The Arrow #155

Hello friends,

Greetings from Montecito.

Since I won’t be writing this newsletter on the 25th, let me take this opportunity to wish everyone a Merry Christmas in advance from a rainy Montecito, CA. I’ll be back in Dallas next Thursday where, I hope, it won’t be raining.

It’s good to get it here though. Been in a drought for a few months. Three days ago a mangy, malnourished coyote ran across the fairway in front of me. Last time a coyote ran across the same fairway in front of me was about five years ago, and he/she had a full coat and was pretty frisky. This one looked barely alive. When there is no rain, the rabbits, rodents, and squirrels become scarce. And the coyotes go hungry.

My Favorite Christmas Story

I can’t remember when or where I read this story, but it is my all-time favorite Christmas story (second to A Christmas Carol, of course, but that’s more of a book.). I’m pretty sure I posted it in The Arrow last year, and I would guess I’ll probably post it next year as well. I put it up on my blog a few years ago, so here it is. I read it every year. Maybe it will become your favorite Christmas story as well. Enjoy!

Another Kind of Christmas Story

Back in 2008 our family started an appliance business. We developed the first sous vide unit for the home cook. Up till then, sous vide had been the province of upscale restaurants throughout the world. The commercial units at the time cost thousands of dollars—we came up with one costing about $400.

It got the attention of Heston Blumenthal, a famous British chef mostly unknown by the average person in the US, but extremely well known in the chef community as we were to find out.

Heston owns and operates a restaurant in Bray, UK called The Fat Duck that is in a class by itself. He is hugely famous over there, but, as I said, not so much so here. He tried one of our units in his test kitchen to compare it against the 25 or so highly expensive others he uses. Ours did as well as the ones in his experimental kitchen, so he reached out to see if there was any way we could work together.

We ended up going on a North American tour with him extolling the virtues of sous vide cooking and basically introducing our product to the world. We started in Vancouver, Canada, a huge foodie city and ended up in New York. It was a different city almost every day for a little over a week.

Heston, though a major star in the UK, is really a down to earth kind of guy. We were all packed into cars driving for hours, and he never complained about anything. He traveled with one of his top chefs Kyle Connaughton, who was also a terrific guy. Kyle ended up leaving Heston to open his own highly acclaimed restaurant in the Napa Valley not long after our big adventure.

Heston is entirely self-trained. He told us he grew up in middle-class London and had no idea what he was going to do with his life. When he was a teenager, his family traveled to France, where Heston got his first taste of fancy French food. He was hooked. He started studying everything he could about food and ended up with one of the most famous restaurants on the planet.

Just about the best part of the tour was going out to eat after the day’s last media presentation. We went to all these great restaurants where the chefs practically genuflected to Heston. As a consequence, we ended up with many fabulous meals. This one below with Heston and David Chang, founder of Momofuku, in his flagship restaurant in New York. At the time, it was impossible to get a reservation there, but David made room for all ten or so of us, and the food just kept on coming. Here I am with Heston and David.

Here are MD and Heston doing their dog and pony cooking show for the foodie crowd.

Heston with Tim Ferriss and me at the show in San Francisco.

And Heston, MD and me at the end of the whole trip. As you can see, MD was happy it was all over.

All this fun is leading up to a video of one of Heston’s famous Christmas dinners released not long after we did this tour. There are a number of these Christmas dinners along with many other videos of Heston’s cooking on YouTube. I just like this one because MD and I had some of these dishes on our one trip to The Fat Duck, which, to this day, is still the most memorable dining experience I’ve ever had. We dined once at The French Laundry, which was good, but to paraphrase Mark Twain, it weren’t shucks to The Fat Duck. If you ever have the opportunity to dine there, do not pass it up. Even though there will be plenty of carbs. Which are okay occasionally. After all, pleasure is a nutrient.

Here is Heston’s Perfect Christmas feast. Watch it, and you can imagine all the chef stuff we picked up on our tour.

It’s a fairly long video, so if it doesn’t work properly on Substack, click here for an online version. By watching this video, you’ll see how much fun we had and how much we learned.

Warning: it is not low-carb.

Sad News

I just heard from a friend that former Beach Boy and all around good guy Jeff Foskett died recently. My friend—the one who got MD into Willy Nelson’s trailer—told me he had just learned Jeff had died. I met Jeff years ago when MD was the president of the Santa Barbara Choral Society. The group put on a huge concert around a Beatles theme in honor of Sir George Martin, the so-called 5th Beatle because he arranged most of their music. Sir George was there and directed the premier of one of his own compositions, called The Mission Chorale. Jeff was scheduled to sing the solos in the Beatles pieces, but due to an unfortunate snafu, it didn’t work out. He would have been great. But at least we got to meet him and hang out with him a bit, for which I’m really glad. A seriously nice guy. RIP

The Bride Down Under

On a more uplifting note, as promised, here is the video of MD giving her talk in Sydney a couple of months ago.

When I put up one of my videos from Sydney last week, I said I would put up hers (which had just become available) this week. What I didn’t realize was that if anyone watched mine, they got fed hers from the side. Many of you wrote to say you had seen it as a consequence. But for those who haven’t, here you go.

A Cholesterol Vaccine

In one of the many medical subscription services I subscribe to was an article about a researcher in the medical school at the University of New Mexico who was developing a vaccine for cholesterol. Yep, if your cholesterol is too high, then you could just get a vaccination for it.

Who would want to do that?

Well, no one with good sense.

Before we get into the why, take a look at this short video about the guy creating the vaccine and a cardiologist (also from the University of New Mexico) who is waiting for the vaccine himself, because he has high cholesterol.

These two are practically quivering with glee over the potential for this vaccine. But they are also non-thinkers unless what they’re thinking about is the money they could make from this vaccine were it ever to come into being.

There is really no mass of data out there proving that LDL cholesterol causes heart disease. It’s a theory. An hypothesis that has not been proven. It’s called the lipid hypothesis in the literature, not the lipid fact. I always like a quote I read in Taubes’s Good Calories, Bad Calories by physicist Pief Panofsky, founder of the Stanford Linear Accelerator Center. Said he, “If you throw money at an effect and it doesn’t get larger, that means it is not real.”

God only knows how much money has been thrown at the lipid hypothesis, and it’s still just an hypothesis. In my view, these guys are stupid to spend the resources and brain power to try to create a vaccine to treat a hypothetical relationship. Why not spend the time and effort working on something real?

And they should know better. Presumably they’ve taken biochemistry somewhere along the way in their training. If they remember any of it, it should give them pause to consider a vaccine that could override the body’s ability to make and transport cholesterol. It is an incredibly important substance—not just some waste product that needs getting rid of.

I went online to look for a brief, snappy video explaining all the good things cholesterol does. I couldn’t find a short one that I was happy with, but I did come across a terrific one that is about 28 minutes long. The video is of a guy drawing out the cholesterol-synthesis pathway in a fair amount of detail. He shows where the critical control points are and how they can be influenced. I know it sounds dreadful, but it’s really well done and keeps your attention.

In discussing all the products that are made of cholesterol, he mentions, but doesn’t emphasize, vitamin D. Vitamin D is made from cholesterol as are a whole lot of other substances we can’t live without.

When you watch the video, pay close attention to where insulin and glucagon work on the various pathways. And notice that one of the starting raw products for cholesterol synthesis is glucose.

It’s a fascinating video, but the guy doing it all is kind of stupid like the guys involved in making the cholesterol vaccine. I’m not talking Hank Johnson level stupidity here, but stupid nonetheless. In the video he stresses all the important compounds made from cholesterol, then when he gets to HMG CoA reductase, the rate-limiting enzyme in the synthesis pathway, he simply states that this is where statins do their job to keep that dangerous cholesterol low and prevent heart disease. Do these people not think?

I know it’s 28 minutes long, but if you have any interest at all in the subject, I encourage you to watch it. Especially if you are being badgered about going on a statin. Or told your cholesterol is too high.

Watch closely when he talks about how insulin and glucagon act on the synthesis pathway. And remember, you can lower insulin and raise glucagon by simply cutting a lot of carbs from your diet.

This video is much like a medical school biochemistry lecture. Everyone who has ever gone to med school has had a lecture like this one. And, if everyone else’s medical school was like mine, they had to memorize every one of these pathways, all the enzymes that catalyze them, and all the intermediate products, then regurgitate them on exams. And then they apparently forget it all and blindly prescribe statins for anyone who walks through the door.

I want to take a sec and focus on one specific part of this talk.

Toward the end, he goes back to discuss HMG CoA reductase, the rate-limiting enzyme in the cholesterol synthesis pathway. The rate-limiting enzyme in any pathway is the one that controls the output. If that enzyme is stimulated, more product is pushed out at the end of the pathway. Conversely, if the rate-limiting enzyme is inhibited, less product is made.

HMG CoA reductase determines the amount of cholesterol the body produces. It is the enzyme statins act on. Statins inhibit HMG CoA reductase, which ends up forcing the body to make less cholesterol.

What else stimulates or inhibits HMG CoA reductase? You can see it here in this blow up of a part of the video.

I’ve put a big red box around it to make it easier to see. That plus sign indicates insulin stimulates HMG CoA reductase while the minus sign means glucagon inhibits it. This is not brand new information. It has been known for years. MD and I wrote about it in Protein Power almost 30 years ago, and it had been known for years then.

It all makes sense, if you think about it. Insulin is a growth hormone. When insulin is up, the body is storing and building. And since cholesterol is a structural lipid, more is needed to fortify cell membranes and to build new tissue. Glucagon is an anti-storage hormone. It releases glucose from the liver and helps move fat out of the fat cells. Under non-storage conditions, glucagon predominates. During times of breakdown, cholesterol will be released, so we don’t need to make as much of it. That’s how the insulin-glucagon axis regulates the increase or decrease in the production of cholesterol: by governing the activity level of HMG CoA reductase.

Glucagon basically does the same thing statins do, but without all the side effects. And it’s free.

Okay, so how do we get glucagon levels up and insulin levels down? What organ regulates those two hormones?

You may be tempted to say the pancreas, since the pancreas makes both insulin and glucagon. But the real answer is different. The actual organ that determines the relative amounts of insulin and glucagon you have circulating in your blood is your mouth. (OK, it’s not exactly an organ, but you get my meaning.)

What you put in your mouth determines the levels of these two hormones. Never forget that you have complete, 100 percent, control over what goes in your mouth.

If you chow down on carbs, your insulin will go up, then your HMG CoA reductase will kick into gear and start making more cholesterol. If you avoid the carbs and stick to protein and fat, just the opposite will happen.

Always remember that. You are in control.

And, BTW, if the cholesterol vaccine should ever hit the market, my recommendation would be to avoid it.

Okay, time for the weekly beg. To become a paid subscriber to The Arrow costs a measly 20 cents per day (19.78022 cents, to be exact, and that’s if you sign up monthly). If you sign up for an annual subscription, it’s only 16.438356 cents per day. A bargain to be sure. I just saw a stat that the average cost of a quart of eggnog is $4.53, which is about 22 day’s worth of The Arrow. And The Arrow won’t raise your insulin. And it might make you just as jolly.

Ultra-Processed Foods (UPF)…Again

As of late, I’ve become obsessed with ultra-processed foods. Mainly because in my own brain I define them differently than most everyone else does.

To me, processing means, well, processing.

Take wheat, for example. You roast it, then grind it. If you grind it coarsely, the flour is less processed than if you grind it extremely fine. An apple is unprocessed. If you purée it, you process it. If you juice it, you process it even more.

The more stuff you do to disrupt the architecture of a natural product, the more processed it is. At least that was my thinking until I started reading more about it. As it turns out, everyone else who is jumping on the UPF bandwagon has a different definition.

To them, apple juice is the same as an apple. And fine flour is the same as course flour.

What then makes an UPF? According to the current writers on the subject, it’s the addition of stuff to the end product. Which, to me, at least, would not be a processing issue, but an additive issue. I would think a better definition by their standard than ultra-processed food would be adulterated food.

I’m reading a book on it now by a UK physician. Here is his definition.

UPF has a long, formal scientific definition, but it can be boiled down to this: if it’s wrapped in plastic and has at least one ingredient that you wouldn’t usually find in a standard home kitchen, it’s UPF. Much of it will be familiar to you as ‘junk food’, but there’s plenty of organic, free-range, ‘ethical’ UPF too, which might be sold as healthy, nutritious, environmentally friendly or useful for weight loss (it’s another rule of thumb that almost every food that comes with a health claim on the packet is a UPF).

The author of the book I’m reading was inspired by a food writer in The Guardian. She led me to a paper by a Brazilian physician, who, as it turns out, is one of the major forces behind the idea that UPF are bad. Here is his definition from a fairly recent (2019) paper:

Ingredients that are characteristic of ultra-processed foods can be divided into food substances of no or rare culinary use and classes of additives whose function is to make the final product palatable or often hyper-palatable (‘cosmetic additives’). Food substances of no or rare culinary use, and used only in the manufacture of ultra-processed foods, include varieties of sugars (fructose, high-fructose corn syrup, ‘fruit juice concentrates’, invert sugar, maltodextrin, dextrose, lactose), modified oils (hydrogenated or interesterified oils) and protein sources (hydrolysed proteins, soya protein isolate, gluten, casein, whey protein and ‘mechanically separated meat’). Cosmetic additives, also used only in the manufacture of ultra-processed foods, are flavours, flavour enhancers, colours, emulsifiers, emulsifying salts, sweeteners, thickeners, and anti-foaming, bulking, carbonating, foaming, gelling and glazing agents. These classes of additives disguise undesirable sensory properties created by ingredients, processes or packaging used in the manufacture of ultra-processed foods, or else give the final product sensory properties especially attractive to see, taste, smell and/or touch.

The entire paper is basically one long definition (the author’s) of what an ultra-processed food is. Here is another short section.

Industrial breads made only from wheat flour, water, salt and yeast are processed foods, while those whose lists of ingredients also include emulsifiers or colours are ultra-processed. Plain steel-cut oats, plain corn flakes and shredded wheat are minimally processed foods, while the same foods are processed when they also contain sugar, and ultra-processed if they also contain flavours or colours.

Generally, the practical way to identify if a product is ultra-processed is to check to see if its list of ingredients contains at least one item characteristic of the ultra-processed food group, which is to say, either food substances never or rarely used in kitchens, or classes of additives whose function is to make the final product palatable or more appealing (‘cosmetic additives’).

So, let me get this straight.

If I eat plain corn flakes in the way most people eat them with whole milk and sugar, I’m eating a minimally processed food? But if I eat corn flakes with a sugar-glaze sprayed on at the factory and whole milk, I’m eating an ultra-processed food?

Sorry, but I don’t see a real distinction here.

I do agree that way too much crap is put in commercially-prepared foods, i.e., factory foods, but I would call that adulteration, not ultra-processing.

But that’s just me.

My definition of an ultra-processed food would be one that would run my insulin up more than the same food would if it were less processed. As I’ve written a few times in this newsletter, it’s the incretin effect—primarily from glucose-dependent insulinotropic peptide (GIP)—that drives insulin so high. For those of you who are new readers, here is a short video of the incretin effect from a part of a talk I gave a few years ago.

The more the integrity of the food is compromised—usually by ever more processing—the greater the incretin effect. And that effect is large.

I wrote a few weeks ago about the difference in GIP and the consequent insulin response to an apple as compared to the same apple puréed and juiced. There are big changes as it is broken into ever smaller pieces.

According to the definitions above, the apple purée would be an unprocessed or minimally processed food as long as it didn’t contain some kind of preservative. Admittedly, I’ve never seen it tested, but I would bet that adding a bit of preservative to the apple purée would not change its incretin response.

Don’t get me wrong. I’m not in favor of adulterating foods. But I’m not sure that adding a bit of flavoring to something makes it ultra-processed.

Kevin Hall, the main proponent of the energy-balance theory of weight gain and loss (calories in vs calories out), felt the same way, so he decided to do an RCT to see if there was anything to the notion that consuming a lot of UPF would cause people to eat more. Which is the hypothesis of all the people who are the main promotors of the idea that UPF are bad.

Hall recruited 20 subjects—10 men and 10 women—to be kept in hospital conditions and fed for 28 days. Each subject would be his/her own control. Each subject would follow a minimally-processed food diet for 14 days, then an ultra-processed food diet for 14 days. During the diets a ton of parameters were to be measured.

Twenty subjects is not a lot. In fact, it would be what you would start with in a preliminary or pilot study to see if you got results strong enough to justify a larger experiment. But it is extremely expensive to do a study in which people are housed 24 hours a day and provided with all their food.

Here is how the experiment worked. The subjects were told it wasn’t a weight-loss study, which it wasn’t. And they were instructed to wear loose clothing, so in case they did gain or lose some weight, it wouldn’t be noticeable.

The staff came up with diets that were similar in macronutrients but dissimilar in the amount of food additives. The subjects were given breakfast, lunch, and dinner and were allowed an hour to consume each meal. And they were provided with unlimited snacks available throughout the day. All they had to do was request them. The study staff collected and analyzed any leftovers from any of the meals, so only the foods actually consumed were measured.

After all the data were tabulated, it turned out that subjects consumed a whopping 500 Calories more per day while on the UPF part of the experiment than they did while on the so-called minimally-processed (fewer additives) part. And while on the UPF diet, the subjects gained significantly more weight than they did while on the minimally-processed diet.

You can read the paper on this experiment here. Also, if you scroll down to the supplemental materials section, you can download a pdf with descriptions and photos of the meals involved.

Here are a few examples.

First, let’s look at the snacks, which are vastly different between the two diets. The left from the UPF phase of the diet; the right from the minimally-processed phase.

As I wrote, a big difference.

Now let’s look at breakfasts from both. UPF on left. Minimally-processed on R

You’ll notice in these meals that NutriSource fiber has been added to the UPF meals. Some believe that the reduction in fiber as a consequence of processing causes many of the issues with UPF, so the researchers added fiber back in, so the fiber would not be a confounder.

Here are the lunches. UPF on left, minimally processed on right.

And here are the dinners.

Unfortunately, when Substack does a gallery view, it truncates the images. I’ll stick the image of the minimally processed ingredients below.

UPF on left; minimally processed on right.

Here is the full menu for the minimally-processed dinner.

Here is a test question for you. Where do you think the extra 500 Calories those in the UPF phase consumed? Snacks? Breakfast? Lunch? Dinner? A combo of all?

Go ahead and guess. I can tell you I was completely incorrect when I guessed.

As it turned out, the vast majority of the extra Calories did not come from the snacks, which was my guess. They came from both breakfast and lunch, more so lunch. Not dinner and snacks.

Where did all the excess calories come from? A little over half came from carbs and the rest came from fat.

The increased energy intake during the ultra-processed diet resulted from consuming greater quantities of carbohydrate (280 ± 54 kcal/day; p < 0.0001) and fat (230 ± 53 kcal/day; p = 0.0004), but not protein (2 ± 12 kcal/day; p = 0.85)

You can see this graphically below.

The fact that protein is almost exactly equal between the two diets gives some credence to the protein-leverage hypothesis, which posits that people will eat to achieve a particular protein intake. If the protein content of the meal(s) is low, people will keep eating until they’ve gotten enough protein even though their total calorie intake is high. In other words, our food drive keys on protein, which makes sense as protein is the only essential macronutrient. But, as far, as I can tell, there have been no well-executed human studies showing the protein-leverage hypothesis to be true.

What this tells me is that the diets were not as balanced as they were supposed to be in terms of calories and macronutrients. Or that the subjects picked and chose on their breakfast and lunch offerings.

Now if the fat content of the UPF diets was mainly PUFA, I could buy that. I’ve given an entire talk on why an increase in PUFA might drive overconsumption and weight gain. But, again, I’m not sure I would call the addition of PUFA an ultra-processed diet. Seems to me that the processing level is the same. There is just some PUFA added to the mix.

Looking at the insulin and glucose curves is a bit telling.

These tests were done at the end of each dietary period, not after each meal. But we can see from the glucose tolerance test at the top that it looks like the glucose levels in the UPF group are crossing the line at the 180 minute mark. Which would make sense since the insulin graph at the bottom shows an increase in insulin starting at 90 mins and continuing. This, one would assume, would cause a decrease in blood sugar beyond the 180 minute mark.

If so, then glucose would be dropping at about lunchtime. And a falling glucose level is a strong stimulus to eat. Maybe that’s why they ate so much more at lunch.

For whatever reason, there was clearly a difference in weight change and fat mass change between the two groups.

I have a few issues with this study. First, it would have been nice to have had more subjects. I know it’s expensive, but now with this data, maybe the NIH would be willing to fund a larger study. Also, there are too many confounders in the substances added to the UPF foods, mainly PUFA. And, finally, there should have been a washout period between the two diets, which there wasn’t. Hall et al is aware of that and wrote about it.

Our study did not include a weight-maintenance run-in period or a washout period between test diets. These design choices were made to lessen the burden to the subjects and reduce the likelihood of dropouts, which was successful because all 20 subjects who successfully screened for the study also completed. To partially address the lack of run-in or washout periods, we compared ad libitum energy intake during the final week of each test diet period and the substantial diet differences persisted. The lack of a run-in period complicates the interpretation of the baseline blood measures in comparison to those obtained at the end of each test diet, and all such diet comparisons were potentially confounded by the substantial differences in energy intake and corresponding weight changes.

I look forward to someone doing a longer experiment with more subjects with washout periods between the two interventions. I do commend Hall et al for doing this study at all.

Before we leave this section, I want to comment on more of a long-term, ancient, observational study on food processing.

The ancient Egyptians consumed a diet based on grain. They used emmer wheat to make bread, which was their primary form of subsistence. The bread was coarse-ground, whole-wheat bread. Grinding the wheat was a major occupation in those days. Once, ten years or so ago, I was left to my own devices in Paris (MD and her sister wanted to go to Versailles—I didn’t want to go anywhere near the place, having been twice or more before), so I wandered over to the Louvre where there was an exhibit of Egyptian antiquities.

There were hundreds of little statues of people grinding wheat. Here are a couple of examples.

There were countless little pieces of statuary like these. It just went on and on.

When the Egyptians ground the wheat, they put a little sand in the mix to help move things along. Then tried to get as much sand out as possible. The bread made from this wheat was not the highly-processed bread we get today. Not even close. Theirs was real stone-ground, whole-wheat bread.

On YouTube, you can find multiple talks I’ve given on this subject over the years. Part of the data I use in the talks comes from the anthropological and archeological and paleopathological literature, but the description of the diet of the ancient Egyptians has come from more popular books written for the layman.

A couple of weeks ago, I got a line on a book about the lifestyles of ancient Egyptians in a working class town near Thebes. A huge trove of written material from the area had been unearthed and translated. The contents of these papyri and ostraca, flakes of limestone used as writing surfaces, included love letters, letters asking for advice, letters demanding payment, letters asking for apologies, statements of payment, and all kinds of other back and forth contemporary to about 3,000 years ago.

When I learned about this book, I had to have it. I found it on Amazon, and it cost a fortune. At least in book-price terms. Here the little turkey is.

Here is an Amazon link, but don’t buy it. It’s a small book, and not really worth the money unless, like me, you need it for professional purposes. For me, it was the find of the century. Now I can really get into the dietary habits of the early Egyptians with confidence.

Why is the diet of the early Egyptians important? Because they left about a zillion mummies from that period. Most of the ancient archeological record is in the form of skeletons only. The Egyptian mummy data includes soft tissue.

The ancient Egyptians followed a diet that would be your typical nutritionist’s nirvana today. A lot of whole-wheat bread, no sugar (sugar didn’t come on the scene until about 1025 AD), minuscule amounts of red meat, if any, vegetables, and a little fish.

And yet the mummies of these ancient folks show that they were crawling with all the diseases today’s nutritional ‘experts’ would tell you eating the ancient Egyptian diet would prevent. They had horrible tooth decay, obesity, inflammation, and atherosclerosis, even at an early age. So, clearly their diet didn’t help them a whole lot.

The book above shows—based on documents from the time—that the average workman received an allotment of emmer wheat from the granary as a salary. Each workman received 4 sacks monthly. A single sack held 300 liters of wheat, which calculates to about 80 gallons. So 4 sacks would be 320 gallons, which is a lot. According to the book, this amount would feed a family of ten with plenty left over to trade. The emmer wheat was traded like currency.

Along with the wheat, the workers received 1.5 sacks of barley, which was used to make beer. Bread and beer were the staples of the Egyptian diet of the time. In addition, they were provided with water, vegetables (mainly beans), a bit of fat (the kind wasn’t specified, but I would imagine some kind of oil), and fish. How much fish? According to the equivalent of pay stubs, about 8.4 kg per month. Which calculates out to a little over 18 pounds. Seems like a lot, but not when it is spread out over an entire family. The beans and the fish were the primary sources of protein.

The emmer wheat was the primary dietary staple, and it was coarsely ground, so all the so-called good parts of the wheat berry were included. Despite its not being particularly finely processed, this bread had to have generated a significant incretin effect, which means insulin was elevated. It had to have been for them to have so many metabolic disorders.

All of this business about the UPF, the study by Hall et al, and the Egyptian data leads me to a couple of conclusions. First, keep carbs to a minimum. And, second, and probably the most important, the best thing you can do for your health is spend more time in your own kitchen. If you make the food, you have control over what goes into it or how much it is manipulated. But if the definition of UPF is foods made with anything not found in your kitchen, you will never eat UPF.

Famous Hall Study Refuted

While I was on a roll writing about Kevin Hall, I was notified by David Ludwig that his group had refuted a paper near and dear to Kevin’s heart.

Here’s the situation.

There wages an academic feud right now about why it is that we gain or lose weight. On one side we have those who believe it’s just a matter of calories in vs calories out. If you eat too much, you’ll gain weight. If you want to lose, you need to eat less and move more. Because it’s all a matter of calories. Now, they’ve recently tried to tart up their argument to make it look like it’s more than it is, but at its root, it is calories in vs calories out. The leading light of this argument at this time is Kevin Hall from the NIH.

The other side of this argument says the gain or loss of weight is a function of insulin status. If you produce too much insulin, you will store fat. By moving the nutrients—fat and glucose—from the blood and storing them in fat, the blood becomes deficient, which sends a signal to the hunger center in the brain saying Eat more. When we do eat more, we drive insulin up, which stores the nutrients, and the cycle begins again. Since what drives insulin up is primarily carbohydrate, the process is called the carbohydrate-insulin model (CIM) of obesity. The academic leader of this argument is David Ludwig at Harvard.

The debate over these issues has been fierce, but congenial. As it should be.

There is one specific point both sides have been trying to demonstrate as proof their theory works. According to the calories in vs calories out (CICO), people who maintain their weight should have the same energy expenditure (calories out) irrespective of the composition of their diet. If they eat 3,000 calories/day, and then burn 3,000 calories/day, their weight should stay the same. Doesn’t matter whether the diet is high or low in carbohydrates. Or fat. It’s all a matter of calories.

The CIM group claims that the type of calories makes a difference. If an individual consumes a high-carb diet, then insulin will drive some of those carb calories into fat and send them to the fat cells instead of burning them. Which means the subject in question will have to consume more calories to be able to fuel their energy needs. Conversely, if he/she consumes a diet low in carbs, say, a ketogenic diet, insulin levels will fall, fat will come out of the fat cells to be burned. So weight can be maintained on a lower caloric intake. Their caloric needs are being made up in part by the fat they’re releasing from their stores. People following such a diet will either need to eat more to maintain their weight, or they’ll lose weight.

To examine these theories, Hall et al did a study, published in 2021, to test the CIM theory. (The paper is behind a paywall. Here is my Dropbox copy.)

The researchers recruited 20 moderately overweight (BMI = 27.8) subjects aged ~30 years old to act as their own controls while following one of two diets for 14 days. These subjects were admitted to the NIH Clinical Center where their intake could be controlled.

For one 2-week period a subject would follow one of two diets, then switch to the other for the second 2-week period. The subjects were randomized into which diet they would follow first. In other words, they did not all start on the same diet for the first week.

The two diets were provided ad libitum, which means all the food the subjects wanted to eat. One was a “minimally processed, plant-based, low-fat diet (10.3% fat, 75.2% carbohydrate) with high glycemic load (85 g 1,000/kcal).” The other diet was a “minimally processed, animal-based, ketogenic, low-carbohydrate diet (75.8% fat, 10.0% carbohydrate) with low glycemic load (6 g /1,000 kcal).”

The researchers wanted to see if either diet induced the subjects to eat less. If the CIM prevailed, then the subjects on the ketogenic diet would consume fewer calories over the course of their two weeks on that diet. They found the opposite.

We found that the low-fat diet led to 689 ± 73 kcal d−1 less energy intake than the low-carbohydrate diet over 2 weeks (P < 0.0001) and 544 ± 68 kcal d−1 less over the final week (P < 0.0001).

Their understated but wry conclusion was

Therefore, the predictions of the carbohydrate–insulin model were inconsistent with our observations.

As you might imagine, this paper got a lot of play everywhere. The authors proclaimed that not only did the ketogenic diet cause people to eat more, the subjects actually lost more fat mass on the two weeks of the high-carb, plant-based diet.

All in all, it was a pretty brutal takedown of the ketogenic diet.

But… There is always a but.

David Ludwig just sent me a new paper from his group that pretty much lays waste to this one.

Hall et al made one critical error in their study. They didn’t use a washout period between the two diets. It’s one thing not to do so in the UPF study we discussed above, but a totally different thing when two so diametrically different diets are compared.

As all of us know, it takes a while to become low-carb adapted. And as we saw in the discussion of females on a ketogenic diet we discussed a few Arrows ago, it takes a while going from a ketogenic diet to a low-fat, high-carb Standard UK diet before the effects of the ketogenic diet begin to fade. So not having a washout period in this study was a major mistake.

Which somewhere along the way Hall et al figured out. And to their credit, they reanalyzed their data and wrote a paper on the reanalysis. They looked at both the UPF study vs the minimally processed food study we discussed earlier and the low-carb vs high-carb study linked above.

As it turned out, the lack of a washout period didn’t really matter with the UPF vs MPF study, but it made a major impact on the one under discussion here. I’ll have to give them credit for publishing their reanalysis, though they didn’t really announce that it was a reanalysis of the study they had made such hay with earlier.

Then Ludwig’s group did an analysis of Hall’s reanalysis and showed that the original data, when viewed from the perspective of what diet went first, was actually an argument for the CIM.

A couple of graphics from the second Hall paper says it all.

This chart is a bit confusing because of the mix of colors and symbols. Here is the legend for the chart.

  • Circles represent the low-carb diet

  • Squares represent the low-fat diet

  • Red means low-carb for the first two weeks, then low-fat

  • Blue means low-fat for the first two weeks, then low-carb

You can clearly see that in those who started off with the low-carb diet then switched to low-fat for the second two weeks there was a carry over effect. That is represented by the red line kind of on the bottom. Subjects starting on the LC diet reduced their caloric intake over the first two weeks, then when the diet was switched to low-fat, because of metabolic priming from the low-carb diet, the decreased caloric intake carried on despite the switch to low-fat for the second two weeks.

If you look at the blue boxes on the left, those tell what happens when subjects started on the low-fat diet. They took in a bit more calories than those starting on the low-carb diet, but then when they switched to low-carb the calorie intake really took off thanks to increased insulin (as shown in the later Ludwig study) generated by the first two weeks of low-fat.

What becomes completely obvious from this graphic, is that the meal order and lack of a washout period invalidated this study.

Here is another graphic. This one looking at changes in body weight.

Same legend as the graph before. If you look at the first two weeks weight loss on those starting off with the low-carb diet (the red circles), you can see a nice steady loss. Then when the subjects switch to the low-fat diet (the red squares), they continue to lose weight, just not as quickly.

Looking at the blue at the top, you can see that those who start with the low-fat diet lose less than those in the red circles below who were following the low-carb diet first. Then when they switch to low-carb (the blue circles), they continue to lose, but not as much as those starting off with the low-carb did.

Here is a link to the Ludwig group’s paper. It’s fairly technical.

Nick Norwitz, one of the co-authors, made a nice video going over the findings.

The take home message is that you really need a washout period to do these types of studies. And the original study that was used to dance on the low-carb diet’s grave, has ended up weighing in on the side of the CIM. Funny how that worked out.

Okay, I’m going to do something different here. I come across all kinds of things I want to post about, or at least mention, but I invariably run out of time. And space. And probably readers’ patience. So I’m going to start sticking these things in as links.

Here’s my first go at it.

Video of the Week

I’ve narrowed it down to two this week, so instead of picking, I’m going to go with both.

First is a video of a guy who has never hunted in his life coming to Arkansas to go on a duck hunt in Stuttgart, which is sort of in the middle-eastern part of the state and one of the rice capitals of the world. I wouldn’t come close to the place in the summer, because the mosquitos are fearsome. In duck season the mosquitos have gone dormant. Or whatever it is mosquitos do when it’s cold. Saw this on The Free Press, one of my fav Substacks.

The next video is of Prime Time, ol’ Neon Deion Sanders himself, talking about what he looks for in players for various positions.

Then there is the weekly poll where you get to rate me just like Deion does his players.

Once again, I’m wishing everyone a Merry Christmas! Keep in good cheer, and I’ll be back next Thursday.

And don’t forget to help me out by clicking the Like button, assuming, of course, that you liked it.

Thanks for reading all the way to the end. Really, thanks. If you got something out of it, please consider becoming a paid subscriber. I would really appreciate it.

Finally, don’t forget to take a look at what our kind sponsors have to offer. Dry Farm Wines, HLTH Code, Precision Health Reports, The Hustle (free), and now The Morning Brew (also free).

Join the conversation

or to participate.