Imagine a political historian discovers that Lyndon Johnson accepted a campaign contribution from a big Wall Street bank. Since Johnson’s policies helped shape the modern Democratic Party, everyone agrees the Democrats are built on a foundation of lies. “Republicans Vindicated; Small-Government Conservativism Was Right All Along”, say the headlines of all the major newspapers.
This is kind of how I feel about the reaction to the latest New York Times article.
How The Sugar Industry Shifted Blame To Fat describes new historical research that finds that the sugar industry sponsored a study showing that fat (and not sugar) was the major risk factor for cardiovascular disease. They tie this into a bigger narrative about how sugar is the real dietary villain, and it’s only the sugar industry’s successful bribery work that made us suspect fat for so long:
The revelations are important because the debate about the relative harms of sugar and saturated fat continues today, Dr. Glantz said. For many decades, health officials encouraged Americans to reduce their fat intake, which led many people to consume low-fat, high-sugar foods that some experts now blame for fueling the obesity crisis.
“It was a very smart thing the sugar industry did, because review papers, especially if you get them published in a very prominent journal, tend to shape the overall scientific discussion,” he said […]
I’m glad researchers have discovered this. But treating it as a smoking gun which exonerates fat and blames sugar is like the political example above. Yes, it’s sketchy for LBJ to take Wall Street money. But this kind of low-level corruption is so universal that concentrating on any one example is likely to lead to overcorrection.
Yes, the sugar lobby sponsors some research, but the fat lobby has researchers of its own. They tend to be associated with the dairy and meat industries, both of which are high in saturated fat and both of which are very involved in nutrition research. For example, Siri-Tarino et al’s Meta-analysis of prospective cohort studies evaluating the association of saturated fat with cardiovascular disease finds that saturated fat does not increase heart disease risk, but it has a little footnote saying that it’s supported by the National Dairy Council. Modulation of Replacement Nutrients, which finds that replacing dietary fat with dietary sugar doesn’t help and may worsen heart disease, includes two authors affiliated with the National Dairy Council and one affiliated with the National Cattleman’s Beef Association.
Mother Jones does a dairy industry expose and finds:
[Industry] ties can sometimes be hard to avoid, since much of the research on dairy is funded by a constellation of industry-backed institutes, including the Nestlé Nutrition Institute, the Dannon Institute, and the Dairy Research Institute, which spends $19 million a year “to establish the health benefits of dairy products and ingredients.” Even Willett acknowledges that he has received a “very small” dairy industry grant. Dairy companies also donate heavily to the American Society for Nutrition, which publishes the influential American Journal of Clinical Nutrition, and the Academy of Nutrition and Dietetics, “the world’s largest organization of food and nutrition professionals.”
Then there are the industry’s donations to politicians. Dairy companies spent nearly $63 million on federal lobbying and gave $24 million to candidates between 2004 and 2014.
As Jim Babcock points out in the comments, some of the agendas are more complicated than I’m making them sound. Dairy was pretty okay with the low-fat craze for a while, because it let them market low-fat milk. But they do seem to be behind a lot of the pro-saturated-fat research going on right now, and their website does promote pro-saturated-fat articles (1, 2, 3). Overall they seem to be taking a low-key approach where they roll with some studies and push back on others.
In any case, claims that the sugar industry sponsored one study back in the 1960s, and this means everything we’ve ever thought is wrong and biased against fat and in favor of sugar, miss the point (especially since there are probably problems with both sugar and fat). Whatever study the New York Times has dredged up was one volley in an eternal clandestine war of Big Fat against Big Sugar, and figuring out who’s distorted the science more is the sort of project that’s going to take more than one article.
Scott, do you have any personal opinions about the health effects of sugar and fat?
I’m not a nutritionist and this isn’t a very well-informed opinion, but my guess is that this is a dumb debate. If you eat lots of calories with few nutrients you are going to have increased risk of obesity and heart disease, and any differences between fatty vs. sugary foods are probably not that important in the grand scheme of things. Sugary drinks may be unusually bad because our body doesn’t read them as food and adjust caloric intake appropriately, but this still isn’t the most important factor.
I recommend these two Stephen Guyenet posts, especially the graph in the first one:
A lot of the concern about high levels of dietary sugar is only indirectly related to obesity. For example, fructose seems to have unique bad effects on uric acid that contribute to renal issues like diabetes.
Fruit has been giving all the people diabetes!
Fructose is more concentrated in fruit juice and high-fructose corn syrup than in regular ol’ fruit.
I’m hypoglycemic, and I can eat fruit without difficulty, but a couple of glasses of fruit juice make me feel bad. And sad.
Fruit has been giving all the people diabetes!
You jest, but if you start reading the small print of the ingredients in all the “low-fat”, “healthy” foods, you find they’re stuffed full of carbs/starch/sugars instead.
And I went along thinking I was being really virtuous switching to drinking pure fruit juices instead of sodas etc. until the nurse practically had a fit and told me they were terrible because fruit juice is full of sugars – yes, even the “not from concentrate, no added sugar, all natural” versions. If you’re going to drink fruit juice, dilute it heavily with water, was what I was told.
So parents probably think they’re doing the right thing making sure Junior drinks only real fruit juice instead of fizzy drinks and the like, with no idea it’s loading the kid up on as much sugar as the manufactured drinks.
I’ll be very interested to see how the proposed “sugar taxes” work out in practice; will they also hit fruit juices or will they be allowed to opt out? This is why companies like Coca-Cola are introducing products using stevia instead of sugar and artificial sweeteners 🙂
Although it makes sense that they’re both as bad when they have as much sugar, fruit juice and sweetened beverages don’t actually appear have quite the same health effects.
What’s the difference between orange juice, fresh, pulp in, and oranges?
I guess most other fruits remove nonfiber solids from the juice. Are those significant dietetically?
Decius: I once tried making my own OJ from oranges bought from the shop. I was surprised by how many oranges went into a small amount of juice.
I think it might be a quantity thing – eating too much fruit is hard, because of all the fibre, whereas drinking too much juice is a lot easier – especially when you’re drinking it to quench your thirst. Hence the advice given to Deiseach to dilute the stuff, I guess – if you do that, you don’t need as much to quench thirst, and it’s harder to drink to excess.
I drink a lot of Diet Coke and Coke Zero. If I want to drink and am out of those or think I’ve been drinking too much of them, one of my standard substitute is water with a little orange juice or grapefruit juice–probably well under a tenth the volume.
That gives the water flavor without giving me very many calories.
Hah! I do the same thing, except always with carbonated water. At soda fountains, I also drop about 5% pineapple or orange soda into my Coke Zero, and it seems just about as sweet as my taste buds can handle.
Fructose is so called because fruit is full of it.
E.g. apples have about 2g of sucrose, 2.4g of glucose, and 5.9g of fructose (per 100g, source). Watermelon, technically a berry, has similar ratios: 1.9g sucrose, 2.4g glucose, 5.2g fructose (source). Pears are very fructose-rich: 1g sucrose, 4g glucose, 9.2g fructose (source).
Okay I must be reading this wrong. 470g of carbs per day = 1880 kcal. From carbs alone, before counting proteins or fat, for the average American.
I don’t see a problem with those numbers. What seems off?
What seems off is why so many people are wondering about why the average BMI is increasing if those numbers are accurate to be honest. I’ve seen wildly differing numbers on average daily dietary food intake for Americans (from ~2250 to 3800 kcal) and after reading the caveats on those numbers stopped paying attention about population stuff (no good baseline data => garbage conclusions).
Actually, technically, you are. “Nutritionist” isn’t a protected term like “dietitian”. You can call yourself a nutritionist if you want to.
That still means that technically he is not, as, technically, he chooses not to.
If I understand correctly, one argument of the pro-fat tribe is that eating excess calories as fat is more difficult than eating excess calories as sugar since fat is more filling than sugar (or at least, it is more likely to be associated with proteins and fiber that are more filling, while sugar often is found in things like sodas).
Is that correct?
Fat also has a different metabolic response then sugar. But I’m not dietician nor do I have a good understanding of insulin resistance.
I’m not Scott, but: I lost a bunch of weight when I eschewed sugar and starch. I don’t want to crap up the comments with buzzwords that are SEO-likely (much like the prohibition on Death Eater references, ’tis an attempt to keep the blog below the general radar as much as possible), but I am a physician, and I rapidly lost 80 pounds when I stopped eating carbohydrates in anything other than trace amounts. Is it an ideal diet? I don’t know. It worked for me, and I have no plans to change. YMMV.
Sure, eliminate a major source of calories in your diet and you’ll lose weight. If you’d also eliminated fats and proteins you’d lose weight even faster but probably would need to heal thyself.
When I wanted to lose weight I cut my portions in half without restricting any particular macronutrient. This worked well also.
Of course losing weight bottoms down to calories/calories out and you can do that by cutting calories anywhere. The stronger claim that I think devilbunny is offering is that cutting out sugar modulated his appetite so he could actually lose the weight. That’s the principle behind a lot of low carb diets, anyway.
I’ve been hearing this response for so long that it’s practically eye roll worthy at this point. Treating things like “how many calories you eat” as completely exogenous variables for the average person is honestly just stupid, as is the equivalent claim that it makes any sense to ignore different foods’ effect on satiety (and thus willpower, etc)
How many calories you eat IS completely exogenous. You don’t have to ignore different foods’ effects on satiety to believe that. Just that ultimately, it is your choice whether or not to put the food in your mouth. People can and have literally starved themselves to death by choice.
It may be easier to cut out carbs entirely than to cut the macronutrients roughly equally or to cut out fats, but all are doable.
Not everyone has the will power to starve themselves to death, however. Also, A life while famished may be less fun than the health risks of being overweight.
Did you eat fewer calories total after cutting out sugar? Or did you replace those sugar calories with other foods? Any idea whether removing/replacing sugar affected your hunger levels?
If you don’t eat any carbs you’ll be in ketosis.
There was a guy who did a Twinkie diet where he ate only twinkies, and lost a lot of weight
An what’s wrong with ketosis?
The hell? Is “Death Eater” a euphemism for something around here?
Congratulation son the weight loss – FWIW, my recollection from the last time I looked at this, is that there are basically two kinds of diets:
1) Diets that measure and limit consumption, either by telling you how many calories you get or by measuring macronutrients like grams of carbs, proteins and fats.
2) Diets that work by changing your eating habits – e.g. give up alcohol, sugar and flour; drink shakes for two meals a day; only eat things theoretically available to cavemen; only eat one meal a day, etc.
The second class basically works by limiting calories, (some may also be slightly better at the margins due to metabolic effects, but if so, it’s basically a rounding error at best compared to the caloric effects). If you eat cabbage soup for two meals a day, it’s hard to load a lot of calories, plus the boredom may discourage eating for pleasure.
The bottom line is that like exercise, the best diet is the one you will do (assuming it’s not affirmatively unhealthy).
That doesn’t mean that sugar is (or isn’t) worse than fat in your diet, but if cutting sugar improved your life, you don’t need to answer that question.
Where would you put fasting in this model? Specifically a one-day-no-calorie fast, which I started maintaining 11 months ago and which caused a loss of 2.5 stone in 14 weeks.
It’s a taxonomic problem, so put it where you want, but if you don’t count calories on non-fasting days, I’d put it in group two.
It’s literally intermittent fasting, although most people use the term to refer to eating one meal per day.
And congratulations! Do you do 1 day a week?
“It’s literally intermittent fasting, although most people use the term to refer to eating one meal per day.”
That’s literally the thought that passed through my head – I’m doing what more says recommend than otherwise, but with giant interval sizes.
And thanks! I originally picked Mondays, but work complications moved it to Wednesdays, where they sit to this day. I generally find it helps to consume as little as possible end-of-story, so I have maybe a few sips of water a day. My brain feels about 3 times quicker. Couldn’t recommend it enough.
edit: For some reason any kind of chevron-type quote marker isn’t working for the J mann quote above. Not sure what’s going on. Using “”s for now.
I started tracking calories thanks to a question I asked in an open thread a few months ago. Lost 10 pounds in 1.5 months. I’ve since increased my calorie intake, because while I still want to lose weight, that felt too fast, and I wasn’t incredibly obese to start with.
It wasn’t knocking out fats or carbs in particular, but just realizing how many calories there are in every day foods. Particularly carbs. I haven’t cut them out, but I need to be careful when eating them. A half-breast of chicken is more filling than two slices of while wheat bread but has fewer calories.
edit to add: just measuring food puts you more in control. I’ve gotten good enough that I can eyeball just what is “1 cup” if I’m not around my measuring spoons.
I know a nutritionist very well, and here is her take:
High levels of sugar and fat are related to obesity inasmuch as high caloric intake is related to obesity (read: strongly related).
High levels of sugar intake are related to insulin resistance and type 2 diabetes.
as for the high fructose corn syrup thing, HFCS is about 55% fructose and 45% glucose. Regular sucrose is 50% fructose and 50% glucose. There are some studies, and their conclusions are probably true, that fructose is slightly worse than glucose. However, the difference in percentages is not very much, and worrying about specific fructose content rather than the amount of sugar intake in general is like making beds in a burning house.
I found this story on the day it made waves via HackerNews, and comments left there as well as in particular this video they linked  were awfully convincing about sugar being the bigger culprit. Would love to hear SSC’s position on the debate, although I realize this post was expressly not to give a position but just to give some historical perspective to the issue. Anyone have other convincing arguments to marshall, particularly on the other side?
My background in nutrition is that I read Food and Western Disease. It is perhaps the nerdiest book I’ve ever read, and I read a lot of textbooks. Key quote:
The book also includes this image with the caption “Fat and carbohydrates – both to the left and to the right.”
My impression echoes Scott’s: fat vs sugar seems to be a red herring. Eating a lot of calories with few micronutrients won’t end well, no matter how they’re proportioned.
And, for what it’s worth, Food and Western Disease advocates for a paleo diet and doesn’t give two darns about macronutrient composition, beyond getting enough protein, even though most paleo blogs treat carbohydrates as the Worst Possible Thing. Also, it convinced me to eat soylent (proper amount of micronutrient to the calorie!) which, from a certain perspective, is literally the furthest thing from paleo. The author, Staffan Lindeberg, has some talks youtube, which are delightfully nerdy and I think would appeal to a lot of commenters here.
Some of the confusion may come from the categories: fat, carbohydrate.
There are lots of different fats and lots of different carbs.
Also, processing changes things. The honey from the market is seldom anything like the raw, unheated and unfiltered honey paleo people ate. Some carbohydrates are digested by enzymes in the mouth and stomach, and some are digested in the hind gut by bacteria etc., the gut flora. The ones digested by enzymes yield simple sugars but the ones digested by flora yield fats such as butyrate, acetate and propionate. Cooking changes those carbs. E.G. a potato can be either depending on its temperature. When raw, or cooked then cooled, its starch is “resistant”, meaning that it resists enzymes and is digested in the hind gut by bacteria.
You may think that you ate carbs, but your body sees fat.
As it happens those fats are very helpful since they maintain the health of gut linings so that they don’t leak and wreak havoc on the immune system.
Worse, people vary. The precise effects have genetic and environmental dependencies, and those are altered by life history. Dietary guidelines aren’t very useful, however we are slowly developing individual diagnoses and regimens.
“Fats” and “sugar” are high-abstraction labels that are not terribly useful once you dive into details. For an obvious example, glucose and fructose, both “sugars”, are processed by the body in vastly different ways. Fats (or fatty acids) come in different varieties (a crude distinction is saturated/monounsaturated/polyunsaturated) with quite different physiological effects. And real food tends to have a lot of bioactive components in it which modify the body’s response to what has been consumed. And some food is for yourself and some is for your symbiotic gut microflora and microfauna…
tl;dr: It’s complicated.
So, you’re saying that, within the set of fats and within the set of carbohydrates, there’s a spectrum, with members range from “good” to “bad”. The ones encountered by early man are “good” ones, because we’ve become adapted to them. And then comes along agriculture and novel cooking techniques and food processing, and now we’re eating the “bad” ones. As such, the divide between “fat” and “carbohydrate” isn’t the salient distinction. Also, people vary, both because of individual variation and because evolution can happen over a short timescale given sufficient pressure (lactose persistence comes to mind).
This seems about right.
I ask because I’m not sure if you’re disagreeing—I would very much like to rid myself of any misconceptions I may have—or adding detail/nuance.
Gary Jones has some good points, but to add a few details. The “carbs” you eat now are not analogous to the “carbs” that were earn X thousand years ago. The ratio of carbs to protein in wheat has shifted dramatically with selection and breeding, and the processing of flour has changed as well. If you were to eat grains 10,000 years ago you would probably either soak them for several days during which they would germinate and start to ferment, or would have cracked them with a couple of rocks (but not ground or milled them to a major degree).
If you want to demonstrate this for yourself, find some wild carrots and dig them up and compare how sweet they are (the aren’t!) when compared to store bought carrots. Eating wild carrot feels closer to chewing on a stick than eating the fruits a root veggies we are used to.
“Food and Western Disease” is a good book. One interesting tip from the book is that the western diet leads to about 2 extra inches of height (along with the gut fat), so maybe don’t make your kids eat paleo.
Tomorrow’s headlines: ‘Doctor Reveals Rampant Corruption in Scientific Publishing, Creationists Right All Along’
If Big Sugar starts reblogging this article, I quit.
Please don’t Dr. Scott! Would you be interested on a speaking role in a 2 week all expenses paid nutrition meeting in Cancun? Nestlé’s representatives will be happy to answer all questions you might have.
Is lunch provided?
Yes, low fat skim milk. And twinkies. There are special ayahuasca tasting sessions with authentic shamans, to help you break your harmful fat eating habits if you wish.
Please don’t go. The drones need you. They look up to you.
I understood that reference.
Sounds like a joke but considering Climategate was a thing…
Actually, the National Dairy Council’s incentives aren’t what you think they are. Milk producers were (and are) incentivized to support the low-fat craze because it creates demand for skim milk, which is a byproduct of cream production. The value of being able to sell the skim milk is larger than the cost of reducing demand for cream.
(Epistemic status: not-terribly-well-researched sketchy conclusion from previously thinking about possible origins and signal-boosters of the low-fat meme.)
Then how come they’re producing all of this saturated fat research?
Also, see eg https://www.dairynutrition.ca/scientific-evidence/experts-summaries/saturated-fat-and-cardiovascular-diseases-the-role-of-milk-products-in-reducing-risk for what looks like the industry’s opinion.
Also, Google “dairy” in https://en.wikipedia.org/wiki/Saturated_fat_and_cardiovascular_disease_controversy
Being able to sell skim milk was a blessing to dairy processors. Being able to once again sell butter, now that margarine is in bad odor, is also a blessing.
They’d probably rather sell the milk fat as icecream and margarine as butter.
All milk isn’t the same. The amounts of fat and protein vary with the breed of cattle, diet, digestive flora, stress etc. etc.
There isn’t a single best formulation. The desired milk component levels vary depending on the products a given processor sells. Some processors make cheese, some sell fluid milk, some dehydrate and sell whey protein or powdered milk (cheaper to ship), etc.
The choice of which product to produce varies with location and regulation. The price varies within nations by complicated formulas intended to…well who knows, it’s a Goldberg contraption.
All of this is dynamic, varying with time and circumstance. It isn’t too useful to seek simple answers to questions such as why producers and processors might fund research apparently designed to support one view, or its seeming opposite, at different places and times.
That Slate article is memorably terrible.
What happened in the 1930’s and 1940’s that might have affected the market for milk?
The answer: the Rural Electrification Act. Milk is extremely perishable: the advantage of butter is that it is not. A farmer can make butter and sell it on Saturday, but milk spoils in a day or two. Only in a few locations with good access to markets for everyday delivery was fresh milk salable. With electricity, milk can be kept cold.
Refrigeration existed way before the 1930s, and was accessible to large dairy producers.
@baconbacon: But how many thirties dairy producers were large?
Refined sugar is probably not good for you–it seems especially bad for diabetics and hypoglycemics.
Refined fat is probably not good for you–it seems to be a factor in heart disease.
Eating an excess of calories from fat, sugar, or protein is probably bad for you (too much protein may a factor in cancer.)
Eating fat, sugar, and protein in reasonable, moderate amounts is probably good for you.
Beyond that, we’re probably just rearranging deck chairs.
also eat good amounts of fiber
You haven’t really brought anything to the table there.
Strange… I thought “Sugar is bad” is one of the few constants of diet advice. I don’t recall ever hearing diet advice to “eat lots of candy.” There might have been an isolated instance or two of this, similar to the one time someone said that nuclear power will be too cheap to meter, but it wasn’t common.
And yet, there’s a ton of “low fat” food out there that is still high in sugar–low fat chocolate milk, low fat yogurt, low fat ice cream, etc. It’s much rarer to see the opposite–full-fat yogurt with no sugar, for example.
As someone who can eat fat without trouble but can’t eat much refined sugar, this is a constant annoyance.
> It’s much rarer to see the opposite–full-fat yogurt with no sugar, for example. As someone who can eat fat without trouble but can’t eat much refined sugar, this is a constant annoyance.
You picked an item well known for lactose content (dairy). Is it really that hard to find satisfying sugar free items? I’m pretty sure the majority of foods i eat are sugar free or close to it and I don’t even try that hard.
I try hard to find them and it is very hard.
The majority of the foods I eat are sugar free. It is very hard, unless I prepare everything from raw ingredients.
>Is it really that hard to find satisfying sugar free items? I’m pretty sure the majority of foods i eat are sugar free or close to it and I don’t even try that hard.
This probably depends on what percentage of your diet is homemade. Sugar is very common in processed foods, even things that don’t fall into the “dessert” category: salad dressing, soup, applesauce, frozen meals.
Does sugar free mean no sugar or no added sugar? Pretty much any fruit has sugar in it.
Practically nothing I eat on a regular basis has added sugar (occasional exceptions: ice cream, chocolate). But then very little of what I eat comes out of boxes, cans, and packets.
Nobody was ever told to “eat lots of candy”. But they were told to be careful about fat – don’t fry, poach or boil or grill instead; trim visible fat off meat, skinless chicken etc.; reduce consumption of dairy fats e.g. switch from butter to margarine, full-cream to skim milk, don’t use cream, cut down or remove cheese from your diet, and so on.
If you’re getting a lot of your protein from meat, and you’re trying to cut back on fat, you’ll probably cut down on meat as well. So that leaves carbs as the filler, and people did tend to load up on bread, potatoes, rice, pasta.
Persuading people to try and eat more fruit and vegetables and cut down on bread and starchy foods was the reply to that, but as pointed out – fruit naturally has a lot of sugars, and root vegetables have starches as well, so you would need to be aware and juggle your diet to avoid the ‘bad’ carbs and increase the ‘good’ carbs, and people really didn’t do that or cut back on fat and left it at that.
Even trying to work out what you should and shouldn’t be eating is a maze; can you/can’t you eat carrots? how starchy are they? Different websites (and also depending if you’re trying to stick to a low-carb diet or not) will tell you different things.
Back when I was (roughly) carb counting, I worked on a basis that 1 serving = 118g = 15g of carbs for vegetables. Things like 12 grapes = 13g of carbs. Adult GDA (Guideline Daily Allowance) of carbs is 230g, so that 470g of carbs per day is way over the top (I stuck to 165g of carbs per day, with 11 servings of 15g each over meals/snacks – so 2 slices of toasted bread for breakfast was my 30g meal allowance of carbs, for instance, and since I can’t eat in the mornings*, a cup of tea and two slices of toast was more than enough for me).
I had reasonablish results – can’t say the weight melted off, but I lost a bit steadily. Unfortunately, between one thing and another, I fell off the diet and never got back on again.
*As in “can’t”, not “shouldn’t” – I don’t feel hungry in the mornings, eating when I’m not hungry makes me feel nauseated, and in an ironic twist on the advice about “want to lose weight – then eat breakfast, it fills you up and you eat less during the day!”, eating in the morning actually makes me hungrier so, for instance, if I do eat breakfast then I have to have something for tea break, I have to eat a mid-day meal, and then I have to eat an evening meal because now my appetite is aroused and roaring.
eating in the morning actually makes me hungrier
Adding my anecdote:
When I break down and snack between meals (because my body says “c’mon, you won’t be able to concentrate without it, anyway it will be less food for dinner”), I’m generally hungrier at dinner.
Your body quickly adapts to what it’s getting. People on hunger strikes only describe hunger pangs for the first day or so. After that, it’s definitely having an effect on your body, but your body gives up on annoying you to eat food.
My experience as well.
To keep my weight down, I usually have only one substantial meal a day, almost always supper. If I don’t eat until then, I feel a little hungry early then it does away. If I have breakfast, I am more likely to feel hungry later in the day.
That, at least, is the pattern I think I see–lots of random noise, of course.
During the 1920s, ‘eat lots of candy’ was marketed as an effective diet trick.
I thought when I read a history of Kellogg’s that at some point the conventional wisdom was that sugar was supposed to fill you up, but I can’t find it now. The only reason I post this is that I hope it triggers a memory in someone else.
This seems like another artificial controversy to me. It seems clear that both excessive amounts of sugar and fat are toxic to humans. The evidence is not from research looking at foods in isolation. The relevant research is on metabolic syndrome and diabetes mellitus type II. That same line of research looks at the confounding variable like BMI changes, activity level and diet. The other relevant factor is the underlying genetics of the population. There is a genotype for example that does not lead to any significant metabolic changes based on the percentage of fat in the diet as long as the total caloric intake is constant. There are numerous genotypes that lead to poor glycemic control and dyslipidemia as a function of low activity levels and and dietary factors that include the fat and glucose content of the diet.
At a clinical level – monitoring patients for these metabolic effects shows dramatic changes:
1. People with very low levels of activity due to severe psychiatric illness or addictions will have baseline elevations in cholesterol, triglycerides and even glucose that resolve with restoring a normal diet and activity level within a few weeks with no other intervention.
2. Some people with very high fat diets (like fatty meats daily) can end up with massive dyslipidemia (cholesterol > 400, TG 800-1,000) that correct completely with changing diet from fatty meats to fish and salads.
3. People with elevated blood glucose who respond to diet and exercise alone compared with people who develop adult onset diabetes with no significant change in body weight.
4. Some people will require cholesterol and lipid lowering therapies if dietary changes and exercise fail.
There are multiple pathways to metabolic syndrome and dietary exposure to glucose and fats are two of them. When an author talks about eliminating one versus the other for health purposes – they are only telling part of the story for the entire population. There are undoubtedly protective factors that are not considered in the large epidemiological studies. A critical one that I have often observed is people working vigorous occupations (lumberjacks, railroad section workers, furniture haulers, etc) who experience extreme metabolic changes just by retiring or becoming disabled.
The support of the sugar industry for a few studies might not sound so consequential, but it occurred at a time when the medical community really started to deal with nutrition based diseases. They looked for the culprit and identified it as fat, neglecting other nutrients or multiple nutrients. It seemed like a switch that either is ON or OFF, either fat or sugar, and it seems we went in a wrong directions for decades.
It seems to me that the relevant question is, “Does it replicate?” If non-biased researchers confirm the results with additional studies, I don’t care at all that the original study was financed by somebody with a strong incentive to produce certain findings.
I think we’d all be best off acknowledging that biased research is a thing that happens, and even unbiased studies produce statistically unlikely outcomes sometimes, and therefore wait until a study has been replicated before we commit to it as the unquestioned truth. In that regard, “the first study on anything is super-biased” probably helps build support for attempts to replicate. Too many people are trying to do groundbreaking research, and not enough are doing the boring grunt work of making sure that yesterday’s groundbreaking research produced reliable results. I say this not because I know how many researchers are involved in original research vs confirming studies, but because I can see where the incentives lie.
A problem with the epidemiological studies is that they find things like flossing adds 6.4 Years to your life, because people who do what health authorities tell them to do have other healthy habits, are smarter, are more contentious etc.
I just don’t get why “eating too much sugar is unhealthy” and “eating too much saturated fat is unhealthy” would be assumed to be mutually contradictory or even just moderate evidence against each other in the first place. (It’s not like we couldn’t possibly get more of our calories from starches, proteins, and unsaturated fats, or fewer calories altogether, and even if we couldn’t it’s not like there couldn’t possibly be an optimal range of sugar-to-saturated-fat ratios that it would be unhealthy to deviate too much from in either direction.)
I’m surprised to see you not emphasize that if fat or sugar are making people obese, then it basically has to be by changing the body’s self-regulatory processes (the body’s weight ‘set point’, hunger responses, metabolic rate etc.).
You’ve made the point very strongly in the past (https://slatestarcodex.com/2015/01/12/the-physics-diet/), but it seems critical to keep in mind here. Feels like the discourse on the topic is doomed to be terrible terrible terrible forever if it doesn’t keep this fact front and center.
An idea occurs – have a research grant application company/agency/etc. that a researcher doing a project can use that will apply for grants for research from everyone, take a cut off the top, and then pass it all along to the researcher anonymously. You can’t be biased towards the people cutting you a cheque if you don’t know who’s cutting you a cheque, after all, and most profs would rather have someone else do the legwork of grant applications anyway.
The people cutting the checks want the bias, so this won’t work.
Based on all the media reports over the years, you cannot eat anything as it will all kill you in various ways 🙂
Well it’s true – laboratory procedure has also been proven to cause cancer in mice. Life has a 100% mortality rate, after all.
On the contrary, life is the only thing we know that can cause 100% life extension. Mice with it will live for *years* more than mice without it.
But all mice with it will eventually contract death, and none without it do. If that’s not a smoking gun, I don’t know what is.
Correlation != causation
The one flaw in my otherwise perfect argument.
Mice without it don’t contract death, but they do have the symptoms of death.
For what it’s worth, kicking the sugar water habit is probably the best lifestyle change I’ve ever made. And by sugar water, I’m including not just soda, but tea, coffee(though I never developed a taste for the stuff to begin with), non-carbonated flavored drinks, and even real fruit juice. Bottled Water now makes up 99% of my fluid consumption outside of fluids included with my food(I don’t drink tap water due to horrible quality).
My diet is largely crap otherwise though as I’m largely dependent on the microwave and foods that can be eaten straight from the package. Wish I knew how to grill steak, burgers, or boneless chicken breast.
Slow cookers are stupid simple and more logistically convenient. Slow-cooking steak would be a waste of good steak, slow-cooking ground beef potentially hazardous, but briskets and chuck roasts cook up fine.
You throw them on the grill and then flip them around when the upper side is warm, that’s all there is to it.
I would recommend doing the steak more raw, but for the other two, that’s all you need.
No significant sugar in coffee (or tea) unless you add it. I usually have mine with just half-and-half, which has a little sugar. Probably one of my more significant sources of dietary fiber, as I don’t eat most vegetables.
Grilling steak or burgers is easy. Preheat grill. Reduce heat to medium high, add burgers or steak, grill for a few minutes, flip them over, let them cook some more. You can pre-salt a steak with coarse salt if you like; I like McCormick’s Montreal Seasoning for burgers or flat-iron steaks. Amount of time depends on grill, thickness of burgers, how done you want it, etc. For small steaks or burgers I usually do about 4 minutes on the first side and 3 on the other to get to medium-rare.
For burgers you can cook them frozen (longer) but it makes it hard to get the right done-ness. For steaks, definitely defrost first.
Only some steaks (mostly the more expensive) are suitable for simple grilling. Sirloin, NY Strip (also called club, cowboy, shell, etc), ribeye (Delmonico), T-bone, porterhouse, tenderloin (filet mignon), flat-iron. You can grill a chuck steak but you need to cook it hotter (you want a really good char for flavor) and it’s quite gristly at best and easy to mess up. Skirt and flank steaks need to be marinated first.
These answers assume people have a grill, which is something I associate with outdoors barbecuing, although others may be using it in other senses.
The equivalent with an ordinary stove is the broiler setting in the oven. You can also pan fry meat. Or roast it. Or stew it. Or, as someone suggested, use a slow cooker.
Grill: I think it’s a US/UK thing. In the UK we basically never say “broil” (in fact I think most of us don’t know what the word means), it’s always “grill”. A barbecue is always a barbecue, we never call it a grill. Your grill is typically a separate compartment in your oven, with a heating element at the top (one that gets red hot) – it can be electric or gas, it doesn’t have a temperature setting, instead there’s a dial with numbers on it, or maybe a big flame and a little flame. At my parents’ house, there’s an oven with two compartments, both can work as ovens, but the top one can be put in grill mode – I suppose that’s what you call the broiler setting.
A common UK thing is to talk about putting something “under the grill” – this might confuse people who think that the grill is the wire grid you put your food on. In this context “under the grill” means “under the heating element”.
The arrangement I’m most familiar with for an electric stove has heating elements at the top of the oven, and you broil something by putting it just under the heating elements and turning the control to “broil” while leaving the door a little open.
At least I’m pretty sure that’s how it works–I haven’t actually done it for a long time.
Nowadays most ovens have closed-door broiling, but aside from that, it’s pretty much the same.
Gas ovens are generally not as good for broiling, and mine’s no exception.
Unless you live in Flint tap water and most bottled water are indistinguishable.
Philadelphia tap water and bottled water are quite distinguishable.
For whom? I never noticed the tap water in the Fairmont area had a distinct taste, and the water outside the city (KOP area) isn’t noticeable.
But taste isn’t the same as quality, and most municipal tap water is a long way from “horrible quality”.
For me. Sure, it won’t kill you, but it’s not even close to indistinguishable from tap water. The Water Department even has a FAQ mentioning the chlorine and musty taste:
I’ve found Philly water to have a far stronger chlorine taste than most other places I’ve been.
I’ve found that Philadelphia tap water has improved drastically in the past year or so. It used to have fits of being awful– a sweetish “off” chemical taste or something that would make my tongue sting.
What’s stopping you? Get a grill, practice. It ain’t rocket surgery.
Might boil down to context and ratio. For the most part natural sugars come wrapped in fiber packages. From the perspective of intestinal flora, there may be a hugely significant difference between an apple, and an apples worth of juice.
I’ve never tried this, but I suspect that I could probably start to chew coca leaves and drink poppy tea on a daily basis without significant ill effect, but daily use of cocaine and heroin would be a train wreck.
It might be more about the quality of the calories than the quantity. Eat whole foods. Avoid refined foods.
Widespread use of antibiotics may also be playing a big role in the rise in obesity.
This wouldn’t be particularly scientific, but it would be interesting to put scales underneath the checkout lanes at supermarkets and surreptitiously weigh the patrons and see what kind of interesting correlations show up between what people buy and what people weigh. From this data you couldn’t know how much people eat, but you would know what they eat, and then if you could somehow link their medical data…
People are terrible liars. You can’t just ask them what they eat and get an honest answer.
Daily poppy tea will leave you in a state of awful, protracted withdrawal that may last a month or more. I strongly recommend against it.
Weighing people can comparing it to what they buy would be noisier than you think. Some people buy for families. Some people have bulemia.
Yes, not to mention people eating in restaurants.
Plus some people (like me) buy different food at different stores. A single store would have a very… skewed idea about my diet 🙂
I do not understand why people would call foods that make you fat unhealthy. They are healthy; that is precisely why they make you fat.
Personally, when I look at prices at the grocery store, I look at the nutritional information as well, and buy the things that have the highest calories per dollar.
This is how I buy fast food. I made the mistake of telling my friends this once, and they laughed at me. I mean, how else are you supposed to decide?
As Dave Barry says, “a ‘calorie’ is a scientific measurement of how good a food tastes.”
> and buy the things that have the highest calories per dollar.
So, I’m guessing you live on generic vegetable oil..? X-)
According to The Omnivores Dilemma, America’s food system revolves around of cheap (ie subsidized) corn. Back of the envelope tells me that maltodextrin gives you about 10x the calories per dollar.
This is, of course, assuming that OP is American to begin with which I’m given to understand is only 63% likely.
(Is also assuming that OP buys food and eats it. Eating waste fryer grease is more cost-effective, if disgusting. Better yet, they could win hotdog-eating contests, which would pay them for eating food.)
Your maltodextrin is $1.28 per lb. Bulk canola oil is about $20 per 35 lbs, cheaper if you buy more. That’s about $0.57 per lb and, of course, maltodextrin is basically glucose (~4 calories / gram) and canola oil is fat (~9 calories / gram).
Huh. Never heard of that site before. It will save me a bunch of money on soylent ingredients. Thank you!
Calories per dollar is the main consideration but obviously there are secondary considerations like variety.
“Healthy” does not mean “provides nutrients”. It’s more like “provides an amount of nutrients per serving that is good for health given typical servings”.
McDougall claims that it is more work for the body to convert an equivalent number of sugar calories into fat than it is for the body to put fat calories in your fat stores, which seems intuitively correct. Because turning it into fat is inefficient, the sugar calories (including starch like potatoes, rice, etc.) tend to be burned off as energy–heat or movement–compared to the fat calories, but, as everyone knows, fat+sugar=kiss of death because sugar raises the insulin and insulin moves the fat into storage. I think this is the reason people seem to be able to lose weight on either high carb-low fat or high fat-low carb, but not high-carb-high fat (though low calorie overall is probably most important).
He also cites a studying claiming to show that simple sugar calories (like in soda, white sugar) are no more likely to cause weight gain, all else equal, than complex sugar calories in, e. g. potatoes.
Overall, I think the real cause of the supposed failure of low-fat diets is that people just added the low-fat, high-sugar snackwells cookies to their roast beef sandwiches, like the people I see at the mall who drink a fruit smoothie as a beverage to accompany their General Tso’s. As a meal, a fruit smoothie is very healthful and promoting of weight loss. As a beverage to accompany a high-fat meal, not so much.
I agree that this isn’t a smoking gun, and obviously “study finding A was biased” is not evidence for !A.
That said, there were a handful of lines in that article that looked genuinely ugly, beyond the usual bounds of funded research. Specifically:
As far as I can tell, this wasn’t just a sponsored study, it was a back-and-forth where early drafts got industry approval and edits before publication. That may not invalidate the data any more than usual for this constantly-sponsored field of study, but it does read as an uncommonly serious ethical lapse by the researchers.
Comparing the weight of evidence found/created by fat advocates and sugar advocates will not yield the truth. Ever.
Some unbiased source is required. Period. Good luck with that.
As with some other areas of science it isn’t the food science that is the real problem, it is the * reporting * of food science that is the problem.
Journalists who work at ESPN are no doubt big sports fans but I don’t think this necessarily affects their coverage too much and there are some pretty definitive measurements in sports that can’t be altered to suit an agenda “Re-analysis shows Seattle Seahawks actually lost last week!”.
Food journalists are different, they tend to be “foodies” or activists of one sort or another and this tends to bring bias to their coverage. One example would be the total disdain for GMO’s and the emotional embrace of organic foods. From a food science perspective they are sending mixed and wrong messages. There are numerous recent examples of certain dogmas being held far longer than the evidence supported. The selection bias of what is reported has reduced credibility to the point where I barely trust anything they say at all anymore. There seems to be some convoluted war on meat from it causes cancer to it is leading to a global meltdown.
Food journalism is probably a pretty boring area to work in. All the low hanging fruit have been picked and real progress now requires seriously large long term studies. They have to do something with their day so they churn out effective tabloid articles on food that abuse just about every area of statistics there is.
I’m surprised anybody is listening.
The problem with this sort of thing is that SOMEONE has to pay for research. Money doesn’t just fall from the sky. And a lot of research is, quite logically, funded by interested parties. Who else is going to devote the money to such research? The NSF only has a limited amount of funding to provide (and isn’t as unbiased itself as one would hope).
The logic behind the “Big Sugar funded this research!!!” is: IF a researcher accepts money from Person X, THEN their results will be skewed in the direction of Person X. There’s a strong “nudge-nudge-wink-wink” implication that the phrase “…and the research is therefore hopelessly biased” is added to the end of it. On a personal note, I have nothing but contempt for people who wish to make such accusations by implication. Fraud is serious business in science–it can destroy a scientific career–and as such should be addressed with a certain amount of openness and seriousness that Argument from Implication does not allow for.
First and foremost, it’s not a given that the logic is true. Nearly every environmental remediation study is funded by some group paying for the clean-up, for example (read the CERCLA implementation guidance, it’s the expected procedure–the EPA oversees, but does not fund, this stuff). If the logic was true, it would mean that every CERCLA site study in history is flawed, with the implication that every company and government agency who’s gone through the CERCLA process is guilty of fraud. This is obviously insane. Ergo, we cannot assume that merely receiving money from an interested party renders the research unusable.
In other words, we have to read the research ourselves and draw our own conclusions.
Second, even if the logic (but not the implication) was sound, it wouldn’t lead to the conclusions people using it wish us to draw. Bias is inherent in science, and stems from numerous sources. In paleontology, the study of taphonomy is an entire field devoted to studying the biases in the fossil record–meaning the biases in the theoretical dataset we have the capacity to sample from as paleontologists. If bias inherently destroyed the capacity for data to be useful, the entirety of historical science–astronomy, paleontology, archaeology, geology, all of it–would have to be declared invalid. Wholesale. If you are willing to drive on a bridge that rests on dirt in any way, shape, or form, you accept that this argument is false (compaction testing being an application of geologic principles).
The trick is to acknowledge actual and potential biases and interpret the results with those in mind. One common way to do this is to set the bar higher for the direction the bias would swing the results. If research paid for by the milk industry very slightly supported the notion that fat was healthy, it’s probably due to random noise–or at least can be treated as such until further data are available. If research paid for by the SUGAR industry very slightly supports the notion that fat is healthy, that’s something to take notice of–it obviously contradicts what the bias would lead us to suspect.
I am sick to death of people complaining that science is broken (or that subfields are) merely because results in publications require thought before they can be applied. And this is an example of that kind of non-thought. Bias happens. Fraud, while relatively rare, happens. We’re humans–passionate humans, vigorously defending our ideas. We’re going to make mistakes. I’ve no sympathy for frauds, but the simple truth is we have to be willing to accept a certain amount of fraud as the price for using humans as scientists (and there’s no other option right now–we accept this, or we abandon science entirely). The ONLY defense we have, the only one we’ve EVER had, is the intellectual rigor of the individuals reading the reports. If we all commit to rigorous reading of scientific papers, such biases as would come from funding would have limited affect–and we will have the tools ready to hand to counter it when it does.
There’s a huge difference between “the funding for this research came from the industry” and “the industry paid for the research, the industry provided the specific papers they wanted to have reviewed and what they wanted the result of the review to be, as time went on the industry continuously received drafts and provided feedback and editing as well as sending additional articles that needed to be refuted by the review”.
This is not even a case of “bias might be creeping in”. By all accounts this is a case of “Here is what we want you to say, please put your name on top of it and do the work needed to make it look good”.
The gold-standard for clinical trials is a double-blind study. Something similar may be needed for funding of reviews.
I’ve lost a lot of weight (100+ pounds) twice in my life, once on a low carbohydrate ad libitum diet and once on a more traditional calorie counting diet. I found the LC diet easier because because fat and protein are more satiating at equivalent caloric values and being in ketosis has a mild appetite suppressant effect, so you’re less hungry. It’s more boring though because of the heavy restrictions on what you’re allowed to eat.
That’s what I tell people now if they ask me whether they should count calories or restrict carbs – it’s basically a choice between hunger and boredom, so just decide which one you hate more. For me personally it’s hunger but I know people who find boredom worse and they do better with calorie counting.
Two beliefs make this article big news. First the theory that high glycemic food = insulin = hunger = high glycemic and so on. This is the low carb or pro fat or anti sugar argument in a nutshell. Second, that for years children were eating a bowl of Frosted Flakes or Cheerios (with skim milk) and a glass of oj for breakfast because the food pyramid led mom and dad to think that was a healthy breakfast.
If you believe both the above are true, you have terrible villains who convinced parents to hurt their own children.
My vague impression is that America’s sugar plantation owners invest more in protectionism to keep out cheaper foreign sugar than in promoting sugar, per se. Sugar kind of sells itself. As somebody who was a child with a sweet tooth in the 1960s, my memories are that the adult world was constantly citing the Authority of Science as a reason I couldn’t have more candy.
In contrast, the corn lobby (and to a slightly lesser extent the wheat lobby) was more dominant in molding the nutrition propaganda of the second half of the 20th Century.
Here’s a theory I’ve never heard before about the increase in obesity (so it’s probably not a very good theory, but here goes):
When I was a child in the 1960s, sugar was seen as leading to cavities, which required painful drilling. I remember the dentist finding seven cavities in a single visit. Ouch.
But then I stopped getting cavities, probably due to fluoride in the toothpaste. Subsequent generations of children didn’t associate candy with cavities, so they felt less fear of painful consequences due to consuming more sugar.
Hah! No fluoride in Hawaii and we have the second lowest obesity rate! The science is settled.
Obviously, whether cavities are there or not, obese and at risk of obese children should undergo dental torture. Problem solved, epidemic averted. Do it for the children.
There is not a shred of evidence that people who eat carbs or drink a Coke a day and are lean suffer untoward health effects. This is all about blaming sugar for the obesity problem in the USA .
I love your writings, and I think this position is perfectly reasonable if you’re starting with little more information than this one NYTimes story, but given a whole lot more information, this position is clearly wrong.
Dietary fat is not only harmless, but actively healthy. There never was any real scientific evidence otherwise. Real scientists correctly called that out from the beginning, and the entire edifice of modern nutrition is a pseudoscientific sham. The salient bit about these new revelations is just that it was an intentional sham — in which the “scientists” who constructed that edifice were secretly lining their pockets — instead of merely an emergent, non-profit pseudoscience.
The shortest article that I’ve yet found which gives a sufficient grasp of the enormity is this one:
“figuring out who’s distorted the science more is the sort of project that’s going to take more than one article.”
Having done that project, I can assure you that the answer is clear. When you read Ian Leslie’s article you’ll probably think “He’s cherry-picking, or slanting, this story to make it more outrageous. The real, objective truth of this story can’t be this bad.”. But I assure you, he’s not. His article is accurate and fair.
I’d be happy to feed you more details if you want to inquire, or challenge this narrative. Because I love your writing, and because I care a lot about the truth coming out, here, and how much good this truth can do for people.
No true scientist
I don’t know if I want to be that categorical. Yes, the anti-fat people are full of the brown stuff and that has been widely known for at least a decade (I think Taubes was the first popular account of the shenanigans of Ancel Keys and company). But humans are complicated and different. Finding the One Bad Thing to hate (see Lustig and fructose) isn’t a reasonable solution either.
Yes, an average American would benefit from eating considerably less simple carbs and more fat. But everything is good in moderation and VLC diets for a long time are problematic, too. You have to get most your energy from carbs or from fat (you can’t get it from protein) and the proper ratio is subject to debate but it’s highly unlikely to be something like 1% to 99%. Plus the individual differences can be large. Plus different kinds of fat (or of carbs) have rather different physiological effects. As I mentioned, it’s complicated.
I agree, though, that the fat-is-bad debacle really underscores how far from actual science most nutrition studies are.
The latter doesn’t imply the former, e.g. you could get 35% of the calories from carbs, 30% from proteins and 35% from fat.
Getting energy from protein isn’t great. You’re basically overworking your liver and have to deal with a bunch of nasty metabolites for no good reason.
Make that 45%/10%/45% then. So long as you are getting a nonzero number of calories from non-fat non-carbs, it’s not mathematically impossible for you to not get most of your calories from fats nor from carbs (though the fewer calories from non-fat non-carbs you get the more fine tuning that’d require).