Even back when keto worked amazingly for me it was still pretty clear that it was probably only intersecting with the real truth in a few crucial places, like you described in your "evaluating diets" post. There were just too many outliers and caveats for insulin theory to actually be the complete, root explanation of modern obesity. Sure, it produced better results than traditional nutrition advice, but it definitely seemed to be oopsing its way into success at least part of the time. It sucks to admit, but the anti-PUFA movement also shows similar signs that, while it's a lot better at explaining things keto never could, it's still not the ultimate one-stop magic bullet explanation.
Slightly confused: when you say a failure diet "allows you to starve yourself slightly better," do you mean "allows you to tolerate hunger and deliberate restriction?" As in anything that forces you to unnaturally stop before you otherwise would? If CICO is just an accounting tautology anyway, it doesn't seem like any diet that ends up with the person reducing their caloric intake should necessarily be considered a failure.
100% agree, insulin probably plays a big role. But I'm not convinced it's the root cause, and it certainly doesn't seem to be a complete theory. It might be 1/3 of the picture or so, which is great! Let's figure out the rest.
And yea, agreed on seed oil theory. It can't explain swamping (yet), or protein restriction, both of which clearly work anecdotally for many people (again, not all and not everyone seems to need it!)
I think that if caloric restriction isn't a great strategy, tricking yourself into doing it is probably also not a great strategy. If a diet enables you to eat less calories (be it via conscious restriction or e.g. intuitively not eating as much) without the side effects, that's good. But simply blunting the hunger signal while maintaining all of the side effects sounds terrible. That's like turning off the smoke detector instead of putting out the fire.
I genuinely think that the wide range of diets that work versus don't work for different people are significantly influenced by genetic inborn errors of metabolism, most of which can only be diagnosed with full genome sequencing. For example, I know I have a genetic issue with very long fatty acids, and some issue with long fatty acids (i.e. most dietary fats). At first I only noticed it on an OmegaQuant years ago, then a year ago I got my full genome done and saw the genetic variant for it. I had my partial genome from 23andMe for over a decade, it just doesn't include most things.
Then I realized there are around 1500 identified genetic inborn errors of metabolism that could dramatically impact someone's ability to handle not just a broad group of macronutrients (fats, carbs, protein) but also very specific fatty acids, amino acids, kinds of carbs, etc. At present, the only way these are ever diagnosed is when someone has the super serious fatal or near-fatal issues at infancy, but you can have less serious versions that still cause big problems and no doctor ever figures it out. You just suffer your whole life trying stuff.
Here is just a single example of how impossible it can be for someone to untangle using conventional diet ideas: Propionyl-CoA carboxylase is an enzyme that acts on a metabolite of all the following compounds when they're broken down: the beta-oxidation of only *odd-chain* fatty acids; the amino acids isoleucine, valine, methionine, threonine; the side chain of cholesterol. It also requires biotin to work. So if you have a genetic issue with that enzyme, OR you have any of a number of genetic issues with biotin, OR if you have a dietary biotin deficiency, the data you get from dieting is going to seem VERY weird and inconsistent. It might seem like you have an issue with "protein," but because you really just have an issue with four amino acids, some "proteins" are much worse than others. You might notice that you struggle with dairy fat (most fats are even-chained but dairy has the ones that are odd-chained) but you're lactose tolerant, e.g. maybe you can eat saturated fat from beef, can't tolerate whole milk, but CAN tolerate skim milk. I'm not even sure what cholesterol markers do in response to diet in such a scenario. And if it's a dietary biotin deficiency, your response to foods could vary dramatically depending on whether you ate them with something high in biotin.
Imagine how invisible that would all seem to such a person. There's not a diet movement that encourages fatty beef but skim milk. Hardly anyone talks about amino acid composition of proteins. All such a person could do (without genetic data) is gauge their response to EVERY SINGLE FOOD. The extent to which they start to believe in ANY diet ideology is going to hold them back.
I'm convinced that most people have at least ONE genetic issue somewhere in their metabolism, and that many people have several. There are always some clowns who pipe up with no expertise whatsoever claiming these genetic variants are "rare," which is based on absolutely nothing, because the vast majority of people have gotten no genetic sequencing done whatsoever. There's stuff that was on my 23andMe report that was thought to be rare a decade ago, then further information suggests those variants are common, sometimes the norm rather than the exception. But even now we can't say what's normal or rare, because most people getting their genes sequenced have severe enough health problems to pay out of pocket for it.
Anyway, I think it makes sense for people to try different kinds of diets based on a LOOSE understanding of how the mechanisms seem like they operate, but they should feel free to abandon anything that isn't working. When they find something that works, they should experiment with adding in SPECIFIC foods and seeing what happens. For example, my dad had double diabetes (that's when you have type 2 diabetes so badly it damages your pancreas and gives you type 1 diabetes) but did great (i.e. went off meds and had good blood sugar) on low carb foods PLUS potatoes and fruit. He died before paleo was a thing but I suspect it would have been a good fit for him. Another example: I did decently on low carb, then was shocked I could lose weight on paleo, then was even MORE shocked that I could drink honey all day on paleo and lost more weight than any diet ever. I had never been able to lose weight with higher carbs, JUST honey. And I have no idea why, my genetic analysis is extremely unsophisticated. It may have nothing to do with genes and more to do with calming some sort of allergy, who knows.
Point is, I appreciate this post because this stuff is SO MUCH MORE INDIVIDUAL than diet ideologues want it to be. When people find something that works for them they think it must be that simple for everyone. In reality, they probably just have genetic issues with other diets, and don't have any genetic issues with their favored diet.
Yea the body is a crazy complex system. I am surprised it works as well as it does!
Unfortunately I am super skeptical of the current state of these genetics tests. A few friends of mine have done them, and, very suspiciously, the result never indicates "you should only eat heavy cream" or "you'd do well on carnivore or keto." It's always whole grains and vegetables, it just ranges basically from mediterranean diet to vegetarianism lol. I forgot if this was 23andme, but one of the more mainstream popular ones.
We seem to know next to nothing about the genes and their impact, and kinda like with other sorts of mechanisms, people pick their favorite one and get hung up on it.
That's why I like black box debugging, like you describe. Strong beliefs, weakly held. And yes it might mean eating the same food 30 days in a row a lot and adding in 1 variable at a time. But that's probably what it takes, cause I'm not holding my breath for them to find out how the genetics actually work.
Oh, to be clear I’ve never really read or paid attention to what any genetic service recommends for diet. I wasn’t even aware that was a thing.
To the contrary of not really knowing what the genes do, it’s often quite clear-cut when a gene effects a nutrient or enzyme negatively. We know what the genes code for, it’s not that difficult to have an idea of what sorts of changes are likely to cause problems, and then amongst those a ton have been confirmed clinically to cause the predicted problems. As I said, the extreme versions of these (meaning either homozygous variants, or just variants that severely change the shape of the things they code for) are literally fatal soon after birth so we’re not puzzled about their effect or anything.
The less extreme variants (meaning either heterozygous versions where a person still has a normal copy of the gene functioning, or else multiple changes to the same gene that combine to result in outputting a very warped shape just like one severe change can) still have negative effects without killing people. For example, the severe version of my VLCAD genetic issue can be fatal; for other people like myself, it just fucks with them forever. We’re not unsure what the genes do, it’s very clear. The accumulation of very long chain fatty acids shows up on my tests as expected.
Similarly, I have genetic issues with B12 and folate and B6 and choline that all predictably show up on tests.
Here is a very well-organized list of the genetic errors of metabolism we’ve identified: http://www.icimd.org/
Here is a page where you can search for different genetic diseases and sometimes see what lab markers are associated with them: https://www.iembase.com
Nah, we know what proteins most genes code for, and when they code for a specific enzyme, receptor, transporter, etc we can have a good idea of how disastrous they can be when they don't work properly. The trickier part until just recently was trying to guess if a variant would make a protein shaped basically the same with similar functionality, or whether the protein would be severely misshapen and react poorly, but for a while we had ways to make good guesses we could research, and then AI just solved the protein shaping problem. But that's too new to help individuals with knowing for certain whether their many small variants result in a severely ineffective protein or not; for now, we can just guess that it might.
The MTHFR variant thing is an example of a solid one though, because we know exactly what enzyme those variants are interfering with. However, with that specific enzyme and some others, there are so many ways to help that enzyme function better that it isn't guaranteed to be a big issue for someone depending on their diet and other genes. But it's a great example of a variant where someone might find that only a narrow range of diets works to maintain health. and some diets are disastrous.
As a side note, one reason I'm convinced we can't just say "oh this country thrives on this diet so something something macros" is because some nutrient-related genetic variants are overrepresented in certain regions. Italy has a couple of pockets of remarkably unusual cholesterol genetics; one is ApoA-I Milano, which gives them crazy low HDL and higher triglycerides, yet is extremely protective against heart disease and obesity, even when they eat a lot of sugar and fat. And the genes of two of my Japanese friends (who aren't related to each other) both contain an issue with vitamin K that can lead to heart arrhythmias, but the traditional Japanese diet can be very high in vitamin K (natto, matcha, other fermented foods) so the symptoms of those genes never actually arise for many Japanese people who have them. Vitamin K can be involved in the electron transport chain.
I wonder if you might be conflating this sort of thing with attempts to point huge concepts like "obesity," "IBS," and "cancer" and link them to specific genes. Those sorts of things are usually multifactorial and so genetics are difficult to untangle, although some cancers are directly due to a specific genetic variant and nothing else. Maybe that's where you've been seeing diet recommendations? If you don't look at the actual genes or know how to evaluate them, then you're reliant on whatever weird assessments genetic services sell. Those are typically like "here's your risk of heart disease" and it just assesses a handful of genes that seem statistically associated, often not strongly. It doesn't tell you anything real because it's speculative to begin with, almost never deterministic, and extremely incomplete. Then they'll spit out the conventional advice for heart disease, IBS, whatever, but none of that advice is based on genetics, they're just providing boilerplate advice for whatever condition you expressed interest in. It might even be in direct contradiction to some of the genetic variants found, depending on what they are.
That's completely different than knowing you have a genetic issue with a biotin-dependent enzyme and that you need high dose biotin to function, or you can't metabolize a certain kind of fat, or you struggle to methylate folate, etc. If you try running your genetic data through something like Nutrahacker you get much more specific advice, although you can have genes with contradictory solutions (e.g. you may need to avoid methyl groups for one gene, but you need them for another; a machine can't square that and tell you what to do in reality), and it flags some variants that are probably not important, and for a few genes it does give brief dietary advice (e.g. low fat, or ketogenic, or fiber, etc) which can also conflict with other genes, so it still requires some discernment.
But again, 23andMe or any other partial genome service is not going to cover the vast majority of inborn errors of metabolism. Only a full genome can do that. And right now there's not a great service that can simply scan your full genome and point those out, to my knowledge; you have to manually check for ones that you suspect you might have, and then it can be confusing digging into the research.
Yea what my friend did was exactly what you described. They took her DNA and then recommended she eat the mediterranean diet, basically. I bet they recommend nearly the same thing to everyone. I doubt they'd recommend 90% fat, heavy cream based keto to me lol.
I'm also probably confused by the factor you mention with MTHFR, which is the one I've tested on myself and which is probably confounded like the Japanese vK issue: there are so many ways to get around it that chances are high you never notice the downside. Unless you happen to do one very specific diet maybe, that doesn't have one of the "fixes."
I have supplemented folate and methylated B vitamins based on this MTHFR thing, and over 3-6 months saw no difference whatsoever. (Except my folate shot up in the "too high" range on blood panels.)
That's why I'm very skeptical of many of these mechanistic speculation/explanation type things. I've yet to see somebody miraculously cure diabetes or obesity by diagnosing MTHFR and then supplementing folate.
I guess there are these cases like you have where it makes a day and night difference, but since I haven't seen a fix like that myself based on genetics, I'm not thinking about it much.
You could argue that my keto/non-24 thing is exactly this. We just don't know if it's genetic (AFAIK the cause of non-24 is totally unknown). There are CLOCK genes, but they just determine your chronotype. Non-24, especially the type I have, doesn't seem to be genetic or at least they haven't found the gene.
Long term keto fat cells clearly become very insulin sensitive, so it is easy to gain weight with the small insulin bump from protein despite the liver making ketones to keep the brain buzzing along. Starting keto, the insulin insensitive lose weight rapidly until the body reaches a new homeostasis, with luck that is at 15% body fat, but for many it is much higher.
I believe that this homeostatic level is down to the level of Omega 6 in the body and the only way to get down to 15% is keto with zero Omega 6, or high carb zero fat, which is naturally zero Omega 6, for many years.
I am a fully paid up member of the keto club because it seems logical and I believe it to be healthy.
But it's very frustrating when other diets such as slime molds, potato, riff creates weight loss that can't be explained with keto rules.
There must be a piece of the jigsaw missing. Something that will allow fat out of adipocytes that will work irrespective of insulin level and possibly inhibit fat storage.
For one, I'd say, we don't actually know the insulin levels throughout the day of a potato dieter. Maybe their insulin is pretty harmless. That's one big weakness of the "Carb Insulin Model" I think, while the insulin -> fat part is pretty well studied and understood, I think our idea of "carbs == insulin" is quite flawed. It's way more complicated than that. E.g. some proteins seem to be as insulinogenic as many carbs in the few studies we've seen.
Is a potato/rice diet really as insulinogenic as a candy-laden high-PUFA S.A.D.?
As a recent carb convert - easily 30% of my carbs are literally candy. Just Red Vines, not M&Ms, Haribo not Snickers. All the HFCS i care to eat, but avoiding candy fat and seed oils.
Haha whatever works :) You are probably right that the mix w/ seed oils (palm oil in candy?) is at fault. Just listened to an Anabology video where he did 1lb of honey a day because honey is pretty much the only pure sugar he can even eat that much of. Apparently pure sugar is not THAT appetizing.
My question is why? Why does keto not work for all?
The makings of a good cake are there :-)
- Low insulin, fat oxidation, relying on your own body fat for energy
I think that the missing pieces are
1) We cant match fat oxidation with fat intake as carbs do. "Chronic undernutrition and metabolic adaptation". And I don't think metabolic rate goes-up in the way Ben and team describe.
2) I believe that Jay has got it right in his recent videos. There are blockages in the system that direct substrate towards fat storage and not usage. And this might be even in the setting of low insulin, high ketones etc. Maybe it is about the microbiome. That would make sense to me.
3) Muscle and exercise are hugely under appreciated.
More interestingly is how do we measure if the diet is appropriate for the person. For me
1) Body composition (not weight)
2) Energy levels ( if you feel good just give it time)
3) Stress as an indicator of lack of energy/high demands (TSH , DHEA, Cortisol, Super High Cholesterol?, high BG maybe a lot more)
- I wonder if a high-protein, high-PUFA diet is actually low insulin for everyone. We can't measure insulin well (lab tests only) so we have no clue. Maybe, in me, beef is incredibly insulinogenic and that's why it makes me fat?
To your questions:
1) Does Ben (Bikman?) say metabolic rate should go up? On keto? Haven't seen that particular thought from him (more from Peaters)
2) Jay Feldman? What you describe sounds a lot like "fuel partitioning" which I'm a huge believer in. I think insulin is clearly one mediator of fuel partitioning, but it might not be the only/key one. I.e. maybe I have very high insulin from protein and we just don't know, or maybe I have low insulin but something else mediates the fuel partitioning.
3) I don't know, I think it's overestimated. I just don't know anyone who's had any success losing fat or fixing diabetes with resistance training. You can burn off the carbs you eat with endurance exercise, but that's sort of a band-aid over your diabetes, I think. I do know people who do that.
I agree on your measurements, although for most people weight is a decent, easy to measure proxy. None of us wake up with 30lbs of extra muscle, so if you factor in water retention/glycogen or don't make drastic changes, weight is easy. Of course check in w/ a DEXA once in a while or similar. Although to be honest, you can tell your body composition in the mirror if you're honest. Nobody accidentally gets to 9% body fat and 220lbs of lean muscle. The people who are these "weight/BMI outliers" put a decade of work into it and they know who they are :)
3) When referring to a diet working or not, I was mostly thinking of the how you can identify possible root causes. If you are on keto and thyroid gets messed-up maybe that is a root cause etc.
Ha the thing is, testing for insulin is insanely annoying since you have to do it in a lab. I spent 6h in the lab that day for the Kraft. Too much effort to repeat it a lot, unfortunately..
3) yea agreed, that could happen. Hasn't seemed to happen in my case, but for sure easy & good to check regularly. Thyroid panel & T panel is very cheap. Just do it quarterly or whatever.
I just don't see that many people with messed up thyroid markers. They do exist for sure, but e.g. mine are actually very hypERthyroid in parts. I've always had high body temperature since childhood, too.
I think you did Keto wrong. AKA, you did "internet keto" but not "prescription-strength keto" (limiting fats). Check out Dr. Eric Westman, Duke University and Amy Berger. Authors of "End Your Carb Confusion"
It's not an "unlimited fat" diet, if you're trying to burn your own body fat, that is. a keto meter's not going to show you if you're burning your own body fat vs. burning the fat you just consumed.
You have to count how much added fat (high calorie foods like mayo, cream, butter, etc.) you use per day. Proteins are unlimited though. I'm doing their program now.
Tons of his blog posts are higher protein experiments in which he reliably gains weight. While it's true that most of his blood ketones are likely dietary and not from stored fat, if he gains weight on both increased carbs and increased protein, that leaves him with eating fat to function.
Yea the high-protein/low-fat "keto" diets just aren't sustainable for me. It's basically an inefficient fast; I tend to get starvation symptoms and I have to stop, and then it all comes back just like with the fasting.
"prescription-strength keto" is limiting fat? The classic ketogenic diet is 90% fat by calories, which I am roughly on.
I would argue that high-protein, low-fat keto is "internet keto."
I guess the diet you describe is a version of keto and it can be prescribed for weight loss, but I have tried higher protein before and it never worked. It just makes me insatiable. I basically have to limit protein to the extreme to reach any level of satiety.
Dr. Eric Westman is as Duke University, he took over for Dr. Atkins (kind of...he uses Dr. Atkins food list - the list Dr. Atkins used for 30 years & now Dr. Westman has been using it for 20 years, so 50 years total! I am using Dr. Westman's/Adapt Your Life Academy's way of eating, and I'm having a great result!! I'm especially happy 2day because I'm down another jean size, and bra size, PTL!
Per Dr. Westman - I'm paraphrasing what he said - the hallmark of a ketogenic diet is: to keep the total carbohydrate LOW (note: not "net carbs" even though I think Dr. Atkins invented the concept of "net carbs"? We don't count net carbs on the Keto Made Simple (KMS) approach). In "Phase One," the intro phase, which I'm doing now, the carbs are kept under 20 grams - that way 100% of the people doing the class are in "nutritional ketosis." Per Dr. Westman, if you go up to 50 grams per day of carbohydrate, only 50% of people will be in nutritional ketosis.
To be honest, at this point, I'm not just doing it for weight loss - I'm not overweight for my height (I'm 62 kgs at 5'7) - but the ketones are changing my life *for the better*!!! I have some chronic pain issues from prior back injuries (herniated disc) and being in nutritional ketosis is like magic for my inflammation - the closest I think I'll ever get to a miracle cure.
Two books you could read: "End Your Carb Confusion" by Eric Westman, M.D. and Amy Berger and
"The Art & Science of Low Carbohydrate Living" by Volek and Stephen Phinney M.D. PhD
Also, feel free to check out my blog iusedtobeatrackstar (dot) blogspot (dot) com
Even back when keto worked amazingly for me it was still pretty clear that it was probably only intersecting with the real truth in a few crucial places, like you described in your "evaluating diets" post. There were just too many outliers and caveats for insulin theory to actually be the complete, root explanation of modern obesity. Sure, it produced better results than traditional nutrition advice, but it definitely seemed to be oopsing its way into success at least part of the time. It sucks to admit, but the anti-PUFA movement also shows similar signs that, while it's a lot better at explaining things keto never could, it's still not the ultimate one-stop magic bullet explanation.
Slightly confused: when you say a failure diet "allows you to starve yourself slightly better," do you mean "allows you to tolerate hunger and deliberate restriction?" As in anything that forces you to unnaturally stop before you otherwise would? If CICO is just an accounting tautology anyway, it doesn't seem like any diet that ends up with the person reducing their caloric intake should necessarily be considered a failure.
100% agree, insulin probably plays a big role. But I'm not convinced it's the root cause, and it certainly doesn't seem to be a complete theory. It might be 1/3 of the picture or so, which is great! Let's figure out the rest.
And yea, agreed on seed oil theory. It can't explain swamping (yet), or protein restriction, both of which clearly work anecdotally for many people (again, not all and not everyone seems to need it!)
I think that if caloric restriction isn't a great strategy, tricking yourself into doing it is probably also not a great strategy. If a diet enables you to eat less calories (be it via conscious restriction or e.g. intuitively not eating as much) without the side effects, that's good. But simply blunting the hunger signal while maintaining all of the side effects sounds terrible. That's like turning off the smoke detector instead of putting out the fire.
Keto has failed? Obviously you didn't try it hard enough. If you're eating anything other than pure coconut oil you deserve to be fat.
Actually true keto is eating 0 fat and only protein. And also it has never been tried.
Most protein comes with fat.
I genuinely think that the wide range of diets that work versus don't work for different people are significantly influenced by genetic inborn errors of metabolism, most of which can only be diagnosed with full genome sequencing. For example, I know I have a genetic issue with very long fatty acids, and some issue with long fatty acids (i.e. most dietary fats). At first I only noticed it on an OmegaQuant years ago, then a year ago I got my full genome done and saw the genetic variant for it. I had my partial genome from 23andMe for over a decade, it just doesn't include most things.
Then I realized there are around 1500 identified genetic inborn errors of metabolism that could dramatically impact someone's ability to handle not just a broad group of macronutrients (fats, carbs, protein) but also very specific fatty acids, amino acids, kinds of carbs, etc. At present, the only way these are ever diagnosed is when someone has the super serious fatal or near-fatal issues at infancy, but you can have less serious versions that still cause big problems and no doctor ever figures it out. You just suffer your whole life trying stuff.
Here is just a single example of how impossible it can be for someone to untangle using conventional diet ideas: Propionyl-CoA carboxylase is an enzyme that acts on a metabolite of all the following compounds when they're broken down: the beta-oxidation of only *odd-chain* fatty acids; the amino acids isoleucine, valine, methionine, threonine; the side chain of cholesterol. It also requires biotin to work. So if you have a genetic issue with that enzyme, OR you have any of a number of genetic issues with biotin, OR if you have a dietary biotin deficiency, the data you get from dieting is going to seem VERY weird and inconsistent. It might seem like you have an issue with "protein," but because you really just have an issue with four amino acids, some "proteins" are much worse than others. You might notice that you struggle with dairy fat (most fats are even-chained but dairy has the ones that are odd-chained) but you're lactose tolerant, e.g. maybe you can eat saturated fat from beef, can't tolerate whole milk, but CAN tolerate skim milk. I'm not even sure what cholesterol markers do in response to diet in such a scenario. And if it's a dietary biotin deficiency, your response to foods could vary dramatically depending on whether you ate them with something high in biotin.
Imagine how invisible that would all seem to such a person. There's not a diet movement that encourages fatty beef but skim milk. Hardly anyone talks about amino acid composition of proteins. All such a person could do (without genetic data) is gauge their response to EVERY SINGLE FOOD. The extent to which they start to believe in ANY diet ideology is going to hold them back.
I'm convinced that most people have at least ONE genetic issue somewhere in their metabolism, and that many people have several. There are always some clowns who pipe up with no expertise whatsoever claiming these genetic variants are "rare," which is based on absolutely nothing, because the vast majority of people have gotten no genetic sequencing done whatsoever. There's stuff that was on my 23andMe report that was thought to be rare a decade ago, then further information suggests those variants are common, sometimes the norm rather than the exception. But even now we can't say what's normal or rare, because most people getting their genes sequenced have severe enough health problems to pay out of pocket for it.
Anyway, I think it makes sense for people to try different kinds of diets based on a LOOSE understanding of how the mechanisms seem like they operate, but they should feel free to abandon anything that isn't working. When they find something that works, they should experiment with adding in SPECIFIC foods and seeing what happens. For example, my dad had double diabetes (that's when you have type 2 diabetes so badly it damages your pancreas and gives you type 1 diabetes) but did great (i.e. went off meds and had good blood sugar) on low carb foods PLUS potatoes and fruit. He died before paleo was a thing but I suspect it would have been a good fit for him. Another example: I did decently on low carb, then was shocked I could lose weight on paleo, then was even MORE shocked that I could drink honey all day on paleo and lost more weight than any diet ever. I had never been able to lose weight with higher carbs, JUST honey. And I have no idea why, my genetic analysis is extremely unsophisticated. It may have nothing to do with genes and more to do with calming some sort of allergy, who knows.
Point is, I appreciate this post because this stuff is SO MUCH MORE INDIVIDUAL than diet ideologues want it to be. When people find something that works for them they think it must be that simple for everyone. In reality, they probably just have genetic issues with other diets, and don't have any genetic issues with their favored diet.
Yea the body is a crazy complex system. I am surprised it works as well as it does!
Unfortunately I am super skeptical of the current state of these genetics tests. A few friends of mine have done them, and, very suspiciously, the result never indicates "you should only eat heavy cream" or "you'd do well on carnivore or keto." It's always whole grains and vegetables, it just ranges basically from mediterranean diet to vegetarianism lol. I forgot if this was 23andme, but one of the more mainstream popular ones.
We seem to know next to nothing about the genes and their impact, and kinda like with other sorts of mechanisms, people pick their favorite one and get hung up on it.
That's why I like black box debugging, like you describe. Strong beliefs, weakly held. And yes it might mean eating the same food 30 days in a row a lot and adding in 1 variable at a time. But that's probably what it takes, cause I'm not holding my breath for them to find out how the genetics actually work.
Oh, to be clear I’ve never really read or paid attention to what any genetic service recommends for diet. I wasn’t even aware that was a thing.
To the contrary of not really knowing what the genes do, it’s often quite clear-cut when a gene effects a nutrient or enzyme negatively. We know what the genes code for, it’s not that difficult to have an idea of what sorts of changes are likely to cause problems, and then amongst those a ton have been confirmed clinically to cause the predicted problems. As I said, the extreme versions of these (meaning either homozygous variants, or just variants that severely change the shape of the things they code for) are literally fatal soon after birth so we’re not puzzled about their effect or anything.
The less extreme variants (meaning either heterozygous versions where a person still has a normal copy of the gene functioning, or else multiple changes to the same gene that combine to result in outputting a very warped shape just like one severe change can) still have negative effects without killing people. For example, the severe version of my VLCAD genetic issue can be fatal; for other people like myself, it just fucks with them forever. We’re not unsure what the genes do, it’s very clear. The accumulation of very long chain fatty acids shows up on my tests as expected.
Similarly, I have genetic issues with B12 and folate and B6 and choline that all predictably show up on tests.
Here is a very well-organized list of the genetic errors of metabolism we’ve identified: http://www.icimd.org/
Here is a page where you can search for different genetic diseases and sometimes see what lab markers are associated with them: https://www.iembase.com
Thanks I'll check it out. I got tested for the MTHRFCKR gene and my impression was "we have no fucking clue what any of this does."
Nah, we know what proteins most genes code for, and when they code for a specific enzyme, receptor, transporter, etc we can have a good idea of how disastrous they can be when they don't work properly. The trickier part until just recently was trying to guess if a variant would make a protein shaped basically the same with similar functionality, or whether the protein would be severely misshapen and react poorly, but for a while we had ways to make good guesses we could research, and then AI just solved the protein shaping problem. But that's too new to help individuals with knowing for certain whether their many small variants result in a severely ineffective protein or not; for now, we can just guess that it might.
The MTHFR variant thing is an example of a solid one though, because we know exactly what enzyme those variants are interfering with. However, with that specific enzyme and some others, there are so many ways to help that enzyme function better that it isn't guaranteed to be a big issue for someone depending on their diet and other genes. But it's a great example of a variant where someone might find that only a narrow range of diets works to maintain health. and some diets are disastrous.
As a side note, one reason I'm convinced we can't just say "oh this country thrives on this diet so something something macros" is because some nutrient-related genetic variants are overrepresented in certain regions. Italy has a couple of pockets of remarkably unusual cholesterol genetics; one is ApoA-I Milano, which gives them crazy low HDL and higher triglycerides, yet is extremely protective against heart disease and obesity, even when they eat a lot of sugar and fat. And the genes of two of my Japanese friends (who aren't related to each other) both contain an issue with vitamin K that can lead to heart arrhythmias, but the traditional Japanese diet can be very high in vitamin K (natto, matcha, other fermented foods) so the symptoms of those genes never actually arise for many Japanese people who have them. Vitamin K can be involved in the electron transport chain.
I wonder if you might be conflating this sort of thing with attempts to point huge concepts like "obesity," "IBS," and "cancer" and link them to specific genes. Those sorts of things are usually multifactorial and so genetics are difficult to untangle, although some cancers are directly due to a specific genetic variant and nothing else. Maybe that's where you've been seeing diet recommendations? If you don't look at the actual genes or know how to evaluate them, then you're reliant on whatever weird assessments genetic services sell. Those are typically like "here's your risk of heart disease" and it just assesses a handful of genes that seem statistically associated, often not strongly. It doesn't tell you anything real because it's speculative to begin with, almost never deterministic, and extremely incomplete. Then they'll spit out the conventional advice for heart disease, IBS, whatever, but none of that advice is based on genetics, they're just providing boilerplate advice for whatever condition you expressed interest in. It might even be in direct contradiction to some of the genetic variants found, depending on what they are.
That's completely different than knowing you have a genetic issue with a biotin-dependent enzyme and that you need high dose biotin to function, or you can't metabolize a certain kind of fat, or you struggle to methylate folate, etc. If you try running your genetic data through something like Nutrahacker you get much more specific advice, although you can have genes with contradictory solutions (e.g. you may need to avoid methyl groups for one gene, but you need them for another; a machine can't square that and tell you what to do in reality), and it flags some variants that are probably not important, and for a few genes it does give brief dietary advice (e.g. low fat, or ketogenic, or fiber, etc) which can also conflict with other genes, so it still requires some discernment.
But again, 23andMe or any other partial genome service is not going to cover the vast majority of inborn errors of metabolism. Only a full genome can do that. And right now there's not a great service that can simply scan your full genome and point those out, to my knowledge; you have to manually check for ones that you suspect you might have, and then it can be confusing digging into the research.
Yea what my friend did was exactly what you described. They took her DNA and then recommended she eat the mediterranean diet, basically. I bet they recommend nearly the same thing to everyone. I doubt they'd recommend 90% fat, heavy cream based keto to me lol.
I'm also probably confused by the factor you mention with MTHFR, which is the one I've tested on myself and which is probably confounded like the Japanese vK issue: there are so many ways to get around it that chances are high you never notice the downside. Unless you happen to do one very specific diet maybe, that doesn't have one of the "fixes."
I have supplemented folate and methylated B vitamins based on this MTHFR thing, and over 3-6 months saw no difference whatsoever. (Except my folate shot up in the "too high" range on blood panels.)
That's why I'm very skeptical of many of these mechanistic speculation/explanation type things. I've yet to see somebody miraculously cure diabetes or obesity by diagnosing MTHFR and then supplementing folate.
I guess there are these cases like you have where it makes a day and night difference, but since I haven't seen a fix like that myself based on genetics, I'm not thinking about it much.
You could argue that my keto/non-24 thing is exactly this. We just don't know if it's genetic (AFAIK the cause of non-24 is totally unknown). There are CLOCK genes, but they just determine your chronotype. Non-24, especially the type I have, doesn't seem to be genetic or at least they haven't found the gene.
Long term keto fat cells clearly become very insulin sensitive, so it is easy to gain weight with the small insulin bump from protein despite the liver making ketones to keep the brain buzzing along. Starting keto, the insulin insensitive lose weight rapidly until the body reaches a new homeostasis, with luck that is at 15% body fat, but for many it is much higher.
I believe that this homeostatic level is down to the level of Omega 6 in the body and the only way to get down to 15% is keto with zero Omega 6, or high carb zero fat, which is naturally zero Omega 6, for many years.
So perhaps start with higher carb to begin with. Perhaps extra gelatine to eliminate excessive BCAAs quicker?
But still, exreme diets might cause more damage than good if nutrient deficiencies are being ignored for to long.
Muscle mass should also help to burn excess PUFA.
I am a fully paid up member of the keto club because it seems logical and I believe it to be healthy.
But it's very frustrating when other diets such as slime molds, potato, riff creates weight loss that can't be explained with keto rules.
There must be a piece of the jigsaw missing. Something that will allow fat out of adipocytes that will work irrespective of insulin level and possibly inhibit fat storage.
For one, I'd say, we don't actually know the insulin levels throughout the day of a potato dieter. Maybe their insulin is pretty harmless. That's one big weakness of the "Carb Insulin Model" I think, while the insulin -> fat part is pretty well studied and understood, I think our idea of "carbs == insulin" is quite flawed. It's way more complicated than that. E.g. some proteins seem to be as insulinogenic as many carbs in the few studies we've seen.
Is a potato/rice diet really as insulinogenic as a candy-laden high-PUFA S.A.D.?
Looks like I'm going to have to get a CGM and do some experimenting.
Even that won't show you the insulin :( I've worn CGMs for the better part of the last 2 years.
As a recent carb convert - easily 30% of my carbs are literally candy. Just Red Vines, not M&Ms, Haribo not Snickers. All the HFCS i care to eat, but avoiding candy fat and seed oils.
Haha whatever works :) You are probably right that the mix w/ seed oils (palm oil in candy?) is at fault. Just listened to an Anabology video where he did 1lb of honey a day because honey is pretty much the only pure sugar he can even eat that much of. Apparently pure sugar is not THAT appetizing.
Nice article as always. Good job.
My question is why? Why does keto not work for all?
The makings of a good cake are there :-)
- Low insulin, fat oxidation, relying on your own body fat for energy
I think that the missing pieces are
1) We cant match fat oxidation with fat intake as carbs do. "Chronic undernutrition and metabolic adaptation". And I don't think metabolic rate goes-up in the way Ben and team describe.
2) I believe that Jay has got it right in his recent videos. There are blockages in the system that direct substrate towards fat storage and not usage. And this might be even in the setting of low insulin, high ketones etc. Maybe it is about the microbiome. That would make sense to me.
3) Muscle and exercise are hugely under appreciated.
More interestingly is how do we measure if the diet is appropriate for the person. For me
1) Body composition (not weight)
2) Energy levels ( if you feel good just give it time)
3) Stress as an indicator of lack of energy/high demands (TSH , DHEA, Cortisol, Super High Cholesterol?, high BG maybe a lot more)
- I wonder if a high-protein, high-PUFA diet is actually low insulin for everyone. We can't measure insulin well (lab tests only) so we have no clue. Maybe, in me, beef is incredibly insulinogenic and that's why it makes me fat?
To your questions:
1) Does Ben (Bikman?) say metabolic rate should go up? On keto? Haven't seen that particular thought from him (more from Peaters)
2) Jay Feldman? What you describe sounds a lot like "fuel partitioning" which I'm a huge believer in. I think insulin is clearly one mediator of fuel partitioning, but it might not be the only/key one. I.e. maybe I have very high insulin from protein and we just don't know, or maybe I have low insulin but something else mediates the fuel partitioning.
3) I don't know, I think it's overestimated. I just don't know anyone who's had any success losing fat or fixing diabetes with resistance training. You can burn off the carbs you eat with endurance exercise, but that's sort of a band-aid over your diabetes, I think. I do know people who do that.
I agree on your measurements, although for most people weight is a decent, easy to measure proxy. None of us wake up with 30lbs of extra muscle, so if you factor in water retention/glycogen or don't make drastic changes, weight is easy. Of course check in w/ a DEXA once in a while or similar. Although to be honest, you can tell your body composition in the mirror if you're honest. Nobody accidentally gets to 9% body fat and 220lbs of lean muscle. The people who are these "weight/BMI outliers" put a decade of work into it and they know who they are :)
1) not really Ben Bikman but he has done shorts on this -https://www.bmj.com/content/363/bmj.k4583
2) May you do instead of a OGTT an PTT ?? :-)
3) When referring to a diet working or not, I was mostly thinking of the how you can identify possible root causes. If you are on keto and thyroid gets messed-up maybe that is a root cause etc.
Ha the thing is, testing for insulin is insanely annoying since you have to do it in a lab. I spent 6h in the lab that day for the Kraft. Too much effort to repeat it a lot, unfortunately..
3) yea agreed, that could happen. Hasn't seemed to happen in my case, but for sure easy & good to check regularly. Thyroid panel & T panel is very cheap. Just do it quarterly or whatever.
Have you ever considered thyroid hormones? Broda Barnes and of course Ray Peat both approved the approach.
Epigenetically our bodies are of worse quality compared to our ancestors. Perhaps an extra hormone boost would help your efforts.
Kind regards
I think Barnes and Peat are wrong about thyroid. What's going on is most likely not "hypothyroidism" even if it has similar symptoms.
I've gotten my thyroid tested and it is not only fine, it is insane. I have lower TSH than all the Peaters. T3/4 are fine too.
Thyroid just as any other organ can malfunction due prolonged years of bad habits.
Why do you think both are wrong? Any particular part?
One thinks its epigeneticaly, the other thinks its more our environment, diet included. If you consider Pottenger's cats, both probably are right.
Our body is just like a cell in a malfunctioned environment, without proper signals it can't work properly.
I just don't see that many people with messed up thyroid markers. They do exist for sure, but e.g. mine are actually very hypERthyroid in parts. I've always had high body temperature since childhood, too.
I have read that thyroid hormone in excess is very catabolic which would make that not a smart thing to do.
For sure as with most things, there is a goldilocks zone. Temperature and pulse should be a good indicator.
I think you did Keto wrong. AKA, you did "internet keto" but not "prescription-strength keto" (limiting fats). Check out Dr. Eric Westman, Duke University and Amy Berger. Authors of "End Your Carb Confusion"
It's not an "unlimited fat" diet, if you're trying to burn your own body fat, that is. a keto meter's not going to show you if you're burning your own body fat vs. burning the fat you just consumed.
You have to count how much added fat (high calorie foods like mayo, cream, butter, etc.) you use per day. Proteins are unlimited though. I'm doing their program now.
Tons of his blog posts are higher protein experiments in which he reliably gains weight. While it's true that most of his blood ketones are likely dietary and not from stored fat, if he gains weight on both increased carbs and increased protein, that leaves him with eating fat to function.
Yea the high-protein/low-fat "keto" diets just aren't sustainable for me. It's basically an inefficient fast; I tend to get starvation symptoms and I have to stop, and then it all comes back just like with the fasting.
"prescription-strength keto" is limiting fat? The classic ketogenic diet is 90% fat by calories, which I am roughly on.
I would argue that high-protein, low-fat keto is "internet keto."
I guess the diet you describe is a version of keto and it can be prescribed for weight loss, but I have tried higher protein before and it never worked. It just makes me insatiable. I basically have to limit protein to the extreme to reach any level of satiety.
Yes!
Dr. Eric Westman is as Duke University, he took over for Dr. Atkins (kind of...he uses Dr. Atkins food list - the list Dr. Atkins used for 30 years & now Dr. Westman has been using it for 20 years, so 50 years total! I am using Dr. Westman's/Adapt Your Life Academy's way of eating, and I'm having a great result!! I'm especially happy 2day because I'm down another jean size, and bra size, PTL!
Per Dr. Westman - I'm paraphrasing what he said - the hallmark of a ketogenic diet is: to keep the total carbohydrate LOW (note: not "net carbs" even though I think Dr. Atkins invented the concept of "net carbs"? We don't count net carbs on the Keto Made Simple (KMS) approach). In "Phase One," the intro phase, which I'm doing now, the carbs are kept under 20 grams - that way 100% of the people doing the class are in "nutritional ketosis." Per Dr. Westman, if you go up to 50 grams per day of carbohydrate, only 50% of people will be in nutritional ketosis.
To be honest, at this point, I'm not just doing it for weight loss - I'm not overweight for my height (I'm 62 kgs at 5'7) - but the ketones are changing my life *for the better*!!! I have some chronic pain issues from prior back injuries (herniated disc) and being in nutritional ketosis is like magic for my inflammation - the closest I think I'll ever get to a miracle cure.
Two books you could read: "End Your Carb Confusion" by Eric Westman, M.D. and Amy Berger and
"The Art & Science of Low Carbohydrate Living" by Volek and Stephen Phinney M.D. PhD
Also, feel free to check out my blog iusedtobeatrackstar (dot) blogspot (dot) com
So, just to reiterate. the hallmark of a ketogenic diet is low carbohydrate - low carbohydrate is what puts us into nutritional ketosis.
You could make an appt with Dr. Westman, or Dr. Tro, or someone.
With carbs equal, less protein leads people to have higher ketones than more protein though.
Well I've been in nutritional ketosis for about 9 years now, with carbs from close to 0 to 50g or so, verified by ketone meter.
I think I know pretty much everything about keto there is to know.
10000 Kcal/day Carnivore
https://vimeo.com/1021683428/fbf7af119e?share=copy&utm_source=convertkit&utm_medium=email&utm_campaign=%F0%9F%93%8D%7BAYLA%20WEEKLY%20NEWS%7D%20%E2%86%92%20Dr.%20Westman%20Reacts%20+%20emotional%20eating%20-%2015403152
Haha tl;dr?
the ai generated image is too good
strong KERTOGERNIC DIET FER OBERSERTEEE vibe
Here's a Ben Bikman vid where he addresses raised metabolic rate tied to low insulin therefore Keto. Start at the 25 min. mark: https://www.youtube.com/watch?v=SUfOrnzVVw0