Ricochet is the best place on the internet to discuss the issues of the day, either through commenting on posts or writing your own for our active and dynamic community in a fully moderated environment. In addition, the Ricochet Audio Network offers over 50 original podcasts with new episodes released every day.
Charmed Substances
It’s been thirteen years this month since John Ioannidis published an article in PLoS Medicine entitled Why Most Published Research Findings Are False. Arguably, this article kicked off interest in the replication crisis. He’s back again, this time in JAMA, to warn us to be skeptical of nutrition research, as if we needed that warning. “The field needs radical reform.”
The majority of nutritional advice advocating the consumption or avoidance of specific foods (superfoods, killer foods) is based on epidemiological studies. Ioannidis notes that “…almost all foods revealed statistically significant associations with mortality risk.” Yet the causal connection is rarely established even as dietary recommendations are made based on such studies. Taken at face value, the results lead to unlikely conclusions that fail the sniff test:
Assuming the meta-analyzed evidence from cohort studies represents life span–long causal associations, for a baseline life expectancy of 80 years, eating 12 hazelnuts daily (1 oz) would prolong life by 12 years (ie, 1 year per hazelnut), drinking 3 cups of coffee daily would achieve a similar gain of 12 extra years, and eating a single mandarin orange daily (80 g) would add 5 years of life. Conversely, consuming 1 egg daily would reduce life expectancy by 6 years, and eating 2 slices of bacon (30 g) daily would shorten life by a decade, an effect worse than smoking. Could these results possibly be true?
One obvious logical error is that “much of the literature silently assumes disease risk is modulated by the most abundant substances; for example, carbohydrates or fats.” Since foods consist of dozens or hundreds of chemical constituents, this is naive. Even “seemingly similar foods vary in exact chemical signatures.”
Ioannidis observes that there is a precedent for such foolishness.
For decades, genome linkage scans struggled to link large chromosomal areas to disease risk. According to current knowledge, these previous efforts were doomed: each chromosomal area contains thousands of genetic variants. Linkage scans resulted in numerous articles, but limited useful information.
In short, the early work was bunk, and for similar reasons. There may be more wisdom about nutrition in a 45-year-old movie.
My favorite charmed substance is the egg. What are yours?
Published in Healthcare
My mother ate traditional Southern food all her 95 years. Never dieted. Never exercised. She had a lovely figure in her middle years that melted into skinniness in her later years. I went on my first diet at about 15, hoping to lose all of 3 pounds. I felt my ideal weight was 102, so the scale showing 105 horrified me.
Years of dieting has rendered me…fluffy, shall we say, and diabetic. I watch what I eat and I exercise religiously. I hope it’s doing something for my health, because I know it has no effect on my avoirdupois. Mama did it the right way.
I think it is not unreasonable to suggest that limiting calories in WILL result in less weight gain. People who have fewer calories available, in broad strokes, are thinner than those with more calories available.
Whether those calories come as sugar or cream or what germ certainly matters, of course – but I agree there is much that is clearly not as well understood as nutritionists would have us think.
Still: there are some pretty obvious causalities at play: Too much food make you fat. Too little food makes you skinny.
I eat 5 eggs for breakfast and 3 slices of bacon for lunch every day (always the same). I’ve been on Atkins for 10 years now and the weight has stayed off, no health problems. I don’t believe dieting research, and I don’t believe any amount of alcohol is healthier than no alcohol at all.
Watch me.
You don’t understand. If she hadn’t eaten bacon every day, she would have lived to be 324,850 years old.
Haha
This is just making excuses in the face of evidence that some foods are less healthy than others.
A small number of subjects does not mean the study has no value. Quite the contrary. If there is a statistically significant difference detected in a study with small numbers, that means that even an underpowered study found evidence for harm.
Also, 1.2 equates to 120 percent. A 20 percent higher risk for one group over another.
And here’s another issue to consider: If the people surveyed are relatively homogeneous, it can be a little bit difficult to detect harm. For example, if you study people who smoke 1 to 2 packs of cigarettes each day, it might be hard to detect harm from smoking 2 packs versus 1 pack. But if the study results still show harm, that is even more damning evidence of the harm of smoking.
Same in nutrition. Nearly all Americans eat meat, most eat meat nearly every day. The fact that statistically significant harm can be found from meat eating can be found in a population where nearly everyone eats meat is even stronger evidence of harm.
And this idea that, “Hey, I still have a pulse. That proves my diet is healthy” is not exactly an example of scientific rigor.
I understand that as a doctor, you don’t want to tell people to eat more spinach and less ice cream. And maybe that really isn’t your job. But we shouldn’t lie to people and tell them that ice cream is just as much of a health food as spinach is. In other words, you don’t have to be a food nag. Basic honesty is all that is needed.
You can’t expect the meat industry to be honest. They need to make a living.
This is outside of my area of expertise but I’m pretty sure the mechanisms of respiratory damage from smoking are understood. Thus, a causal relationship has been established. In short, your statement is false: epidemiological studies are not the only basis for recommending against smoking.
Not that kind of doctor, but OK.
Same for nutrition. Epidemiological studies are not the only basis for recommending that people eat less meat, less saturated fat and more whole fruits and more vegetables.
And in the case of nutrition, there have been randomized intervention trials whereas no intervention trials were attempted with cigarette smoking.
We should look at the scientific evidence, not make excuses for why our favorite foods don’t seem to be as healthy as broccoli and winter squash. Sure, the truth isn’t always fun to learn. But we are better of knowing it in any case.
Health care professionals sometimes need to tell their patients what they need to hear, not just what they want to hear.
Nina Teicholz can sell a lot of books by telling people that fatty foods are good for you. That’s good for her bank account. But if you’ve had a heart attack and don’t want to have another one, you need the truth, not feel good Woody Allen movie clips.
Never mistake quantity for quality. The replication crisis encompasses tens of thousands, perhaps hundreds of thousands, of papers. Science is not a democracy. One good study is superior to a thousand bad ones. Since you seem to agree that all the studies you’d quote are open to valid criticism then the answer is better studies, not more of the same.
Harking back to Ioannidis’s example of genetics research, there were certainly many crude genome linkage papers published before sequencing. As he noted, they were basically useless.
It only would have seemed like it.
All without any causal understanding: correlation only. That was the point.
I’d be interested if you had any counterpoints to the arguments made by Ioannidis as described in the OP and in his papers. So far, you have simply repeated standard talking points. Have you read any of the references?
As someone once said, all these health nuts will feel pretty silly some day lying there dying of nothing.
Well, I want some mechanisms. I do not want to know which food is good, I want to know why. While my MS is not specialized in metabolic biochemistry, I can follow a review article just fine. If nutritional science is solid, we should be able to pin down the causative agent. What triggers a higher basal metabolic rate?
We have mechanisms for the carcinogenic effects of Benzopyrene and other PNA/PAHs, after all, and they are components of tobacco tar.
This is not in dispute. Fortunately, we understand the thermodynamics of metabolism well enough to be able to establish a causal relation between caloric intake and weight gain. The subject of the OP is quite different.
For myself, I’ve learned that if I’m always hungry I lose weight.
Sometimes you feel like a nut, sometimes you don’t.
Studies? How about Cheese! Glorious Cheese! Hard and soft, aged and fresh, buttery and skim, pungent and mellow. A whole world’s worth of protein/calcium/flavor/fat, it’s fantastic stuff. (but never, ever goat)
I am not saying that the studies I could link to are bad studies. I am just saying that all studies have limitations, which are usually discussed in the studies themselves.
For example, let’s say we want to do a diet trial on heart disease. We could decide that the only people we will study will be people who have suffered and survived a heart attack within the last 2 years. Then we could randomize them into two groups, an intervention group and a control group. And we could study them for 2 years.
When we report the results of our study, we might mention the fact that our study participants, heart attack survivors, are not representative of the entire population.
Also, other questions would have to be asked: Did the study participants randomized into the intervention group actually follow the intervention diet? Or did they drift back to their prior eating ways after a month or two? Did the control group modify their diet because they knew they were participating in a heart disease study? Was there a high dropout rate? Were the people who dropped out of the study different than those who stayed in?
These and other questions are all legitimate questions to ask of diet trials.
Still, when you look at lots of scientific studies on nutrition, it’s clear that their are certain dietary patterns associated with bad health outcomes and other dietary patterns associated with superior health outcomes.
But let’s face it, even many doctors prefer ice cream to broccoli. Most doctors don’t go to PubMed.com reviewing all kinds of studies on nutrition. If they visit PubMed.com at all, they are more likely to look at drug trial studies.
Doctors are trained in pills and procedures, not the knife and fork. So, no wonder that they often give bad advice or no advice in nutrition to their patients.
Also, a doctor is more popular with his patients if he tells them what they want to hear than if he tells them that they could improve their health outcomes if they changed life long habits. Life long habits are very difficult to change.
About 100 years ago, a doctor in the Netherlands was assigned to be in charge of hospitals in the Dutch colonies. So, he went to the colonies and he was surprised that there were no heart disease patients in these hospitals. He did some blood draws and checked the serum cholesterol.
He found that the serum (blood) cholesterol in this population was quite low compared to the serum cholesterol in the people who lived in the Netherlands. He speculated that the difference might be genetic. But he also realized that the people in these colonies, being poor, didn’t eat very much meat, dairy or eggs at all. Their diet was basically vegetarian due to historical and economic circumstance.
So, he conducted a feeding trial. He put one of the people from the colonies on a Dutch diet, with generous amounts of meat. Then he tested their serum (blood) cholesterol again. The results were higher than when they were on their near vegetarian diet.
This was one of the first indications that many of the diseases that people in Western nations suffer from are heavily influenced by dietary patterns. Change the dietary pattern, change health outcomes.
Now, that’s just one doctor, who published his results in Dutch about 100 years ago. But since then, there have been tens of thousands of other studies published in the peer reviewed scientific literature. So now we know so much more about the impact of food on health than we did 100 years ago.
Therefore, we should not put our heads in the sand and pretend that “the experts were wrong” and “those studies are all bad” and ice cream is health food.
That might be good for a laugh in a Woody Allen movie. But it’s not funny when someone you love is being told to continue on a bad dietary pattern and suffers and/or dies as a result.
Just google “Medical Marco Polos” and you’ll find other stories of doctors similar to the one I described above.
I just googled Marco Polo …
https://www.youtube.com/watch?v=rtSfY-CWT-o
So not only do we have the lies of the food pyramid to blow off, now we’re supposed to care about a study conducted 100 years ago? I suggest people do their own research. Do not listen to anyone else quoting endless studies, go find them yourself, check sources, who funded the study, dig deep. The information is out there, be open minded and do your homework. I’m disgusted to hear the nutritional party line repeated again and again by the same person, again and again.
Writing unnecessarily long responses to justify a predetermined position doesn’t make a person right.
Vegetables aren’t food. Vegetables are what food eats.
We never had margarine in our house growing up. My mom refused, as it was all she had for about a decade during and after WW2 in Germany.
The dangerous combination seems to be consumption of refined carbohydrates and fats, particularly trans fats and excessive polyunsaturated oils. Probably also an imbalance between intake of omega 3 and 6 fats, as well. Note that both tend to be polyunsaturated, aka more prone to rancidity.
You’ve managed to string together a couple of straw men. To your points,
What’s missing is the causal connection with disease. I’m sure you understand that cholesterol is a essential to life and that human cells synthesize it. It would never be a goal to reduce cholesterol to zero. You’re probably also aware lipids transported by HDLs are not equivalent to those transported by LDLs. Hence, total serum cholesterol is not a meaningful metric of anything by itself.
What you may not understand is that serum cholesterol and dietary cholesterol are weakly connected. Serum cholesterol saturates at fairly low levels of dietary cholesterol (cf. Effects of dietary cholesterol on serum cholesterol: a meta-analysis and review). In other words, increasing cholesterol intake beyond a low level has a slight effect on serum cholesterol.
The trick of just invoking cholesterol! as some sort of demon substance has worn a bit thin. Almost everyone understands that there’s a bit more to it. If you want to proselytize for vegetarianism, be up front about your agenda.
*The last clause of this sentence is a joke. For some reason, I feel this needs to be pointed out explicitly.
@HeavyWater
I was asking for mechanisms. I am sure that you are aware that cholesterol is not primarily derived from the diet, but it is synthesized in the liver? You keep on talking about dietary studies using various epidemiological metrics like cohort and case-control studies. Those can be useful if confounding factors are properly controlled. You do not seem to look at study quality or the biochemical cause for a given nutritional finding, which is what people were asking about.
For reference: are you a vegetarian or vegan? In my experience, many vegetarians have a religious zeal in conveying their message that makes me want to have a steak right there.
If you’re looking at classic hunter-gatherers, Loren Cordain researches their diets extensively. Apparently the tribes living that life on the Great Plains consumed over well over 100 species of plants in addition to what their food ate.
If you look at out mechanisms for absorbing calcium from food, they look better at limiting calcium intake than maximizing it.
Now I speculate: How do you get a excessively calcium rich diet? Eating small birds and mammals bones and all, and consuming insect exoskeletons are likely contributors.
Were insects eaten by H. sap. in various locations into the modern era? Yes. And everything old is new again. Behold EXO, the cricket protein company.
Small birds and mammals?
Dormice:
Ortolans.
Fixed it for you.