3 ms·

Yes. BMI is your weight divided by height squared. Which is really weird when you consider that humans are 3D objects! So you should naturally expect that a

Yes. BMI is your weight divided by height squared. Which is really weird when you consider that humans are 3D objects! So you should naturally expect that a taller person will have a higher BMI than a shorter person even with the exact same proportions. In practice humans who are taller tend to be skinnier so we'd ideally use something close to the 2.5th power but that math would have been impractical for the 19th century clerks who were the first consumers of BMI information.

meew0 4y ago
Humans are 3D objects, but fat is not distributed equally throughout the body. Most fat is subcutaneous fat, i.e. located right below the skin. Its abundance can therefore be approximated by body surface area, which is proportional to height squared.
- naniwaduni 4y ago
  This might be a nice explanation of it didn't fail to agree empirically with reality, where in fact there is a significant cubic component which gives rise to the "something close to the 2.5th power" that GP suggests.
sorenjan 4y ago
A few years ago I read a paper where they checked the correlation between CT measured body fat percentage and a lot of easily measured and calculated figures, in the same style of BMI (w/h²). I haven't been able to find it again, but if I remember correctly BMI was actually pretty good, although worse for women than for men, and the best measurement included waist circumference.
- edflsafoiewq 4y ago
  Waist-to-height ratio is surprisingly good: your waist should not be more than half your height. Simple to measure too, since you only need a piece of string.
- Buttons840 4y ago
  Actually, it appears BMI is more accurate for women than for men. I'm guessing because men have more variability in their muscle mass. Overall, I think BMI looks like a pretty useful metric considering how easy it is to measure. That would explain why it sticks around despite being frequently criticized. See charts 2.A and 2.B in: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2877506/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2877506/
- scoofy 4y ago
  Why do we assume that fat is bad for you, but muscle is always good for you. The blood necessary to maintain excessive muscle is still excessive to normal humans, and may put strain on the heart. Only our skin grows with our muscles, other organs do not grow in tandem. I've always thought that while BMI is crude, it still is reasonable for measuring metabolic syndrome, considering that many of the things associated with heart problems involve the needing a powerful heart, when adding any mass whatsoever will necessarily increase the work the heart must do.
  - jhanschoo 4y ago
    Do you have anecdotes or other evidence of someone experiencing heart issues because they were too muscular? It seems to me that in the usual case where a person develops muscle mass by working out, this isn't a problem because whatever "blood necessary to maintain excessive muscle" is negligible in the face of the body adapting to improve the cardiovascular system in order to serve muscular oxygen needs during workout. Of course, heart problems due to an unhealthy person ramping up exercise too quickly is a concern, but this is separate from your concern. One also observes that athletes in general have lower heart rates than the average person; whatever additional burden their muscles impose, their cardiovascular system is way more adapted to handle that due to exercise.
    - scoofy 4y ago
      My argument is that we should not assume that extra mass from muscle does not impose similar health problems from fat. If we are going to say that, it should be demonstrated. A fit person doesn't need to weight 30+lbs more than standard overweight section of the BMI. If they do, they will be a very non-standard human. Here are some things I found in a brief glance at google scholar. There appears to be a condition called Athlete's heart, but I dunno: https://pubmed.ncbi.nlm.nih.gov/26187713/ https://pubmed.ncbi.nlm.nih.gov/26187713/ https://www.degruyter.com/document/doi/10.1515/jom-2020-0046/html https://www.degruyter.com/document/doi/10.1515/jom-2020-0046...
      - jhanschoo 4y ago

meew0 4y ago

Humans are 3D objects, but fat is not distributed equally throughout the body. Most fat is subcutaneous fat, i.e. located right below the skin. Its abundance can therefore be approximated by body surface area, which is proportional to height squared.

naniwaduni 4y ago
This might be a nice explanation of it didn't fail to agree empirically with reality, where in fact there is a significant cubic component which gives rise to the "something close to the 2.5th power" that GP suggests.

sorenjan 4y ago

A few years ago I read a paper where they checked the correlation between CT measured body fat percentage and a lot of easily measured and calculated figures, in the same style of BMI (w/h²). I haven't been able to find it again, but if I remember correctly BMI was actually pretty good, although worse for women than for men, and the best measurement included waist circumference.

edflsafoiewq 4y ago
Waist-to-height ratio is surprisingly good: your waist should not be more than half your height. Simple to measure too, since you only need a piece of string.
Buttons840 4y ago
Actually, it appears BMI is more accurate for women than for men. I'm guessing because men have more variability in their muscle mass. Overall, I think BMI looks like a pretty useful metric considering how easy it is to measure. That would explain why it sticks around despite being frequently criticized. See charts 2.A and 2.B in: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2877506/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2877506/
scoofy 4y ago
Why do we assume that fat is bad for you, but muscle is always good for you. The blood necessary to maintain excessive muscle is still excessive to normal humans, and may put strain on the heart. Only our skin grows with our muscles, other organs do not grow in tandem. I've always thought that while BMI is crude, it still is reasonable for measuring metabolic syndrome, considering that many of the things associated with heart problems involve the needing a powerful heart, when adding any mass whatsoever will necessarily increase the work the heart must do.
- jhanschoo 4y ago
  Do you have anecdotes or other evidence of someone experiencing heart issues because they were too muscular? It seems to me that in the usual case where a person develops muscle mass by working out, this isn't a problem because whatever "blood necessary to maintain excessive muscle" is negligible in the face of the body adapting to improve the cardiovascular system in order to serve muscular oxygen needs during workout. Of course, heart problems due to an unhealthy person ramping up exercise too quickly is a concern, but this is separate from your concern. One also observes that athletes in general have lower heart rates than the average person; whatever additional burden their muscles impose, their cardiovascular system is way more adapted to handle that due to exercise.
  - scoofy 4y ago
    My argument is that we should not assume that extra mass from muscle does not impose similar health problems from fat. If we are going to say that, it should be demonstrated. A fit person doesn't need to weight 30+lbs more than standard overweight section of the BMI. If they do, they will be a very non-standard human. Here are some things I found in a brief glance at google scholar. There appears to be a condition called Athlete's heart, but I dunno: https://pubmed.ncbi.nlm.nih.gov/26187713/ https://pubmed.ncbi.nlm.nih.gov/26187713/ https://www.degruyter.com/document/doi/10.1515/jom-2020-0046/html https://www.degruyter.com/document/doi/10.1515/jom-2020-0046...
    - jhanschoo 4y ago