12 ms·

I worry sometimes about the actual effects of exercise on our health, and the legitimacy of these studies. It's impossible to detach these studies from the conf

I worry sometimes about the actual effects of exercise on our health, and the legitimacy of these studies. It's impossible to detach these studies from the confounding factors - if your health declines (for any reason), you will stop exercising. I feel like we have to take any studies that use correlations with a grain of salt. How do you study the effects of exercise when there are such massive confounding variables?

nemonemo 7y ago
"Health" is a loosely defined word and there are many goals and metrics. We can definitely study some metrics of "health" and study them with high confidence if the goal is well defined and it can be monitored with fine granularity. For example, if many people exercise and then the blood glucose drops for the next day, we can see its good effect, and probably it's easier to figure out what hormonal mechanism or neural pathway affect them. If the goal is higher-level health issue, like longevity, it is trickier because of, as you said, many confounding issues. I suspect there would be cyclic causality between exercise and longevity/overall health, and it would be quite difficult to separate the cause and the effect. IMO, maybe the best way is, making educated guess based on the studies with clearer metrics.
zingermc 7y ago
Not a real researcher, but perhaps randomized trials could help?
kpU8efre7r 7y ago
Same with diet studies which is why mechanisms inside the body are important to study and identify.
- closeparen 7y ago
  Seems like you could A/B test diet for people getting all their food from an institutional cafeteria. Students, soldiers, prisoners, etc.
novok 7y ago
You seperate out the variables and study people who can exercise, but don’t?
- barry-cotter 7y ago
  People who exercise are different from those who can but don’t, just as those who smoke are different from those who could, but don’t. People who exercise almost certainly have higher energy levels to start with and are less likely to have lingering minor injuries, just as you’d expect smokers and former smokers to exhibit higher levels of addictive behaviour like high coffee or cola consumption than never smokers. Epidemiology and nutrition have massive problems teasing out causation because people who X are different from people who don’t X in many more ways than in just that difference. The health benefits of wine over beer probably reflect socioeconomic differences in consumption for just one example.
  - lm28469 7y ago
    If you think like that you quickly stop looking at studies, it's impossible to control for every variables, even for the most simple phenomenons. We do our best and most of the time it's enough.
throwaway3627 7y ago
For every lifestyle decision, there are always potential risks and potential benefits. For example, people who run excessively tend to lose fast-twitch/bulk muscle mass and wear their hips, knees and ankles faster than most people. For example, one of the worst things you could do to your body would be to repeatedly run fast, downhill while letting your feet slap the ground. And, you're never going to be able to measure all of the pluses and minuses intellectually or quantitively, which is why these choices tend to be primarily qualitatively-based with some quantified-self bits on the margins.
- justinator 7y ago
  For example, people who run excessively tend to lose fast-twitch/bulk muscle mass and wear their hips, knees and ankles faster than most people. Pay for the exam, I'll debunk this one right away. And I believe the whole, "Running is bad for your knees" argument has been debunked, https://omrf.org/2016/09/12/is-running-bad-for-your-knees-surprisingly-science-says-no/ https://omrf.org/2016/09/12/is-running-bad-for-your-knees-su... For example, one of the worst things you could do to your body would be to repeatedly run fast, downhill while letting your feet slap the ground. I think falling off a cliff repeatedly would be a slight bit worse. Or taking up a meth hobby. Or other things that seems fairly obvious. I do a lot of trail running on steep trails (or no trails!) My knees are fine. As mentioned, lots of worse things to do to your body. Here's another one: take up skiing.
  - rimliu 7y ago
    Not the most unbiased source, I'd say. And no link to the research.
    - nradov 7y ago
      Here's a link to research. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5179322/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5179322/
      - rimliu 7y ago
wiremine 7y ago
I'm not a researcher, but it seems your logic goes the other way, too: if you walk and your health improves, then there seems to be a positive correlation between health and walking. And, I think we can control for the health decline. I.e., Stopping exercise because you have cancer vs. stopping because you have broken leg. And even then you can control for changing up the type of exercise. My wife broke her foot last year, and shifted from running 20 miles a week to doing "sit and bit fit" exercise. She actually improved her cardio capacity. Finally, I think the logical conclusion of your argument is: exercise has no impact (or maybe even a negative impact) on health. Studies and every day experience don't seem to bear this out.
voidhorse 7y ago
Well, it's entirely anecdotal, but there's a book devoted to the walking habits of many prestigious philosophers and other artists and thinkers: https://www.versobooks.com/books/1865-a-philosophy-of-walking https://www.versobooks.com/books/1865-a-philosophy-of-walkin.... Sure, any conclusions you can draw from this don't have the rigor of those drawn from carefully accumulated scientific observations and controlled experiments...but it's also possible to be such a stingy devotee of scientism that it paralyzes you and divorces you from truths that are otherwise intuitive and simply a part of being human. If you try to incorporate exercise into your daily routine you'll also have a direct experience of its benefits. We don't necessarily need science for everything and we also don't need to establish proofs for everything. A clouded, unflagging allegiance to the scientific method is what distinguishes a dogmatic acolyte from a rational thinker (just as it does the religious fanatic and the rational practitioner)--it's just as important to circumscribe the space of the actual utility of science just as much as it is to stand firmly by its methodological principles.
moonhorse 7y ago
I exercise a lot. But I do it mostly for the intrinsic fun part. I doubt it has a big impact on our health and longevity. Once a research says regular medium intensity exercise could potentially prolong our lives by 3 to 5 years. But that is it. If you exercise for 50 years, 3 times a week, and 1 hour per session plus 30 minutes preparation, that is actually 12150 hours in total and translates into about 2 years of productive life. So its ROI from a longevity perspective is really not as high as some of us think.
- btrettel 7y ago
  The ROI argument is one reason why I commute by bike. 7 days a week no problem, no real preparation time. Takes up time I would otherwise be wasting sitting in a car or public transit. Last week according to my log I rode for 213 minutes. That's a bit more than 30 minutes per day.
  - koyote 7y ago
    I have done the same and it's great. The main issue is that I used to spend the commute reading books. So I have basically exchanged "mental" excericse with physical excercise. I feel that listening to an audiobook while cycling is a bit too dangerous for an urban commute like mine.
    - padraic7a 7y ago
      I really sympathise with this. I changed job about two yeas ago and switched from an hour long train ride to a 30 minute cycyle. The shorter commute is great, and the in built exercise is great, but man I really miss that chance to read, or just relax.
      - Freak_NL 7y ago
        Why not get into the habit of reading in bed for a while before sleep? It helps the brain wind down, and is easy to maintain as a habit.
    - 7y ago
WhompingWindows 7y ago
You're just asking how do we control for confounding factors in epidemiologic research. There are many ways, included adjusted logistic regression modeling, propensity scores, randomized clinical trials, cohort studies, on and on. I know it may seem like there's too many studies, but you shouldn't worry that there aren't extremely intelligent statisticians and epidemiologists working on these very issues - I collaborate with them daily :)
epiphanitus 7y ago
I hate how hard it is to sort out truth from fiction when it comes to new science regarding health and especially dieting. One study says one thing, another says another. How does anyone sort out fact from fiction on these things if they've never been to medical school?

nemonemo 7y ago

"Health" is a loosely defined word and there are many goals and metrics. We can definitely study some metrics of "health" and study them with high confidence if the goal is well defined and it can be monitored with fine granularity. For example, if many people exercise and then the blood glucose drops for the next day, we can see its good effect, and probably it's easier to figure out what hormonal mechanism or neural pathway affect them. If the goal is higher-level health issue, like longevity, it is trickier because of, as you said, many confounding issues. I suspect there would be cyclic causality between exercise and longevity/overall health, and it would be quite difficult to separate the cause and the effect. IMO, maybe the best way is, making educated guess based on the studies with clearer metrics.

zingermc 7y ago

Not a real researcher, but perhaps randomized trials could help?

kpU8efre7r 7y ago

Same with diet studies which is why mechanisms inside the body are important to study and identify.

closeparen 7y ago
Seems like you could A/B test diet for people getting all their food from an institutional cafeteria. Students, soldiers, prisoners, etc.

novok 7y ago

You seperate out the variables and study people who can exercise, but don’t?

barry-cotter 7y ago
People who exercise are different from those who can but don’t, just as those who smoke are different from those who could, but don’t. People who exercise almost certainly have higher energy levels to start with and are less likely to have lingering minor injuries, just as you’d expect smokers and former smokers to exhibit higher levels of addictive behaviour like high coffee or cola consumption than never smokers. Epidemiology and nutrition have massive problems teasing out causation because people who X are different from people who don’t X in many more ways than in just that difference. The health benefits of wine over beer probably reflect socioeconomic differences in consumption for just one example.
- lm28469 7y ago
  If you think like that you quickly stop looking at studies, it's impossible to control for every variables, even for the most simple phenomenons. We do our best and most of the time it's enough.

throwaway3627 7y ago

For every lifestyle decision, there are always potential risks and potential benefits. For example, people who run excessively tend to lose fast-twitch/bulk muscle mass and wear their hips, knees and ankles faster than most people. For example, one of the worst things you could do to your body would be to repeatedly run fast, downhill while letting your feet slap the ground. And, you're never going to be able to measure all of the pluses and minuses intellectually or quantitively, which is why these choices tend to be primarily qualitatively-based with some quantified-self bits on the margins.

justinator 7y ago
For example, people who run excessively tend to lose fast-twitch/bulk muscle mass and wear their hips, knees and ankles faster than most people. Pay for the exam, I'll debunk this one right away. And I believe the whole, "Running is bad for your knees" argument has been debunked, https://omrf.org/2016/09/12/is-running-bad-for-your-knees-surprisingly-science-says-no/ https://omrf.org/2016/09/12/is-running-bad-for-your-knees-su... For example, one of the worst things you could do to your body would be to repeatedly run fast, downhill while letting your feet slap the ground. I think falling off a cliff repeatedly would be a slight bit worse. Or taking up a meth hobby. Or other things that seems fairly obvious. I do a lot of trail running on steep trails (or no trails!) My knees are fine. As mentioned, lots of worse things to do to your body. Here's another one: take up skiing.
- rimliu 7y ago
  Not the most unbiased source, I'd say. And no link to the research.
  - nradov 7y ago
    Here's a link to research. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5179322/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5179322/
    - rimliu 7y ago

wiremine 7y ago

I'm not a researcher, but it seems your logic goes the other way, too: if you walk and your health improves, then there seems to be a positive correlation between health and walking. And, I think we can control for the health decline. I.e., Stopping exercise because you have cancer vs. stopping because you have broken leg. And even then you can control for changing up the type of exercise. My wife broke her foot last year, and shifted from running 20 miles a week to doing "sit and bit fit" exercise. She actually improved her cardio capacity. Finally, I think the logical conclusion of your argument is: exercise has no impact (or maybe even a negative impact) on health. Studies and every day experience don't seem to bear this out.

voidhorse 7y ago

Well, it's entirely anecdotal, but there's a book devoted to the walking habits of many prestigious philosophers and other artists and thinkers: https://www.versobooks.com/books/1865-a-philosophy-of-walking https://www.versobooks.com/books/1865-a-philosophy-of-walkin.... Sure, any conclusions you can draw from this don't have the rigor of those drawn from carefully accumulated scientific observations and controlled experiments...but it's also possible to be such a stingy devotee of scientism that it paralyzes you and divorces you from truths that are otherwise intuitive and simply a part of being human. If you try to incorporate exercise into your daily routine you'll also have a direct experience of its benefits. We don't necessarily need science for everything and we also don't need to establish proofs for everything. A clouded, unflagging allegiance to the scientific method is what distinguishes a dogmatic acolyte from a rational thinker (just as it does the religious fanatic and the rational practitioner)--it's just as important to circumscribe the space of the actual utility of science just as much as it is to stand firmly by its methodological principles.

moonhorse 7y ago

I exercise a lot. But I do it mostly for the intrinsic fun part. I doubt it has a big impact on our health and longevity. Once a research says regular medium intensity exercise could potentially prolong our lives by 3 to 5 years. But that is it. If you exercise for 50 years, 3 times a week, and 1 hour per session plus 30 minutes preparation, that is actually 12150 hours in total and translates into about 2 years of productive life. So its ROI from a longevity perspective is really not as high as some of us think.

btrettel 7y ago
The ROI argument is one reason why I commute by bike. 7 days a week no problem, no real preparation time. Takes up time I would otherwise be wasting sitting in a car or public transit. Last week according to my log I rode for 213 minutes. That's a bit more than 30 minutes per day.
- koyote 7y ago
  I have done the same and it's great. The main issue is that I used to spend the commute reading books. So I have basically exchanged "mental" excericse with physical excercise. I feel that listening to an audiobook while cycling is a bit too dangerous for an urban commute like mine.
  - padraic7a 7y ago
    I really sympathise with this. I changed job about two yeas ago and switched from an hour long train ride to a 30 minute cycyle. The shorter commute is great, and the in built exercise is great, but man I really miss that chance to read, or just relax.
    - Freak_NL 7y ago
      Why not get into the habit of reading in bed for a while before sleep? It helps the brain wind down, and is easy to maintain as a habit.
  - 7y ago

WhompingWindows 7y ago

You're just asking how do we control for confounding factors in epidemiologic research. There are many ways, included adjusted logistic regression modeling, propensity scores, randomized clinical trials, cohort studies, on and on. I know it may seem like there's too many studies, but you shouldn't worry that there aren't extremely intelligent statisticians and epidemiologists working on these very issues - I collaborate with them daily :)

epiphanitus 7y ago

I hate how hard it is to sort out truth from fiction when it comes to new science regarding health and especially dieting. One study says one thing, another says another. How does anyone sort out fact from fiction on these things if they've never been to medical school?