentence a bit, we see that Taubes was going for more than just this one point with this reference. Take note that I will be interchanging the names of similar nomadic cultures since they were quite similar with respect to diet and because Taubes treats them as a single entity as well. Taubes tells us that George Mann was a prominent diet-heart opponent who studied the lipids of the Masai. This experience shaped his beliefs about fat, consequently bringing him into conflict with Ancel Keys, who said that fatty animal foods raise cholesterol. Look at my second block of text. Taubes is relating a response from Keys to Mann about these pastoralists. “Such diets, consumed at a bare subsistence level, would be consistent with the serum cholesterol values achieved.” Taubes then comments on this quote.

“Keys, however, had no reason to assume that either the Samburu or the Rendille were living at a bare subsistence level,” he said. The Samburu and the Rendille are African nomads as well, and Taubes is essentially accusing Keys of having fabricated his claim that they lived at only a subsistence level. It seems to me that Taubes is the one doing the fabricating here.

This is the paper Keys probably had in mind. The Samburu, for most of the year, had a dietary intake that was called “just sufficient” for their nutritional requirements. Keys did not make that up.

The authors stated in this one that during the dry season the amount of milk they drank fell considerably, and milk was one of the most important elements of their diet.

I think it is likely that Taubes has read this study because he referenced it in his book. Did he not notice this line? In a typical year after the wet season, what follows is “…six to seven months of decreasing milk supply and a dry and harsh period of three to four months when milk is in short supply.” That’s pretty clear language yet Taubes apparently didn’t understand it. Ancel Keys isn’t around to defend himself against someone with such poor reading comprehension.

Taubes also later selected a different study about the Samburu to support his claim that they had among the lowest cholesterol ever measured.

If you look at that study, it talks about the curtailment of their food supply due to the dry season and the very hard work that was required of them to survive in their harsh surroundings. The author said “periods of frank famine are not unknown in the Samburu territory.” The phrase “frank famine” does sorta imply a certain lack of calories, does it not, Mr Taubes?

The same study also says that at a mean height of almost five feet nine inches, their men only weighed 126 pounds. Please realize Taubes picked this reference. I didn’t.

Now that you’ve seen all that, and considering that you have reason to believe that Taubes saw all that, too, can you believe that he had the temerity to fault Keys for claiming that the Samburu lived at a bare subsistence level? Once again, Mr Taubes, I’m not asking much.

TAUBES: Just, stay honest, Dean. That’s all I’m asking, ok?

He’s got some nerve. Once again, we see Taubes trying to pass off an old study as supportive of his views that instead reveals his incoherent nutritional concepts as well as his biased reporting. I’ll have more about the barely adequate caloric intake of the African pastoralists in my second Ancestral Cholesterol video.

The Swiss Alpine farmers are next in this sentence.

And this is the paper that supposedly supports Taubes’ ideas. It is reported here that Swiss alpine villagers had low cholesterol levels despite their diets which were high in saturated fat. Once again, what is his point? Need I say more? Near the bottom of this paragraph, the authors stated that the low cholesterol of these Swiss could not be explained by the high altitude at which they lived.

A different comment of theirs about altitude is a bit more accurate. “No data on the effect of altitude on the same groups have been found in the literature.” It may have been true in 1962 that not much was known about altitude and cholesterol but that’s not so true today.

High altitude lowers LDL or bad cholesterol...

and raises HDL or good cholesterol, so high altitude probably does protect against heart disease.

This study found beneficial effects for cholesterol levels attributable to high altitude in India.

A similar observation was made in Venezuela. In both places, high altitude was associated with lower LDL. The authors of the Swiss paper shouldn’t have been so quick to write off the effects of altitude.

I used a couple of the slides you just saw in my old videos about the Masai. These Swiss farmers had more in common with the Masai than just animal foods and high altitudes. Low carbers seem to idealize the lives of historic animal fat eaters. They should realize how hard these people had to work to survive. They didn’t live like modern low carbers. Read through this description of the physical labor that was required of these Swiss. Elderly women were said here to have carried 60 pound loads, for goodness sake!

Because of their hard labors to feed themselves, many of them were considered to be underweight, which is not how today’s low carbers are often described.

Despite that, their cholesterol levels were not especially low, with the men exceeding 200 in middle age. The authors of this paper were impressed by their low-ish cholesterol, but they hadn’t seen my Ancestral Cholesterol videos like I know you will. You’ll see real low cholesterol in those.

Let’s finish off this sentence now by looking at the Benedictine and Trappist monks.

This is the study Taubes picked for them. The lead author of this is J.J. Groen. He wrote that despite the fact that the Trappists consumed a frugal vegetarian diet, there was no difference between them and the Benedictines with respect to various measures of heart disease.

You see that we are once again dealing with isolated cholesterol measurements in this study, so we are once again likely to have statistical noise in our data. If you don’t know what I mean, I’ll explain better in my Measurement Problem video.

The authors admitted that their study was a bit lacking in its design. It had too few participants and it was not conducted longitudinally. Read this and realize that Taubes has picked a study that he says Keys just “explained away” which didn’t even have the full support of its authors.

You can see why they felt this way. In only one Trappist did they have a clear diagnosis of a heart attack. In that one case, cholesterol was extremely high. That’s one person and he had high cholesterol. There were no heart attacks among the Benedictines. Do you think this study had any statistical power to speak of? They did note that the Benedictines, with their higher cholesterol levels, seemed to suffer from angina at an earlier age than the Trappists. We also see that the Trappists who did experience angina also had high cholesterol. Once again, Ancel Keys had nothing to worry about here. But, you may ask, why did any Trappists have high cholesterol? They were vegetarians, right?

Well, just like today, the word “vegetarian” could mean a lot of different things back then. Interestingly, you can see that Taubes was once again selective in his interpretation of data. The Trappists consumed much more buttermilk and cheese. Also, they consumed much less sugar. Remember, we are dealing with two groups of men with what we are told was a low rate of heart disease. Shouldn’t we have seen a difference in heart disease with such a clear difference in sugar intake?

Here are their cholesterol numbers. The Benedictines, who ate more meat, had higher cholesterol and developed angina at a younger age. That isn’t very surprising, is it?

Here is another paper that looked at these studies of monks. These authors noticed that there was a regression toward the mean in their data. They commented on intra-individual variation. They noted that diet was not the only factor which could affect serum cholesterol. That’s not bad considering this was written in 1963.

Here you see their effort to combine the results from two studies of Trappist monks. Notice the line that I have underlined. The Trappists who ate eggs had higher cholesterol. Remember this when you hear people say that eggs won’t affect your cholesterol. I’ll have much more to say on that topic in another video.

Ancel Keys did explain the monk studies adequately in my view. Because the nominally vegetarian Trappists consumed so much butterfat, they still ate plenty of saturated fat.

If Mr Taubes likes J.J. Groen’s research and if he believes saturated fat doesn’t raise your cholesterol, perhaps he should have a look at this paper by Groen. Here, he demonstrated the effect of a diet loaded with meat, cheese, and gravy on cholesterol levels. Naturally, this fatty diet caused cholesterol levels to rise. He noted that his participants had low cholesterol levels on their normal diets despite the fact that they did not consume a lot of polyunsaturated fats. In other words, a low-fat diet might also lower cholesterol. A fatty diet based on vegetable oils wasn’t the only option available. It would have been nice if people had paid more attention to this observation years ago. One need not consume a diet loaded with refined oils to lower one’s cholesterol. Unfortunately, as you will soon see, this was not an observation that informed future diet trials. Researchers got it in their heads that diets had to be high in some kind of fat, and so polyunsaturated fatty acids must be the right tool for lowering cholesterol. I’ll come back to this in my Oil-Based Nutrition videos.

Believe it or not, I have delved into still more flimsy science that Taubes stuffed into Good Calories, Bad Calories. We continue with this in my last anomaly hunting video.