3 ms·

Just a couple of days ago, I wrote I automatically flag any submission with any kind of "dietary" studies. I'm not saying there is no one study well done, but d

by f1shy 2y ago

Just a couple of days ago, I wrote I automatically flag any submission with any kind of "dietary" studies. I'm not saying there is no one study well done, but doing it well, is just (almost) impracticable. Not only the people have literally no idea what they eat, they forget and misreport, also a human living normal life in the society has just TOO MANY variables. There is no way to keep the other variables like, sport, social interaction, stress and such out of the study.

Sebb767 2y ago
> There is no way to keep the other variables like, sport, social interaction, stress and such out of the study. Sure, but with sufficiently many people, these effects will usually average out. I'm not saying the situation is great, but it's still an important field of study and we need to make progress in some way.
- echoangle 2y ago
  It only averages out if the factors are unrelated though. If a lot of asians eat rice and don’t have a high alcohol tolerance, your study would still show a correlation between eating rice and alcohol tolerance when looking at every single person on earth.
- constantcrying 2y ago
  >Sure, but with sufficiently many people, these effects will usually average out. No they won't. If you have two correlated factors and only measure one of them you can easily get to totally wrong conclusions. If you have a food that is more often eaten by people doing a lot of sports, you will measure that eating that food is correlated with being more healthy. But it would obviously be fallacious to conclude that this food is more beneficial to health than other foods.
- jcranmer 2y ago
  > Sure, but with sufficiently many people, these effects will usually average out. In the case of dietary studies, not really. There are a few factors which are known to have a big effect on your health--being wealthy, active, and moderate in particular--and a lot of the big studies are really just uncovering yet another proxy for those factors. Of course, you can turn that around and make the realization that your diet doesn't really matter: there's no diet that will magically make up for being a couch potato. And outside the main well-known interventions (e.g., eating less calories), the solution is generally to just be more active and things like that rather than trying to tweak your diet.
- f1shy 2y ago
  No if they correlate strongly: people eating more vegetables are more likely to do sport, and care about sleeping. Not to mention visiting a doctor much often. That is just one example.
  - amanaplanacanal 2y ago
    Yeah this is a major issue. The first study that reports a link between some specific thing and health pollutes the data for all follow-up studies, because the folks that care the most about their health are going to change their behavior based on it. So after that you will always see a correlation with all the other things that have been reported to be healthy.
- skirge 2y ago
  Compare people with vegetarian diet from India (over 1 billion, a good sample!) with European meat eaters, what will be the conclusions? Do effects "average out"? Or people drinking alcohol with millions of muslims? There are some obvious criteria which should be used for example divide people in age, income and cultural groups (my grandfather used to eat and did different things I did, including avoiding doctors, despite living in same country and even same home).
bluGill 2y ago
I have long said there are two kinds of diet studies: those that don't apply to you because you are not confined to a hospital bed or prison cell; and those that conclude despite our best effort we couldn't get people to eat their assigned diet.
- chikere232 2y ago
  So what you're saying there's a great business opportunity in people paying to get locked in a diet-cell?
  - bluGill 2y ago
    Maybe, though I suspect there are probably a lot of laws around what you can and cannot do so better get several good lawyers to check what the laws really are around this. Drug treatment programs often work like this so that is the first place to look for laws to watch out for. I've heard of other attempts at things like this. Generally you are not locked into a cell, you are removed to a very remote location by bus so that if you want to leave you have to go through a formal withdrawal process - while waiting for the bus - during which they convince you to stay). They then not only control your diet they also give you exercise (often lead by military drill instructors) thus being a healthier environment than a diet cell. I have no idea how much money they make.
    - f1shy 2y ago
      But even if the people will be confined, you have to be careful to take a broad enough population. I can expect people willing to participate in such study maybe are already orthorexic diet freaks? Or very poor people (which have a diet deviated from “mean”) Doing studies with humans ist just hard!
      - bluGill 2y ago
        The topic has changed from a study to a self selected set of people who want to lose weight badly enough they are willing to pay to be confined in some setting where they cannot access food outside of what is given to them, and they are forced to follow an exercise plan of your (not their) choosing. You can of course study these people, but the only study anyone is interesting in is how different changes in conditions affect how much people lose and how much money you can get out of them.
hombre_fatal 2y ago
We decided that our analytic tech was good enough to figure out that smoking and pollution were bad for us despite infinite confounders. Most people dismiss dietary research because it simply condemns their favorite foods. They accept causal inferences made from epidemiology everywhere else.

by f1shy 2y ago

Sebb767 2y ago

> There is no way to keep the other variables like, sport, social interaction, stress and such out of the study. Sure, but with sufficiently many people, these effects will usually average out. I'm not saying the situation is great, but it's still an important field of study and we need to make progress in some way.

echoangle 2y ago
It only averages out if the factors are unrelated though. If a lot of asians eat rice and don’t have a high alcohol tolerance, your study would still show a correlation between eating rice and alcohol tolerance when looking at every single person on earth.
constantcrying 2y ago
>Sure, but with sufficiently many people, these effects will usually average out. No they won't. If you have two correlated factors and only measure one of them you can easily get to totally wrong conclusions. If you have a food that is more often eaten by people doing a lot of sports, you will measure that eating that food is correlated with being more healthy. But it would obviously be fallacious to conclude that this food is more beneficial to health than other foods.
jcranmer 2y ago
> Sure, but with sufficiently many people, these effects will usually average out. In the case of dietary studies, not really. There are a few factors which are known to have a big effect on your health--being wealthy, active, and moderate in particular--and a lot of the big studies are really just uncovering yet another proxy for those factors. Of course, you can turn that around and make the realization that your diet doesn't really matter: there's no diet that will magically make up for being a couch potato. And outside the main well-known interventions (e.g., eating less calories), the solution is generally to just be more active and things like that rather than trying to tweak your diet.
f1shy 2y ago
No if they correlate strongly: people eating more vegetables are more likely to do sport, and care about sleeping. Not to mention visiting a doctor much often. That is just one example.
- amanaplanacanal 2y ago
  Yeah this is a major issue. The first study that reports a link between some specific thing and health pollutes the data for all follow-up studies, because the folks that care the most about their health are going to change their behavior based on it. So after that you will always see a correlation with all the other things that have been reported to be healthy.
skirge 2y ago
Compare people with vegetarian diet from India (over 1 billion, a good sample!) with European meat eaters, what will be the conclusions? Do effects "average out"? Or people drinking alcohol with millions of muslims? There are some obvious criteria which should be used for example divide people in age, income and cultural groups (my grandfather used to eat and did different things I did, including avoiding doctors, despite living in same country and even same home).

bluGill 2y ago

I have long said there are two kinds of diet studies: those that don't apply to you because you are not confined to a hospital bed or prison cell; and those that conclude despite our best effort we couldn't get people to eat their assigned diet.

chikere232 2y ago
So what you're saying there's a great business opportunity in people paying to get locked in a diet-cell?
- bluGill 2y ago
  Maybe, though I suspect there are probably a lot of laws around what you can and cannot do so better get several good lawyers to check what the laws really are around this. Drug treatment programs often work like this so that is the first place to look for laws to watch out for. I've heard of other attempts at things like this. Generally you are not locked into a cell, you are removed to a very remote location by bus so that if you want to leave you have to go through a formal withdrawal process - while waiting for the bus - during which they convince you to stay). They then not only control your diet they also give you exercise (often lead by military drill instructors) thus being a healthier environment than a diet cell. I have no idea how much money they make.
  - f1shy 2y ago
    But even if the people will be confined, you have to be careful to take a broad enough population. I can expect people willing to participate in such study maybe are already orthorexic diet freaks? Or very poor people (which have a diet deviated from “mean”) Doing studies with humans ist just hard!
    - bluGill 2y ago
      The topic has changed from a study to a self selected set of people who want to lose weight badly enough they are willing to pay to be confined in some setting where they cannot access food outside of what is given to them, and they are forced to follow an exercise plan of your (not their) choosing. You can of course study these people, but the only study anyone is interesting in is how different changes in conditions affect how much people lose and how much money you can get out of them.

hombre_fatal 2y ago

We decided that our analytic tech was good enough to figure out that smoking and pollution were bad for us despite infinite confounders. Most people dismiss dietary research because it simply condemns their favorite foods. They accept causal inferences made from epidemiology everywhere else.