4 ms·

And to play devil's advocate, the opposite findings from just this June https://www.sciencedaily.com/releases/2018/06/180622012101.htm https://www.sciencedaily

And to play devil's advocate, the opposite findings from just this June https://www.sciencedaily.com/releases/2018/06/180622012101.htm https://www.sciencedaily.com/releases/2018/06/180622012101.h...

gboudrias 8y ago
Dang, I was about to buy a block of salt to lick for dessert :'(
2muchcoffeeman 8y ago
The article isn’t that biased either way. And it looks like many commenters haven’t read the article. My understanding from reading the article is the lower threshold is higher than we think. There are still negative effects at what we consider high levels of salt intake. They also mention same team who did this study published a previous study that met a lot a criticism and they apparently failed to address the criticisms. So you should probably moderate your salt intake still.
macspoofing 8y ago
Nutrition Science is a huge mess even though it is a core part of human health. As in, it is a total disaster. Here's something depressing to think about: if the Keto people are correct, mainstream nutrition science was a key contributor to American obesity crisis and resulted in countless premature deaths and major suffering over the last few decades. Anyway, recently I wanted to look into whether or not eggs are good for you because those same Keto supporters make noise about how eggs no longer affect (bad?) cholesterol levels and therefore won't contribute to heart disease. You'd think it would be clear. Nope. It's about 50-50 split judging by various links (from seemingly reputable doctors and nutritionists). Even those docs that pay lip service to the changing attitudes towards eggs, don't know what a safe amount is and caution about eating more than X per day or week ... though not because of any specific studies, but just as a precaution. A simple question like "Can I eat 3 eggs for breakfast everyday" has no consensus answer. What the fuck.
- miguelrochefort 8y ago
  What's the solution? Should we heavily invest in Quantified Self as a society? I'm sure we have the technology to make it worthwhile, but I'm not sure most people would be willing to do so due to privacy concerns.
- jarfil 8y ago
  Nutrition is, for the most part, not science at all. Studies are not just financed by obvious interests in the result, but also they use really poor population samples with horrible falsability and repeatability issues. The fact that we still don't have a fully functional human body model to test theories on, doesn't help either.
  - nonbel 8y ago
    "The fact that we still don't have a fully functional human body model to test theories on, doesn't help either." What does this mean?
    - barry-cotter 8y ago
      Model as in model organism, like fruit flies in genetics, or mice in biology. Lacking a model organism means we can’t do much in the way of experiments and that means we learn much slower.
      - nonbel 8y ago
        They said "we still don't have a fully functional human body model", as if one was pending or expected to be developed already.
        mmirate 8y ago
        Well, I think he/she explained rather well the reason why - and the extent to which - such a thing would be useful. That plus knowing the amount of effort expended in the rest of the field of medical research today ... it is indeed surprising that nobody has bothered to even try to produce a human body model. The feat probably wouldn't even require genetic engineering. (Then again, the amount of rote concepts in biology has steered me clear away from studying it beyond a single required high-school class, so what do I know.) My best guess is that there must be some kind of fear, among that field's practitioners, that some stupid politicians (gah, redundancy!) would take one look at the project and throw a hissy fit or worse.
hrktb 8y ago
I found the article hard to read with the constant switch from very assertive summaries and very nuanced quotes, so I might be missing something. Is it actually finding the opposite ? For instance: > Our findings indicate that inaccurate measurement of sodium intake could be an important contributor to the paradoxical J-shaped findings reported in some cohort studies. Epidemiological studies should not associate health outcomes with unreliable estimates of sodium intake Basically, this seems to me to argue that the error range in the methods used would be bigger than the relevant range of results discussed, so to draw any conclusion measurements need to be redone. That would be a bit different from a denial of a specific thesis, although I am sure the researcher must have an opinion on the matter.
Xcelerate 8y ago
Coming from a hard science background, I just ignore nutrition studies altogether nowadays. The reproducibility seems even worse than psychology research. This isn't a knock on the researchers themselves — these are hard subjects to make definitive conclusions in — but we do need to at least be able to accurately quantify the uncertainty on the results. It's not acceptable to have so many studies contradicting each other, particularly when the general public is making lifestyle decisions on the basis of the nutrition findings.
- amanaplanacanal 8y ago
  Nutrition research is really hard. You want a good study, you have to take a large population, randomly assign them to two different diets, then make sure they follow those diets, for some long period of time. This is incredibly expensive. One of the ways people have tried to get around this expense is to take a large population and ask them what they have been eating over their lives. This has two major problems that make it almost useless: People don't remember what they actually ate, only what they think they ate. And there are too many co-founding variables. Some people try to be healthy, so they make lots of changes to their lifestyle, and you can't separate out the effects from diet, exercise, stress levels, and everything else people do.

And to play devil's advocate, the opposite findings from just this June https://www.sciencedaily.com/releases/2018/06/180622012101.htm https://www.sciencedaily.com/releases/2018/06/180622012101.h...

gboudrias 8y ago

Dang, I was about to buy a block of salt to lick for dessert :'(

2muchcoffeeman 8y ago

The article isn’t that biased either way. And it looks like many commenters haven’t read the article. My understanding from reading the article is the lower threshold is higher than we think. There are still negative effects at what we consider high levels of salt intake. They also mention same team who did this study published a previous study that met a lot a criticism and they apparently failed to address the criticisms. So you should probably moderate your salt intake still.

macspoofing 8y ago

Nutrition Science is a huge mess even though it is a core part of human health. As in, it is a total disaster. Here's something depressing to think about: if the Keto people are correct, mainstream nutrition science was a key contributor to American obesity crisis and resulted in countless premature deaths and major suffering over the last few decades. Anyway, recently I wanted to look into whether or not eggs are good for you because those same Keto supporters make noise about how eggs no longer affect (bad?) cholesterol levels and therefore won't contribute to heart disease. You'd think it would be clear. Nope. It's about 50-50 split judging by various links (from seemingly reputable doctors and nutritionists). Even those docs that pay lip service to the changing attitudes towards eggs, don't know what a safe amount is and caution about eating more than X per day or week ... though not because of any specific studies, but just as a precaution. A simple question like "Can I eat 3 eggs for breakfast everyday" has no consensus answer. What the fuck.

miguelrochefort 8y ago
What's the solution? Should we heavily invest in Quantified Self as a society? I'm sure we have the technology to make it worthwhile, but I'm not sure most people would be willing to do so due to privacy concerns.
jarfil 8y ago
Nutrition is, for the most part, not science at all. Studies are not just financed by obvious interests in the result, but also they use really poor population samples with horrible falsability and repeatability issues. The fact that we still don't have a fully functional human body model to test theories on, doesn't help either.
- nonbel 8y ago
  "The fact that we still don't have a fully functional human body model to test theories on, doesn't help either." What does this mean?
  - barry-cotter 8y ago
    Model as in model organism, like fruit flies in genetics, or mice in biology. Lacking a model organism means we can’t do much in the way of experiments and that means we learn much slower.
    - nonbel 8y ago
      They said "we still don't have a fully functional human body model", as if one was pending or expected to be developed already.
      - mmirate 8y ago
        Well, I think he/she explained rather well the reason why - and the extent to which - such a thing would be useful. That plus knowing the amount of effort expended in the rest of the field of medical research today ... it is indeed surprising that nobody has bothered to even try to produce a human body model. The feat probably wouldn't even require genetic engineering. (Then again, the amount of rote concepts in biology has steered me clear away from studying it beyond a single required high-school class, so what do I know.) My best guess is that there must be some kind of fear, among that field's practitioners, that some stupid politicians (gah, redundancy!) would take one look at the project and throw a hissy fit or worse.

hrktb 8y ago

I found the article hard to read with the constant switch from very assertive summaries and very nuanced quotes, so I might be missing something. Is it actually finding the opposite ? For instance: > Our findings indicate that inaccurate measurement of sodium intake could be an important contributor to the paradoxical J-shaped findings reported in some cohort studies. Epidemiological studies should not associate health outcomes with unreliable estimates of sodium intake Basically, this seems to me to argue that the error range in the methods used would be bigger than the relevant range of results discussed, so to draw any conclusion measurements need to be redone. That would be a bit different from a denial of a specific thesis, although I am sure the researcher must have an opinion on the matter.

Xcelerate 8y ago

Coming from a hard science background, I just ignore nutrition studies altogether nowadays. The reproducibility seems even worse than psychology research. This isn't a knock on the researchers themselves — these are hard subjects to make definitive conclusions in — but we do need to at least be able to accurately quantify the uncertainty on the results. It's not acceptable to have so many studies contradicting each other, particularly when the general public is making lifestyle decisions on the basis of the nutrition findings.

amanaplanacanal 8y ago
Nutrition research is really hard. You want a good study, you have to take a large population, randomly assign them to two different diets, then make sure they follow those diets, for some long period of time. This is incredibly expensive. One of the ways people have tried to get around this expense is to take a large population and ask them what they have been eating over their lives. This has two major problems that make it almost useless: People don't remember what they actually ate, only what they think they ate. And there are too many co-founding variables. Some people try to be healthy, so they make lots of changes to their lifestyle, and you can't separate out the effects from diet, exercise, stress levels, and everything else people do.