I’ve been eagerly awaiting this study. It was a 12-month randomized controlled trial comparing non-calorie-restricted low-fat and low-carbohydrate diets. Both diets placed a major emphasis on diet quality. I’m not going to review the findings in detail because Examine.com has already beat me to it. But here are some highlights:
- Participants achieved large differences in fat and carbohydrate intake for the duration of the 12-month trial, although the difference narrowed somewhat over time (as it does in every diet trial).
- The first primary (most important) outcome was about insulin. Genes influencing meal-related insulin secretion were unrelated to a person’s response to the low-fat or low-carbohydrate diets. [Correction 2/21/18: These genes aren’t directly related to insulin; they’re more closely related to fat metabolism and previous preliminary evidence suggested they might impact the response to low-fat/carbohydrate diets]
- The second primary outcome was also about insulin. A person’s baseline insulin response to carbohydrate was unrelated to his/her weight loss response to the low-fat or low-carbohydrate diets.
- Both diets caused similar weight and fat loss at 12 months (-5.3 kg [-12 lb] for LF vs. -6.0 kg [-13 lb] for LC). These are better 12-month results than most diet trials.
Here are a few reasons why this study is a big deal:
- It was funded by the Nutrition Science Initiative, a nonprofit involving Gary Taubes that has been funding additional research to test the carbohydrate-insulin hypothesis of obesity. This is the second of three studies NuSI has funded. The first study showed that a very-low-carbohydrate ketogenic diet doesn’t accelerate fat loss or meaningfully increase metabolic rate relative to a high-carbohydrate, high-sugar diet of equal calories and protein, despite large differences in insulin levels (implying that ketogenic diets promote weight loss by causing people to spontaneously eat fewer calories). So far, these studies are doing a remarkably effective job of further refuting the carbohydrate-insulin hypothesis.
- It was large and well-designed. In included 609 people and the research plan was registered in advance, including which outcomes would be the most important (primary). This guarantees that the researchers couldn’t tweak the methods after the fact or use selective outcome reporting to obtain a preferred outcome.
- It didn’t involve deliberate calorie restriction in either group. One of the (valid) complaints of the low-carbohydrate community is that in randomized trials, non-calorie-restricted low-carbohydrate diets are often compared with calorie-restricted low-fat diets. Some have speculated that if low-fat dieters were allowed to eat until full rather than controlling portion size, they wouldn’t lose weight, and they might even gain. This is not the case.
- It focused on diet quality in both groups. All participants were told to focus on unrefined foods and cook at home whenever possible. This is in contrast to the wimpy low-fat control groups in many randomized trials of low-carbohydrate diets, which I call the “Snackwell diet”.
One typical (somewhat valid) critique about randomized trials like this is that the low-fat diet contained too much fat, or the low-carbohydrate diet contained too much carbohydrate. In this study, both arms were started on very-low-fat (20 g/d) or very-low-carbohydrate (20g/d) diets, and told that they should keep fat/carbohydrate to the lowest level they can sustainably maintain. While it is true that fat and carbohydrate intake rose throughout the 12-month period, this isn’t an indictment of the study: It’s a fact about human nature. This trial showed that even with extensive support, the average person is unable to stick with a very-low-fat or very-low-carbohydrate diet. It is true that they probably would have seen better results on both diets if they had been able to stick with a greater degree of restriction, but the participants demonstrated with their forks that it wasn’t sustainable for them.
Sugar intake declined substantially in both groups, although more in the low-carbohydrate group. I predict that those who are constitutionally unwilling to give up the insulin hypothesis will focus on this as a way to discount the results. Please allow me to inoculate you against this weak argument with the following facts:
- Sugar has little impact on insulin levels relative to other types of carbohydrate unless it’s consumed in unusually large amounts.
- High sugar intake doesn’t seem to impair weight loss, both in settings where calories are strictly controlled, and in settings where they aren’t.
Sugar probably does contribute to obesity, but the evidence suggests it’s very unlikely to be the primary driver of changes in body fatness. I have no illusions that this study will change the minds of the diehards, but I do expect it to impact everyone else.
https://examine.com/nutrition/low-fat-vs-low-carb-for-weight-loss/
Nice post!
Brilliant. Thanks for keeping us enthusiasts informed. Without experts like you, where would we be?
If it is uninteresting, people will lose weight. Book idea: the Tastes Like Cardboard Diet [TM].
I lost 30 pounds in a couple of months on a high sugar diet. But the sugar was in the form of raw fruits and vegetables, and I was eating everything without additional flavors. No salads or spicing. To eat apples, eat apples. To eat oranges, eat oranges. To eat lettuce, eat lettuce. To eat honey, eat honey. No Waldorf salad. No apple pie. No sticky buns.
But lots of total sugar.
Lost too much weight. Had to stop.
Carl, this is my experience as well. The less I alter the food, the more satisfied I am. I’ve been developing this model for quite some time. Its been great to discover Stephan Guyenet and learn that my anecdotal experience has scientific basis
At the end of the 12 months, those on a low-fat diet reported a daily average fat intake of 57 grams; those on low-carb ingested about 132 grams of carbohydrates per day….
this is low carb ?? …. ridiculous study
Hi Mike,
I know some people just need a way to dismiss this study’s uncomfortable findings, but you might want to consider this. Those in the low-carb arm were eating 247 g of carb at baseline. This dropped to 97 g at 3 months, then gradually crept up to 132 g at 12 months. Even at 12 months their carb intake was nearly cut in half. Furthermore, this was the maximum degree of restriction these people found sustainable. This is a study of how diets work in real life when people are given good support.
I’m honestly unclear about why anyone should find this study uncomfortable, except for a few scientists with pet theories. As I see the results, this is good news for just about everyone. Whole food lfhc diet is safe and effective. Whole food lchf diet is safe and effective. Stick to whole foods and you are fee to use whatever macronutrient ratio is personally most comfortable and effective. Am I missing something?
I mostly agree Jodie and I don’t think you’re missing anything. The only caveat I would add is that one could debate the word “effective”. The average person went from ~213 lbs to ~200 lbs. It’s a significant improvement (and pretty good for a 12-month outcome) but not enough to reverse obesity.
Very interesting.
I emailed Gary Taubes not long ago, offering him information I thought might interest him. He was quite friendly. I sent him two articles. The first showed that insulin does not inhibit satiety (POMC) neurons as his friend Lustig believes, but activates them, and previous results are likely due to contamination of insulin preparations with zinc, which inhibits them. You will remember that paper. It confirms what obesity researchers believe about insulin and satiety of course, and goes a long way towards proving Taubes and Lustig wrong.
You will remember the second article too. It’s the one about fat breakdown being dependent on copper, and suggests it isn’t insulin that prevents fat breakdown as Taubes believes, but copper deficiency.
Mr Taubes told me he would stick with the insulin paradigm. I suppose the poor man is trapped.
I saw that article of his in the BMJ about sugar and fatty liver. Fatty liver is likely due to copper deficiency. I can’t imagine what the BMJ is thinking.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4183666/
https://www.sciencedaily.com/releases/2016/06/160606200439.htm
https://www.nature.com/articles/ajg2010170
All well and good, but don’t you see that you are just shifting the reductionist nutritional scapegoat? I don’t want to suggest that there is no role for such mineral imbalances in the obesity pandemic or that the studies you cited don’t warrant further research, but both experience and the weight of the evidence tell us that we deal with a complex and multifactorial phenomenon related to fundamental chages in our environment/food culture and that it would be foolish to point at any single factor as its main cause.
@Timar
“don’t you see that you are just shifting the reductionist nutritional scapegoat?”
No.
“we deal with a complex and multifactorial phenomenon”
Yes indeed. It requires much study of the scientific literature. I have spent more than 30 years doing just that. How long have you spent?
“it would be foolish to point at any single factor as its main cause”
Not if a single factor really is the main cause. That single factor is oxidative stress. We eat a diet which is overloaded with the main pro-oxidant nutrient (iron) and deficient in antioxidant nutrients such as manganese, magnesium and copper. The most important one is manganese, which is not only removed during refining of carbohydrate foods, but is very low in animal foods. Lifelong marginal deficiencies of these three minerals can explain the epidemics of degenerative disease we see today.
If you have evidence or arguments to the contrary, please present them.
I would like to read some of your research articles from your 30 years of research into the role of oxidative stress and minerals in obesity. Where can I find those?
@Bruce
As I expect you have already found out, I have not written any papers since 1981, when the last of my PhD and postdoc work was published
My research was on tissue regeneration, which is part of the same phenomenon as maintenance and repair, the process which keeps us healthy. Our bodies are constantly breaking down old components and replacing them with new ones.
When my experiments were finished, I spent several years in the Oxford biochemistry department library reading current journals in biochemistry, physiology, molecular biology, cell biology, neurobiology, immunology and genetics. .I found out that many of the enzymes I thought would be important in regeneration were activated by manganese or copper. Then I read about the astonishingly healthy Hunza, and McCarrison’s experiments showing their health was due to their diet.* They ate no refined carbohydrate and very little meat. I concluded modern disease was due in large part to mineral imbalance caused by a diet based on meat and refined carbohydrate.
I tried to tell my Oxford college what I had found, and my fellowship was terminated. I was considered insane. Fortunately I had inherited some money and could continue my reading.
You asked about my ’30 years of research into the role of oxidative stress and minerals in obesity’, but that is not really what I have done.
*I adopted the Hunza diet at this time, and the food cravings which had been making my life a misery disappeared, never to return. Food cravings are arguably due to oxidative stress in the hypothalamus (POMC neurons) and midbrain (dopamine neurons).
What exactly is the Hunza diet jane?
There appears to be a summer diet and a winter diet, with lots of grain, root vegetables, dairy and some meat. All sheep and goat.
http://biblelife.org/hunza.htm
@thhq
Groan. Not that execrable Biblelife article again. Didn’t you notice it doesn’t mention McCarrison?
http://journeytoforever.org/farm_library/Wrench_WoH/WoHToC.html
The diet given to the rats was chapattis, or flat bread, made of wholemeal wheat flour, lightly smeared with fresh butter, sprouted pulse, fresh raw carrots and fresh raw cabbage ad libitum, unboiled whole milk, a small ration of meat with bones once a week, and an abundance of water, both for drinking and washing. [the Hunza diet also includes fruit]
In this experiment 1,189 rats were watched from birth to the twenty-seventh month, an age in the rat which corresponds to that of about fifty-five years in man. The rats were killed and carefully examined at all ages up to the twenty-seventh month of life by naked-eye post-mortem examination.
The result was very remarkable. Disease was abolished. This astonishing consequence, however, must be given in McCarrison’s own words in the first of two lectures given at the College of Surgeons in 1931.
“During the past two and a quarter years there has been no case of illness in this ‘universe’ of albino rats, no death from natural causes in the adult stock, and, but for a few accidental deaths, no infantile mortality. Both clinically and at post-mortem examination this stock has been shown to be remarkably free from disease. It may be that some of them have cryptic disease of one kind or another, but, if so, I have failed to find either clinical or macroscopical evidence of it.”
By putting the rats on a diet similar to that of certain peoples of Northern India, the rats became “hunzarised,” that is they “enjoyed a remarkable freedom from disease,” words used by McCarrison in 1925 of the Hunza. They even went further. Except for an occasional tape worm cyst they had no visible disease at all.
This month-long Hunza Diet experiment is interesting.
https://www.lifehacker.com.au/2016/01/12-diets-in-12-months-the-hunza-diet/
Dropping breakfast doesn’t sound too good to me. And I don’t like the weight gain part either.
After reading about the historic Salish diet and culture , I’m skeptical of diets producing disease free humans jane. There are too many other factors in play.
The rat diet you describe sounds about like what I’d expect the Hunza diet to be. Lots of whole grains and legumes, dairy, a little meat and root vegetables, and some cherries and apricots. It’s not a bad diet. But after reading lifehacker’s account it would be hard to stick to the diet long term unless you were diligent about it.
A problem is that the likely high amount of sweating is not considered in relation to healthy population groups like the Hunza.
Nowadays many people spend almost all their time in perfectly comfortable temperatures resulting in minimal sweating, i. e 100 ml or less over 24 hours. Sweating can be a way to excrete large amount of salt, but also potassium and byproducts from protein catabolism. Without sweating the kidneys must handle this by themselves. Interestingly this article https://www.tandfonline.com/doi/full/10.3109/0886022X.2013.832318 suggests that for people with kidney failure, hot bath treatment may almost be as effective (and possibly superior in some ways) to dialysis. Also it mentions that during hot baths, the body can excrete as much as 2000 ml sweat per hour, superior to 1300 ml for sauna.
https://www.ncbi.nlm.nih.gov/books/NBK236237/ suggests that athletes walking for 5-7 hours would sweat about average 250 ml per hour in normal environment, 700 ml in hot wet environment and 1200 ml in hot dry environment (https://www.ncbi.nlm.nih.gov/books/NBK236237/table/ttt00009/?report=objectonly). How much would the Hunza have sweated in the summer at 30+C over 24h doing manual labour, farming etc (no air conditioning obviously)?
Sweat may supply 700-1500 mg sodium per liter (wikipedia), and this can make studies measuring only 24h urine as a proxy for salt intake highly misleading. It may in fact be that exercise (in so far as it increases sweating) would be associated with *less* sodium in urine while a lack of exercise means more sodium in the urine. Indeed it seems that taking sauna can be as effective to lower blood pressure as cutting salt intake.
It is unfortunate that commonly prescribed diets for weight loss can be extremely nutrient dense relative to calories with large amounts of low fat dairy, whole grains, legumes and lean meat/fish. Nuts are not so nutrient dense though. The Hunza diet would also be nutrient dense but probably much better still and likely with a more preferable macromineral balance. But unless we live exactly like them it may not be optimal.
Studies have suggested exercise as superior to dieting too. A reason could be that exercise involves sweating and leads to less muscle loss: https://mobile.nytimes.com/2018/03/05/well/move/heart-disease-exercise-weight-loss.html
It is also a problem that *all* research animals differ from humans in that they do not use clothes. Humans use clothes as a means of creating very comfortable temperatures and hence minimize calorie loss (this in addition to cooking would have given us an advantage over other species evolutionary). I would also assume that temperatures in the labs where these research animals dwell are now closer to a constant around 20C-ish temperature, but it wasn’t like that a hundred years ago. But 20C for a mouse isn’t the same as for a human: https://www.ncbi.nlm.nih.gov/m/pubmed/29122558/
guiseppe at 2500m elevation they’d be focused on staying warm most of the time. Right now it’s minus 11C in the Hunza valley.
https://www.worldweatheronline.com/lang/en-us/hunza-weather-averages/northern-areas/pk.aspx
30C is a rare day there. Their valley is higher than a normal mountain peak.
The growing season is short. Tender greens and vegetables are out of the question.
Regarding the health benefits of living at 2500m I found this interesting.
https://www.denverite.com/coloradans-age-better-10611/
You have to be acclimatized to see the health benefit of altitude. Given the thousands of years the Hunzas have occupied their valley, and the 30% infant mortality rate, there has probably been some natural selection to favor survival. Stronger hearts and larger lungs for instance.
https://en.m.wikipedia.org/wiki/High-altitude_adaptation_in_humans
thhq
Don’t know much about the Hunzas and the climate there, just looked briefly at accuweather. It suggests the historical temperature for today is 12C day and 2C night. It’s much colder than normal now. For June, July and August accuweather suggest 29-31C daytime and 14-18C night temperature. I’m not sure if these figures are very accu-rate. Also as I mentioned, dryness vs humidity would matter regarding sweating. In any case also cold temperatures can induce beneficial health effects such as in relation to diabetes: https://www.nature.com/articles/nm.3891
I believe the change in ambient temperatures to near constant all the time, and less time spent outdoors, have contributed to the obesity epidemic and many other health problems.
Elevation is really interesting. I think a study suggested that in the US life expectancy difference was 2-3 years longer for those living at higher vs lower altitude.
guiseppe the Hunza climate is a lot like lower elevation Alaska, Norway or Finland. Growing seasons are short, so root and cruciferous vegetables, hardy tree fruits and berries are all you can count on. I’ve seen tomatoes grown at 1500 meters, but they’re tough skinned little things.
The cold climate is great for growing giant cabbage.
http://www.amusingplanet.com/2015/10/alaskas-giant-vegetables.html
One of the other things I looked into was the tendency of hucksters to pick up on elements of the Hunza diet and distort them. One of these concerns the health benefits of apricot pits. The other is distorted longevity claims, which the Hunzas themselves had a habit of doing.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3062986/
A final part for consideration is the 8000 foot elevation of the Hunza valley. A highly active population, living on the verge of starvation much of the time, in a low oxygen atmosphere….these are factors which would make them extremely hardy without regard to their diet.
@thhq
The Hunza did eat breakfast.
Your link says
“The Indigenous Hunza people in the hills of Pakistan don’t eat until lunchtime, existing on a mostly raw diet of fruit, nuts and seeds with a little yoghurt thrown in. There is one cooked meal per day — dhal with wholemeal chapattis.”
The Wheel of Health says
“The daily eating is given by Schomberg as: nothing before going out in the early morning to the fields; after two or three hours of work, bread, pulses and vegetables with milk; at midday, fresh fruit or dried apricots kneaded with water; in the evening these same foods, with meat on rare occasions.”
In other words, your link consists of the usual misrepresentations and half truths. Nobody wants the Hunza to have been as healthy as McCarrison found them to be. Or if they were, it wasn’t their diet.
Please tell us what the Hunza diet consists of Jane, and the sequence that you eat it in. McCarrison’s rat diet sounds fine to me, but it doesn’t appear to be much different from following the USDA Food Pyramid.
If I labored for 2-3 hours in the AM before eating I would be skipping the meal I call breakfast. And I would be cranky.
The study is a huge confirmation for the fructose/sugar hypothesis, while at the same time refuting the much too broad carbohydrate hypothesis. The carb hypothesis was never realistic anyway, as so many healthy traditional diets all over the world have been high in carbs. What changed is really sugar/fructose content. (The links provided by Stephan don’t refute this at all). With low sugar, you can eat high carb or high fat, doesn’t matter, you will stay healthy literally from teeth to toe!
Hi Samuel,
This study doesn’t support that hypothesis. Table 2 reports the sugar intake of the two groups over the 12-month period. From baseline to 12 months, the low-fat group reduced sugar intake by 33 percent. The low-carb group reduced sugar intake by 56 percent. There was no difference in weight loss.
The previous NuSI study demonstrated that under calorie-matched conditions, a high-sugar diet allowed faster body fat loss than a low-carbohydrate ketogenic diet.
As I’ve said, I do think sugar contributes to obesity and cutting it out can help with weight loss and maintenance, but there is really no reasonable interpretation of the evidence that suggests it’s the dominant factor that determines health and body fatness.
There was no (sig.) difference in weight loss, because both diets were relatively low in sugar. If you’re low in sugar (excluding fruit), you’re fine, no matter how many carbs or fat you eat. (And you won’t get cavities either, which is the first sign of an obviously bad diet).
The first Nusi-study didn’t refute the sugar/fructose hypothesis either, because in practice it was somewhat hypocaloric [1], and even the high-carb diet contained very little added or liquid sugar [2], and the sugar it did contain was mostly glucose and maltose (which is fine), and the little fructose it contained was mostly from fruit (also fine). Take a look at the high-carb baseline menu of the first Nusi study below (from the sup. material, which almost nobody checked).
The first Nusi study did however likely refute the “processed food” hypothesis. What really changed in Western diet is not the bread and noodles, but the amount of added sugar/fructose. The second Nusi study now confirmed this.
They major flaw of the first Nusi study was that they simply said “25% sugar” but didn’t mean sucrose/HFCS at all, but instead simple carbs (glucose/maltose) and fruit. This is a whole different story, which almost nobody realized back then. They should have determined total added fructose, but they didn’t.
Quoting the 2016 Nusi study [2]:
“Both the BD and KD menus contained minimal quantities of processed food, and,
despite the large differences in macronutrient composition and sugar content, the BD did not include large quantities of added or liquid sugars. In that regard, the BD may have differed somewhat from the customary diets of these subjects. Sample menus for 1 d of each diet are included in the Supplemental Materials.”
Sample baseline diet menu [2]
Breakfast: Egg and potato hash with berries
Morning Snack: Peanuts, oil roasted, salted; Chewy granola bar, chocolate chunk, low fat; Lemonade, prepared from frozen concentrate
Lunch: Turkey burger with hot potato salad
Afternoon Snack: Pretzel sticks, Wheat crackers & Cheese, American, spread, processed
Dinner: Cheese steak sandwich and pineapple
[1] https://wholehealthsource.blogspot.com/2016/07/nusi-funded-study-serves-up_6.html
[2] https://academic.oup.com/ajcn/article/104/2/324/4564649
ps: I wrote “no matter how many carbs or fat you eat”. Of course I meant no matter the carb/fat composition, not the total amount.
By the way, Robert Lustig showed a few years ago in his pilot study with obese children, that if you feed them a high-carb junk diet (pizza, hot dogs etc.) but eliminate the added sugars (sucrose/HFCS), their liver and blood values improve within 10 days, unrelated to calories!
https://www.ucsf.edu/news/2015/10/136676/obese-childrens-health-rapidly-improves-sugar-reduction-unrelated-calories
It amazes me how the authors of the first Nusi study could ever call their baseline diet “high sugar” when it contained almost no added sugar (the exception being the one granola bar and one serving of lemonade). It simply was a high carb diet (including glucose and maltose “sugar”/simple carbs), both whole and processed food, which of course is perfectly fine.
Hi Samuel,
Weight loss can be produced by cutting almost any major source of calories out of the diet. It doesn’t have to be sugar. Vegan diets cause weight loss, gluten-free diets cause weight loss, even high-sugar fruitarian diets cause weight loss. https://www.ncbi.nlm.nih.gov/pubmed/4928686
Also, refined sugar isn’t very fattening in animal models.
Why do you make a distinction between total sugar intake and added sugar? Making that distinction suggests that you don’t actually think the sugar per se is responsible, but rather that how it’s packaged is more important (calorie density, nutrients, and/or fiber). If that’s the case, then it doesn’t make a lot of sense to only focus on the packaging of the sugar, but rather the packaging of all the food that was eaten. i.e., the fact that the entire diet was composed of unrefined foods.
Attributing the entire effect of the Gardner study to sugar is simply fitting the study’s results to your own preexisting beliefs.
Hi Stephan,
weight loss: exactly, that’s why I noted NUSI 1 was slightly hypocaloric. But more importantly, the NUSI 1 high-carb diet was very low in fructose-based sugars. It wasn’t a “high-sugar” (sucrose/HFCS) diet. This is the key point most researchers overlooked.
animal models: don’t work for fructose/sugar studies, because rodents still possess uricase enzyme to metabolize fructose-derived uric acid, thus countering fructose-induced insulin resistance (and other effects). If you block uricase in rodents and feed them fructose/sugar as in a Western diet, you get MetS in no time. See the many excellent studies by Richard J Johnson.
total vs. added sugars: no difference. The metabolic difference is fructose (including sucrose and HFCS) or not (glucose, maltose, lactose). Added sugars are almost always fructose-based (sucrose, HFCS), thus dangerous. Whole foods are mostly low in fructose, apart from some fruits, but whole fruit counters the metabolic effects of fructose (starting with cavities). In sum, the real risk is added (fructose-based) sugar.
All I say is both NUSI 1 and 2 perfectly support the sugar/fructose hypothesis. They both clearly refute the carb hypothesis, and NUSI 1 (as well as the Lustig trial) also refute the “refined carb”/”processed food” hypothesis (as long as they are low in fructose).
I find it a bit misleading to conclude from Lustig’s study that it’s necessarily all about fructose. Replacing sucrose with refined grains means replacing a completely empty calorie with a food that provides vitamins, minerals, protein etc. Additionally grains are typically fortified with various vitamins including folic acid. The latter seems to help with fatty liver (https://www.ncbi.nlm.nih.gov/m/pubmed/28705466). It would have been more interesting to see what would have happened if two groups were given the same diet but one supplying for example 30% of energy as sucrose and the other 30% as maltose (a disaccharide with two glucose molecules).
For ad libitum dieting to cause weight loss, calorie intake has to be reduced somehow. The intentional suppression of added sugar is one effective way to do it. The weight loss in this study is what you would expect for a sustained 100 kcal/day reduction in eating. About 2/3 of a can of regular Coke a day.
I have found over 11 years that my counting compliance has shifted. I do not count calories with the accuracy that I did when I was losing 1 kg a week. In becoming casual with the food counting I undercount, and to maintain weight I need to stay at a calorie deficit of 100-200 kcal/day. My exercise counting is a more accurate than my food counting, but in similar fashion I round off the miles in my favor. I naturally gravitate towards eating more and doing less, but I know how to control it to maintain weight.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5274550/
“Notably, the study was not designed to identify which of the two study diets was the one best for weight loss, but rather, which diet was best for which individuals for weight loss.”
It appears to me that your analysis does not address the core research area of this study. To me, the graph of weight loss among each diet’s participants pretty clearly demonstrates that there was perhaps 10% of the group that did very well with their assigned diet and 10% that did very poorly. Determining what characteristics lead people to do better or worse on these diets is the real question, and it seems that this study was intended to primarily drive additional research.
So I just read the full study text including supplements. As I suspected, and as was already the case with the first NUSI study, both diets – including the high-carb diet – were very low in added sugars. Quote:
“Both diet groups were instructed to (1) maximize vegetable intake; (2) minimize intake
of added sugars, refined flours, and trans fats; and (3) focus on whole foods that were minimally processed, nutrient dense, and prepared at home whenever possible.”
What the first and second NUSI study really showed is that a diet low in added sugars (sucrose/HFCS) is a healthy diet, regardless of carb/fat composition, and regardless of “refined carbs” (e.g. bread, pasta, etc.).
This is exactly what Robert Lustig previously found when he fed obese children a high-carb, low-sugar “junk” diet (pizza, hot dogs etc.) and their weight, liver and blood values immediately improved (https://www.ucsf.edu/news/2015/10/136676/obese-childrens-health-rapidly-improves-sugar-reduction-unrelated-calories)
It is crucial to realize that already in the first NUSI trial, the high-carb diet contained almost no added or liquid sugars. The sugars it did contain were mostly glucose/maltose, that is simple carbs, no fructose/sucrose/HFCS (see my previous post).
In other words: NUSI 1 and 2 both vindicated Lustig. Gary Taubes also realized his error and switched from the carb to the sugar/fructose hypothesis a few years ago. The carb hypothesis was never tenable anyway, since there are so many healthy traditional high-carb societies all over the world. Of course, they are all very low in added sugar/fructose.
Excuse me Samuel but it’s not the case at all. I can guarantee you that sucrose is not the culprit for the rise of obesity. I mean, it’s obvious.
just have a look.:
PLENTY OF WHITE SUGAR and huge weight loss and diabetes cured
https://www.drmcdougall.com/2013/12/31/walter-kempner-md-founder-of-the-rice-diet/
and the FIRST STUDY OF NUSY found that with 25% OF WHITE SUGAR metabolic rate increased.
Garty taubes is smart man and was very nice when I have asked him for stuffs. But he was wrong. Period.
you’re wrong: NUSI 1 was NOT 25% white sugar, but 25% total “sugar”, and most of it was glucose/maltose (= simple carbs) in whole foods , not sucrose/HFCS. In other words, NUSI 1 was not high sugar at all.
The rice diet was never reproduced. not scientific.
Samuel,
In the UK sugar intake has decreased substantially over the past fifty years (https://www.czarnikow.com/sites/default/files/uk_per_capita_online_0.jpg), yet obesity rates among adults have gone from just around 2% to almost 30%.
In countries like India and China the rate of diabetes are now as high as in the US. According to https://www.washingtonpost.com/news/wonk/wp/2015/02/05/where-people-around-the-world-eat-the-most-sugar-and-fat the per capita intake of added sugar is just 5 gm in India and 15 gm in China, compared to 125 gm in the US. In India average BMI is quite low.
I suspect the uric acid issue from fructose can largely be alleviated by potassium, i.e. a more «alkaline diet» (meat and salt also seems to increase uric acid). But I’m not sure that whole grains, being net acidic, is ideal in this situation, even though it is a «whole food». It may be better to replace them with potassium rich potatoes/tubers. Kempner’s rice diet was net alkaline being extremely low in salt, phosphorus and protein, but moderately high in potassium (approx 3 gm/day was found to be excreted in the urine/24h). However it wasn’t necessarily overly rich in fructose. Average was 100 grams of added sucrose/dextrose, but up to 500 grams was used in some cased. Additionally 700-1000 gm of fruit/fruit juice was taken. Of course dextrose is only glucose, so it is difficult to know how much fructose they obtained. Kempner’s results have been published in respected medical journals. I don’t know any studies having refuted his findings.
I think there’s too much focus on sugar and «refined» carbohydrates and too little focus on the phosphorus/phosphate additives, and also salt and sodium additives that so often accompany these foods. Phosphoric acid found in coca cola may also increase uric acid levels for example, and it also has salt added.
It is estimated that in the US, average intake of phosphorus from such additives is now over 1000 mg/day. This is twice as much as 20-30 years ago. Unfortunately the issue is largely ignored, similarly I suppose as trans fats 50 years ago (the phosphate additives are sometimes called «the new trans fats»). I think they play a role in the overeating/obesity epidemic. https://www.washingtonpost.com/lifestyle/wellness/why-phosphate-additives-will-be-the-next-taboo-ingredient/2017/03/29/7dd3247a-02cf-11e7-b1e9-a05d3c21f7cf_story.html
yet a 2014 study shows clearly that mortality is minimal at high salt intake (5g/ day). Potassium mortality flattens at 2 grams. Very surprising results. My guess: given that sodium helps with insulin, and good stomach acid cover for a lot of dietary shortcomings, high salt is not an issue. We are truly a society were health issues are dominated by insulin, and possibly by bad digestion, which has also the secondary effect of reducing nutrient absorption.
For uric acid, citric acid in the bloodstream really helps (that is, citrus in the diet). Plenty of citric acid is produced in mitochondria, but does not get out. You will never get gout if you eat one lemon per day.
glib,
If it could be shown that for example a 1:1 potassium to sodium ratio was better than 2:1, I would agree, but this is not the case (see for example https://www.mdedge.com/familypracticenews/article/36494/cardiology/high-sodium-potassium-ratio-ramps-mortality-risk). Furthermore replacing sodium chloride with potassium chloride seems to offer major health benefits, especially for those with high blood pressure. I found it interesting though that human milk is quite low in sodium but relatively higher in chloride (sodium chloride has a 1.55 chloride/sodium weight ratio while in human milk it seems to be around 3). The potassium to sodium ratio is typically 3:1 in human milk. Maybe any benefits to salt is more related to chloride than sodium, and as you suggest a reason could be that chloride is needed to produce stomach acid. But why then can carnivores that according to what I read somewhere secrete 10x more stomach acid than humans, very low pH needed to digest large amounts of raw meat, manage with so little salt? From what I just read the potassium requirement for dogs and cats is 2000-3000 mg/2500 kcal while for sodium it’s just 200-300 mg (in other words a 10:1 ratio).
It is also worth considering that salt additionally (in most cases) comes along with iodine, perhaps this can be a reason for any potential benefit.
Re diabetes and carb type (yam, cassava, maize, wheat), I found this study interesting: http://www.idb.hr/diabetologia/07no2-3.pdf
Re uric acid, I was thinking about the fact that a sucrose rich diet tend to lower urinary pH (and soft drinks with phosphoric acid even more) and perhaps also pH in mouth, whereas potassium rich foods, for example potatoes alleviates this. But at the same time I am aware that citric acid can help dissolve kidney stones that however was brought about by a net acidic diet in the first place, and also perhaps can help with gout, vit c also seems to prevent uric acid. Uric acid is also a powerful antioxidant, so perhaps it’s related to more of that as a replacement when vit C intake is low or something. But yeah, I don’t know so much about all of this and the whole acid base balance theory is complicated.
A number of prospective cohort studies have however found an association between an acidic diet measured by the potential renal acid load (PRAL, which focuses on the balance of macrominerals including sulphur from protein) as well as dietary phosphorus intake, with diabetes, insulin resistance, cvd, mortality etc:
https://www.ncbi.nlm.nih.gov/pubmed/24225358
https://www.ncbi.nlm.nih.gov/pubmed/28818343
https://www.ncbi.nlm.nih.gov/pubmed/24232975
https://www.ncbi.nlm.nih.gov/pubmed/27052540
https://www.ncbi.nlm.nih.gov/pubmed/28539378
https://www.ncbi.nlm.nih.gov/pubmed/27390726
https://www.ncbi.nlm.nih.gov/pubmed/25863769
For example, among French women (study #2 above) it was found that those with the highest phosphorus intake (1700-1868 mg) had significantly higher risk of diabetes than those with the lowest (1086-1203 mg). I think it is interesting to consider this in light of the increasing use of phosphate additives over the past 50-60 years which correlates with the diabetes (and obesity) epidemic.
I will have to modify what I wrote a little.
First, at least according to http://www.dogcathomeprepareddiet.com/minerals.html, the requirement for cats per 2500 kcal is 2000 mg potassium, 250 mg sodium and 950 mg chloride. For dogs it is 3000 mg potassium, 375 mg sodium and 575 mg chloride. Hence as can be seen, cats require much more chloride than dogs. Perhaps this is a reflection of the fact that the cat is more of a true carnivore and produces much stronger stomach acid. However, according to https://en.m.wikipedia.org/wiki/Gastric_acid, stomach acid «is composed of hydrochloric acid (HCl), potassium chloride (KCl) and sodium chloride (NaCl)». Additionally, sodium bicarbonate is (later) excreted to neutralize the stomach acid.
So both sodium and chloride is needed in this process. But probably most of the sodium and chloride is recycled for later use, hence daily dietary requirement can be very low. But I’m not sure how applicable this is to humans, especially on grain, sucrose, legume and dairy rich diets. Could those foods increase risk of H Pylori infection, or cause a lack of zinc, which then messes up the whole thing?
As mentioned before, added salt is also typically a source of iodine. A large part of the general population could actually become iodine deficient if they avoid iodized salt especially if they do not eat seafoods or seaweeds or drink milk. Cheese appears to have an average of 75% less iodine than milk relative to calcium content. Diets like a high meat-low seafood paleo diet, or the vegan diet could therefore result in insufficient iodine intake: https://www.ncbi.nlm.nih.gov/m/pubmed/28901333. Iodine is obviously an extremely important mineral, which is why it was added to salt in the first place. One can only read Price’s «Nutrition and physical degeneration» to realize how much importance the population groups put on obtaining enough iodine. During pregnancy and especially during breastfeeding, the demand is greatly increased. If it turns out that lowering your added salt from for example 5 grams to 1 gram results in a suboptimal iodine intake, I can see how this is undesirable for some people even though it could also lower blood pressure and perhaps reduce the risk of diabetes (https://www.prevention.com/health/salt-intake-and-diabetes-risk). In this case it would be better to increase potassium intake than to reduce salt. Also the consumption of goitrogenic foods like broccoli, cabbage, brussel sprouts etc could increase iodine demand (even though hunter gatherers rarely est such foods, many paleo dieters and their gurus do), as could an excess exposure to fluoride, such as in the drinking water. Iodine may also have some therapeutical functions on its own, like being an antibiotic: https://www.researchgate.net/publication/228090108_Iodine_A_support_against_infections_and_emerging_as_an_antibiotic/amp
Finally: while the potassium/sodium ratio in human milk is typically 3:1, its relatively higher chloride/sodium ratio than found in sodium chloride, would in practice mean that if all the sodium and chloride came from regular salt, the dietary ratio should be 2:1 to make the potassium/chloride ratio similar as in human milk. And this fits the official recommendations.
People with kidney problems may however be adviced to reduce potassium to 3000 mg, sodium to 1500 mg, phosphorus to 1000 mg and protein to 60 grams. This is approx twice the amounts found in 2000 kcal human milk, and I think is probably ideal. The relatively high 4700 mg potassium recommendation for the general population in the US may be related to an excess of other minerals like sodium, including from food additives. If those are low it may well be better to cut the potassium intake. But at least we probably have evolved to handle a relatively high potassium intake as a result of a likely high intake of fruits and tubers during the stone age. For example 2000 kcal of fruits, potatoes or avocados supplies rougly 6000-7000 mg potassium and 700-900 mg phosphorus, but just around 100 mg sodium. For comparison brown rice would supply 800 mg potassium + 1500 mg phosphorus, and 2000 kcal oats 2200 mg potassium and 2700 mg phosphorus. I think the type of «whole food» diets used in the NuSI-2 study and now almost universally praised by experts and dietitians alike could actually be quite undesirable in the long run especially for diabetics due the potential excessive amounts of phosphorus from whole grains, legumes, lean meat and so on. Native hawaiians adoping a low fat diet where most of the calories came from tubers saw a dramatic decline in glucose levels in just three weeks. Later experiments using a similar diet but now with more whole grains, didn’t produce such results.
But an excess of salt can also be excreted through sweat, and perhaps this is a major reason for the benefits of physical exercise, taking saunas or just dwell in a hot country in primitive conditions, no air conditioning, but manual labour on a farm etc. This would also involve drinking plenty of water.
For example, hot tub therapy may offer great benefits for diabetics (http://www.nejm.org/doi/10.1056/NEJM199909163411216), and according to:
http://medicalsaunas.com/healthjournal/benefits-sauna.php «Intermittent hyperthermia has been demonstrated to reduce insulin resistance in an obese diabetic mouse model. Insulin resistant diabetic mice were subjected to 30 minutes of hyperthermic treatment, three times a week for twelve weeks. This resulted in a 31% decrease in insulin levels and a significant reduction in blood glucose levels, suggesting re-sensitization to insulin.»
I think – to bring up a point from the previous discussion we had – that a diet supplying rougly something like the following:
2500 kcal
300-500 ml milk/yogurt
1 1/2 – 2 eggs
100-125 gm meat/fish
1/2 ounce walnuts
1/2 ounce lentils/legumes (dry weight)
3 tbsp honey or sucrose
250 gm each of fruits/berries, potatoes/tubers and vegetables
1 glass red or white wine
1/4 tsp salt
+
white rice, olive oil (with vinegar) for calories
would offer an ideal amount and balance of the macro-minerals and only slightly above the kidney disease recommendation. As I suggested back then, milk and egg yolks could supply various beneficial nutrients, including phosphatidyl choline, which could help to keep the kidneys (and other organs) in good shape in old age, and hence they would be better able to handle an excess of minerals. And so there would be less risk of the higly elevated serum phosphorus which can result in soft tissue calcification (taken from the bones of the person) and perhaps alzheimer’s disease, or an excess of sodium which can cause hypertension and so on.
So much to reply and i am too busy at work to really spend time. A point on chlorine: humans have not completely evolved to carnivorous animals. They spent those 300,000 years of megafauna-based diet basically eating fat as much as possible. That was not enough to change dentition, but a cursory look at a pH table
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4519257/
shows that we are true carnivores in that respect, and all agree that low stomach pH is very protective (against allergies, infections and poor nutrient absorption), it is the original mineral pill. But I wonder how we managed to do that given our inferior genes, and it is possible that salt did play a role. Carnivores have a well developed recycling system for chlorine, and we may not.
i get that iodine is a confounding variable. Here are data which agree with your study but where iodine is not a confounding variable (Japan mortality versus Na/K ratio), due to high iodine intake in Japan, but it is only 22% quintile to quintile, with a minimum in the second quintile, corresponding to a Na/K ratio of 1.64! which is high
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4947715/
Thanks for the comment, glib.
I am wondering if the low intake of phosphorus and phosphate additives in Japan compared to the US could mean that their kidneys would be relieved of a great burden and hence better able to handle more salt.
My understanding is that stomach acid release is triggered by protein. On high carb grain type diets, less is released. And so feeding cats grains cause lower amount of stomach acid and makes them more prone to infections, whic can be a serious problem.
It is however possible that salt by itself and especially iodized salt can have an antimicrobial effect. Traditional healthy grain eating populations often consume the grains fermented including as beer, or along with other fermented foods which offer protection against some harmful microbes. In Japan the intake of soy sauce and miso can be quite high and this is also a major source of salt in their diet (so high salt intake may be associated with high intake of those.) Additionally such grain diets can be rich in other anti microbials: spices/herbs, garlic, vinegar, wine etc. Conversely fruits and tubers have so many anti-microbials on their own and are packed such that it would give the H Pylori or oral bacteria a hard time. I tend to think that it is better to add anti-microbials and eat tubers/fruits than to use too much fermented foods, as an excess of lactic acid and other bacteria «beneficial bacteria» may actually cause problems. We’re not designed to have a lot of those in the intestines or mouth, and certainly not in the small intestine.
Interestingly the approach used here: http://www.dogcathomeprepareddiet.com/diet_and_gastrointestinal_diseas.html for dealing with digestive disorders in cats and dogs involve a low fat gluten free diet relatively high in carbohydrates from white rice and potatoes. A reason why fat is avoided, especially vegetable oils, is to reduce inflammation. I recall thhq posted a study suggesting that participants of Kempner’s rice diet practically cured their digestive problems in just one week.
Another issue is whether higher salt intake leads to an increased caloric intake and hence is a contributor to overweight and thereby numerous health problems.
It is telling that Walter Kempner literally had to whip some of this patients to make them eat the extremely low salt rice diet. Conversely there’s no such problem with the high protein high carb very low fat Sumo wrestler’s diet. I think both salt/chloride and phosphorus, as well as protein, can make food taste better, sometimes a reason for their use as additives in the first place. And if they taste better, they are likely to be consumed in higher quantities. Although perphaps protein leads to better satiety so that caloric intake is not increased overall. However 50 years ago the Japanese may have consumed only half as much animal protein as today, and they were slimmer, but fat intake was also lower.
And then on the other side of the balance is potassium and maybe calcium, the «alkaline minerals» which may have a satiating effect. At least studies indicates that adding high potassium low protein low phosphorus low salt foods like potatoes (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4775821/), avocado (https://www.ncbi.nlm.nih.gov/m/pubmed/24279738/)
raisins (http://calraisins.org/wp-content/uploads/2012/06/Satiety-in-Children.pdf) and others like prunes and baobab fruit, may increase satiety and lead to a reduce caloric intake at the next meal. It appears that fruit is more satiating than a low sugar whole grain based snack even though for example raisins supply less than half as much fiber per calorie as oats. There are obviously many other differences than just the mineral balance.
About salt:
1) it does increase gastric juice production. That in itself has an anti-bacterial effect.
2) It helps with insulin sensitivity. A very good thing in Japan where so many calories are from rice. It is not a surprise that the minimum mortality there (although the data are not conclusive) is at a higher Na/K ratio than in the West, and that their mortality curve is flatter.
There are a number of foods and spices that increase gastric production. Things like chicory, cardoon, or garlic. But historically, the best and most heavily used stimulant is gelatin, or broth containing it. It is noted that glycin is also heavily used for bile production (although bile release is triggered by fats), and so glycin helps produce both stomach acid (pH=1.5) and bile (pH=10.5) in the process creating an approximately neutral environment in the small intestine. It could be the reason why eating fat with proteins is associated with better digestion and possibly less inflammation. We are for sure mostly glycin-deficient.
glib,
The idea that low sodium induces insulin resistance seems rather inconclusive: https://www.ncbi.nlm.nih.gov/pmc/articles/ But even if low intakes like less than 1000-1500 mg sodium/day did, would that be an argument for consuming Japanese like amounts like 5000 mg/day?
At least salt does seem to increase risk of diabetes significantly and high potassium low salt foods like raisins and avocados improves metabolic syndrome, see for example: http://calraisins.org/wp-content/uploads/2015/07/Raisin_Diabetes_Bays-Anderson-2-15.pdf, https://www.ncbi.nlm.nih.gov/m/pubmed/28393409. I believe that a low fat low salt diet where almost all of the calories comes from potassium rich potatoes/tubers may well cure cases of diabetes in a month or so, but it’s too much to go into the evidence here and now.
Also four of the eight persons using hot tub therapy for 3 weeks I linked above, saw average 25% decline in fasting glucose (similar as in the low salt Hawaii study where most of the calories came from tubers) but not much change in body weight. Not sure why, but the salt in their bodies would very likely have decreased as a result of massive sweating such baths would induce. So that could have been one factor.
I think probably 1500-2400 sodium (approx 4-6 grams of salt) is ideal depending on the health of the person and caloric intake as well as his/her amount of sweating (and then an ideal amount of protein could be 60-90 gram, potassium 3000-4700 mg, and phosphorus 1000-1550 mg on a 2000-2500 kcal diet). So I’m pretty okay with the official recommendations. This is already 5 times more as the recommendation for dogs per 2500 kcal. Do we really need more?
Same I’m sure for stomach acid; if you don’t get enough salt you won’t get enough stomach acid. But does this mean large amounts of salt would be a great idea?
H. pylori infection is a major issue – said to affect half the world’s population (in the US it is especially prevalent among Hispanics and African Americans, affecting almost 60%), which then neutralizes stomach acid. And it’s not like those with a high salt intake suffer less. But the solution isn’t more salt and more stomach acid, although maybe traditionally made miso and soy sauce (rich in salt) could help. A grain free diet with fruits and tubers, even honey, plus anti-microbials like garlic and chicory root which you mentioned, seems a better idea. Roots often have strong anti-microbial compounds as a defense against yeasts/fungi and bacteria in the wet soil; ginger root is also helpful for digestive disorders.
I am also aware of the benefits of gelatine rich stocks to fix digestive problems – and try to make this regularly from chicken wings especially. Glycine/gelatine/collagen is very popular in the paleo world, and paleo gurus sell products made with collagen. And there’s probably some biased and exaggerated statements floating around regarding the benefit.
But 2000 kcal human milk supplies just 750 mg glycine, about 2.5% of total protein. Same percentage for cow milk. In eggs it’s around 3%. Hard cheese about 2%. Whole chicken and sardines about 5% and ground beef 7-8%. Taking up to 30 grams of collagen per day – as some gurus recommend – might supply something like 7-10 grams of glycine. I think that’s too much, especially if on top of lots of beef: https://honey-guide.com/2015/11/26/bone-broth-gelatine-oxalates-and-kidney-stones/
If if it works it may be more like a drug or the result of some other problem/deficiency like lack of choline and perhaps some other nutrients/compounds found in egg yolks, organ meats and milk, but missing from egg whites, muscle meat and cheese.
Glycine is also involved in methylation. Betaine = tri-methyl-glycine. Choline can be turned into betaine, but has other functions such as the mentioned very important phosphatidyl-choline which is generally depleted from cells/organs as we get older. Maybe glycine could help spare choline.
glib,
Sorry about these endlessly long posts. I think this will be the last. But did some more digging today and what you said about glycine and bile is now starting to make a lot of sense.
It appears that phosphatidylcholine/lecithin is a constituent of bile (wikipedia: «The composition of gallbladder bile is 97% water, 0.7%[1] bile salts, 0.2% bilirubin, 0.51% fats (cholesterol, fatty acids and lecithin),[1] and 200 meq/l inorganic salts.») and thereby a limiting factor for its production. However maybe betaine or glycine could help spare choline. Beet juice, rich in betaine, has been used to stimulate bile flow.
It also seems that bile could be effective in eradicating/preventing H pylori, but given that we nowadays often consume egg whites and cheese rather than egg yolks and milk and additionally have (mostly) ditched organ meats, choline intake is rather low. In the US about 30-40% less than recommended. Hence perhaps we do not produce optimal amounts of bile as a result.
However in the west we also eat wheat which can be a reasonable source of betaine, around 400 mg/2000 kcal (about 10% more in whole vs refined). But I think a minimum of 500-750 mg choline and 250-500 mg betaine (1000 mg combined) is needed long term for good health, although much more short term for therapeutical reasons. I suspect those with H pylori infection obtain less choline and betaine than those without as the infection can be associated with junk food diets rich in empty calories like sugar and vegetable oils.
In Asia, consumption of rice – absent in betaine – is very high. In India (where per capita salt intake is around 10 grams) the rate of H pylori infection seems to be around 80% and the rate of diabetes is now as high as in the US, but much higher relative to BMI (this is thought to be due to genetics, but maybe diet plays a large role).
This diabetes may well have something to do with fatty liver which could possibly have been alleviated with more protein/glycine/betaine/folate/choline. And getting more of those would maybe also have reduced H pylori infection rates.
I think that proteins in milk and eggs are essentially structural, to build fast a young body, but in adulthood you just replace losses. And 80% losses are from skin and gut, so glycin rich tissues.
Plus glycin is used for other tasks, principally bile. We make a quart of the stuff every day. I am sure it is recycled but recycling costs you. Compare that quart with methionine, which is also used for other tasks, mostly methylation, which is a few grams a day type of process.
Further, glycin can be had from the serine-glycin cycle, but that is mediated by folates and I doubt that we intake enough folates to make a dent (a gorilla or cow might). Glycin is also a main activator of mTOR, even compared to other amino-acids, which is bad, but I think it is best to eat all the protein you need, then fast a bit if you think you need it. So all in all I am convinced that most people do not eat enough gelatin.
glib
I just think the amount of glycine consumed during the stone age, in traditional diets and what is found in human milk, could give an indication of what the body is comfortable with. If your’e going much above that, it’s more like a drug which could have side effects (such as the mentioned issue with kidney stones).
Over the past decades so many nutrients have been touted for their magical properties if taken in gigadosages. (And I guess it was allways «backed» by some biochemical argument of the type you mentioned.) But few of them stood the test of time. And some turned out to cause harm.
It is also worth considering that many meat products like ground beef, sausages and various canned and processed meats/products often have a high glycine content as the cheaper collagen rich parts (like connective tissue) may have been added, or gelatine. This gelatine may actually come from the boiling of pork skin, bovine hides and bones for long periods of time.
These foods have less % of protein as methionine and a higher glycine/methionine ratio than regular meat or fish. It may be undesirable for people eating much such foods to take extra gelatine, especially if they also eat the chicken with skin or the cheaper parts like wings which is very rich in collagen. But people eating mostly chicken breasts, fish filets and steaks may well benefit from 5 grams or something of extra gelatine.
Organ meats (ex skin) and shellfish have a high glycine/methionine ratio, i.e 2-3 compared to 1.6-1.8 in fish, chicken breast and steak. However it’s less than the 3 or so ratio in ground beef or pork sausages. An analysis of 15 species of whole fish suggested a 2.2 ratio, higher than the 1.8 ratio for sardines with bones listed on nutritiondata. The ratio in human milk is 1.25, eggs and caviar 1.15, cow milk 1.0 and swiss cheese 0.65.
So often my analysis ends with a big question mark around cheese. In this case it’s the very low glycine/methionine ratio. But basically cheese has been stripped of several nutrients found in milk such as riboflavin, pantothenic acid and choline. It turns out organ meats and human milk are also particularly rich in these three relative to others, so maybe an answer can be found there: we’re missing certain nutrients which causes imbalances which then increases need for for example glycine. I think the fortification of grains with some nutrients and not others creates imbalances which messes up everything (and hence may even contribute to obesity and other problems). As I’ve understood it in the US of the b-vitamins, cereals are typically fortified with thiamine, riboflavin, niacin and folic acid, but not B6, pantothenic acid, choline and biotin. It would be better to eat unfortified refined grains and take a more balanced vitamin supplement.
Forgot to mention vitamin c. Human milk is generally low in nutrients however is particularly rich in vitamin c, typically 140 mg/2000 kcal. If the diet is rich in vitamin c, at least from natural sources, it may go as high as to 300 mg/2000 kcal which seems to be an upper ceiling which is not exceeded no matter how much vitamin c is consumed. This may be comparable to what’s found in gorilla milk.
Point anyway is that vit c is needed to synthesize collagen. Interestingly vit c seems to improve conditions that are also helped with glycine/collagen such as poor skin or wrinkles. I’ve noticed many of the diet gurus who frequently write about sleep have mentioned how glycine/gelatine can be helpful, but I haven’t seen it mentioned that a lack of vitamin c has been linked to poor sleep. A study showed dramatic improvements in sleep/quality with ingestion of two kiwifruits (approx 150 mg vitamin c) an hour before bedtime over 4 weeks, for example. https://www.ncbi.nlm.nih.gov/m/pubmed/21669584/
Hi Samuel,
I’m a bit confused by your analysis here. If the study instructed people to limit refined flour, how can you say the results show the diet is healthy regardless of refined pasta and breads? Wouldn’t limiting refined flour (as well as processed food in general) also limit refined pasta and breads?
Its too bad they only tested the extremes in this study — Low carb vs. Low fat. I would love to know whether whole food moderate fat/moderate carb would have worked as well. In other words, is it the “wholeness” or something about limiting a macro-nutrient that is key. Any thoughts?
Hi Jodie,
My opinion is that it’s probably both.
I’ve noticed that you trash Taubes at least once a year about his belief in the insulin theory, but I always feel like defending him because his books over the last ten years have changed my health, Especially his view that when third world countries started importing sugar, flour, and vegetable oil people started starting getting fat and getting western diseases.
Hi Peter,
What you wrote is the primary form of pushback I get against my critiques of his writing. If there’s one thing I can credit him for, it’s giving people permission to try low-carb by popularizing the idea that it isn’t going to kill them and it can have benefits. What I object to is the science abuse– seeing it published in high-profile places raises my hackles periodically.
please Samuel just have look to this:
http://raypeat.com/articles/articles/sugar-issues.shtml
and all of Stephan Guyenet have demonstrated.
I just would like to say that eating 100G of WHITE SUGAR A DAY I’m leaner than was on low carb.
I’m not saying that low carb is not a great tool if you want to lose weight. Just that in this case, it’s not because your cut back on sugar.
good luck with your 100g of white sugar a day. you may be lean because your intestinal fructose absorption is very low. or because you already lost most of your teeth and can’t eat anything else 😉
The major problem with Taubes is that he is a reductionist. He’s not alone. So are many Paleos, Vegans, Fruitarians and megavitamin proponents. Metaphorically they see their single-issue as a towering sequoia when in fact is’s a sickly dogwood.
The better approach is empiricism. Using the same metaphor, empirics looks at the forest of individual health factors. Here are two examples based on very large populations:
https://www.thecalculator.co/health/Framingham-Risk-Score-Calculator-for-Coronary-Heart-Disease-745.html
https://www.livingto100.com/?mobile=0
The use of the second model allows one to look at the validity of most of the dietary schemes, by comparing “what if” cases. In my case, my lifespan is projected to be 90 years. If I quit eating my daily pastry and chips I could raise this to 92. And I REALLY don’t care.
11 years ago when I was obese and diabetic my lifespan projected at 72, the same as Robert Atkins. The extension of my lifespan ftom 72 to 90 is a result of weight loss and exercise, not an obsession with macronutrient ratios.
Stephan you didn’t answer to my question (even if I can understand it;)
so I try again.
I frankly dont know why, but I’m leaner eating white chocolate than eating a meditreanean diet. I know this seems crazy or false, but In my case it is true. I’m leaner now than when I was low carb and low carb Mediterranean. I suspect that the perfect combo for me to lose weight is stearic acid plus sugar (I know I know, it seems strange)
I check my blood test recently. perfect.
so should I switch to a Mediterranean diet?
if I ask you this question this is not just form me. it is a general question.
if a diet with many vitamins and nutrients makes you leaner and allow you to have better blood pressure etc, do you still need to follow a Mediterranean diet?
thanks!
Hi all,
Forgive me if this is a ridiculous question, but we all seem to be taking the superiority of whole vs. processed/refined foods for granted. It seems reasonable, but so much of what seemed reasonable just a few years ago, is now up for question. Are there any studies comparing isocaloric diets comprised of whole food vs. refined/processed foods? So far, all I can find is this:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4095661/
This particular study doesn’t seem to show much difference between the two diets. Is it possible that it really does all come down to calories, that the whole food diet “works” because its low in food reward and that you could get a similar effect from eating less than appetizing processed foods? That would explain the success of the liquid tube feeding diet that Stephan has often discussed.
Hi Jodie,
Kudos for thinking outside the box. I feel pretty strongly that whole foods *tend to be* healthier than refined foods, such that a totally unrefined diet will end up being much healthier than a highly refined one over the life course. But I think a major reason for that is that we tend to overconsume calories when we eat a refined diet. When calories aren’t overconsumed the consequences aren’t necessarily bad. For example, the Japanese diet was something like 80% white rice when their heart attack rates were extremely low in the mid-20th century (they’re still low but the diet has changed). They were very lean. I’m not claiming they were the paragon of health in all ways (e.g., Japan has high rates of hemorrhagic stroke, although I don’t know what it was back then).
Tagging on to jodi’s remarks, on your blog, you mention that in mice, isocaloric diets of unrefined chow vs refined and rewarding D12492 produce differing adiposity. Presumably, the refined/rewarding diet raises the body fat set point and ends up causing a reduction of energy expenditure to defend it. Is this a robust finding in rodents? Any evidence of this in humans, or does reward in humans really only only affect the calorie input side of things? Perhaps a calorie matched study of a varied junk food diet vs nutrament. All refined, but differing in reward.
Hi Joe,
Yes, in my former lab we showed a number of times that the obesity produced by refined high-fat diets in rodents isn’t just about increased calorie intake. What we see is that they overeat for 7-10 days, then their calorie intake goes back to normal, or almost. But they continue gaining fat, implying that their energy expenditure is lower. This has been replicated by other groups. I vaguely recall that eventually calorie intake creeps back up a bit though, once they’ve developed pronounced obesity.
I’m not aware of any evidence that this energy expenditure effect happens in humans, and I wouldn’t confidently predict that it does. In humans, body fatness is regulated primarily by the energy in side of the equation, while in rodents it’s more balanced between in and out. The energy expenditure of rodents can vary considerably even without physical activity (they have a huge capacity for nonshivering thermogenesis) but human energy expenditure is much more stable. Furthermore, people with obesity have exactly the energy expenditure you would expect for their tissue mass (i.e., their energy expenditure isn’t low).
That said, I would love to see the study done that you described. I actually advocated for something like this a few years ago (not publicly). Calorie-matched diets but very different in reward value. I predict that differences would arise. fMRI might detect differences in brain activity suggesting differences in the activity of energy homeostasis circuits and responsiveness to food cues. Self-reported hunger would probably differ. And perhaps there would be a small energy expenditure difference– I don’t know.
I believe that there is a marginal advantage to isocaloric whole foods which is achieved at end-of-life. Two years longevity between ALL diets IMO.
I focus on longevity because of people like Ancel Keys and Jack LaLanne. Keys was still publishing scientific papers at 97, LaLanne stll exercising until the day he died at 95. The obesity crisis is an immense social and economic burden. I want to be like those two. I choose not to be a burden on my family.
The blog post says early on “It was funded by the Nutrition Science Initiative, a nonprofit involving Gary Taubes that has been funding additional research to test the carbohydrate-insulin hypothesis of obesity.”
This is only partly true. The study methods had already been published when NuSI came late to the game, kicking in an additional $3 million to make it “more rigorous and more impactful. The specific factors agreed on were to make the study larger (more participants) and more extensive (more secondary and exploratory outcomes measures) without extending the timeline (faster pace of enrollment).” Original study had $2.5 million funding.
See p. 2 of https://jamanetwork.com/data/Journals/JAMA/936761/JOI180008supp2_prod.pdf
So I don’t think it’s fair to imply that NuSI designed the study.
Hi cavenewt,
I didn’t state or imply that NuSI designed the study. What I said is that NuSI funded the study. Thank you for pointing out that it had non-NuSI funding as well.
NuSI believed, prior to the results arriving, that this would be an informative test of their hypothesis. Hence the funding. That’s the only thing I’m implying.
And yet, we have self help low carber forums frequented by millions. I am sure people’s health and metabolism is improved if they switch from grains to avocado, beans and sweet potato. But with an estimated half of the population being diabetic or pre-diabetic, for many this is not enough. I myself never normalized my blood glucose on a carb-rich diet which included lots of apples, rice, banana and potatoes, as well as beans and roots.
Keep in mind that we have a vague notion of “mediterranean diet” floating around our society, that many interpret to mean pizza and pasta and veggies. And also a food piramid based on grains. Big pharma is still getting its profits through non-decreasing consumption of refined carbs and seed oils . One can expect this thing to appear in the press as proof that grains and fruits are good for you.
It’ll never appear in the mainstream press at all glib.
We have moved on, from the Food Pyramid to My Plate. That’s the mainstream. You see the psters for it on the wall in grade school cafeterias.
https://www.choosemyplate.gov/myplate-graphic-resources
It’s not too hard to see that grains take up 1/4 of the plate.
Just curious…..according to Examine .com analysis TG decreased in both groups but significant different lowering in LCHF compare to LFHC. At same time HDLc increased signficantly in LCHF group. TG/HDL ratio…often considered a proxy for insulin resistance ……is this significant?
I have heard that HFLC shifts HDL up, but have had better success raising HDL with exercise hap.
I’ve been puzzling over this report on glycogen depletion and appetite.
https://www.physiology.org/doi/pdf/10.1152/jappl.1999.87.3.947
The continuous blood glucose profiles are really interesting. Refeed on HFLC produces somewhat longer and lower blood glucose peaks than refeed on LFHC. What’s more interesting to me is the triggering of appetite by falling blood glucose. Is the ghrelin release a result of blood glucose levels? Or is it more complicated?
I’m wondering now how much appetite is trained. An athlete with fasting BG of 70 isn’t hungry, but a 200 BG diabetic at 70 BG might be ravenous.
Stephan’s earlier post discusses the relative unimportance of blood glucose for hunger.
http://wholehealthsource.blogspot.com/2016/09/do-blood-glucose-levels-affect-hunger.html?m=1
But the effect of intestinal glucose is noted as important. This study uses duodenal glucose injection to demonstrate it
https://www.ncbi.nlm.nih.gov/m/pubmed/8770012/