My 1,000 word essay on why BMI isn't useful for a sample size of 1

Themajez

AnvilHead

https://community.myfitnesspal.com/en/discussion/10681147/bmi-agree-or-disagree/p1

stevencloser

Who decided that 25% bf was the cutoff point for excess adiposity? In men that's already pretty damn high imo.

lorrpb

So for 75% the numbers are in sync.

Themajez

lorrpb wrote: »

So for 75% the numbers are in sync.

Bob's body fat percentage is 15%, while Roger's percentage is 25%. Both have a BMI of 22.5.

AnvilHead

I guess it all depends on how you choose to define “useful”.

kimny72

Useful for what?

lynn_glenmont

There were two people in the sample with 0% body fat. Is that really possible? I mean, I assume you had to be alive to get into the sample.

AnvilHead

lynn_glenmont wrote: »

There were two people in the sample with 0% body fat. Is that really possible? I mean, I assume you had to be alive to get into the sample.

Looks like 8 people between 2% and 3% bodyfat too, and a total of around 24-25 people at 5% or less bodyfat. And as @lorrpb observed, it looks like the vast majority of subjects were (unsurprisingly) not in the two quadrants identified as outliers.

kpsyche

It looks like a typical elliptical galaxy to me

AnnPT77

I want to know what the person with 3% body fat and BMI 26 actually looks like.

L1zardQueen

AnnPT77 wrote: »

I want to know what the person with 3% body fat and BMI 26 actually looks like.

amusedmonkey

You would know an extreme outlier at a glance. Most people aren't. No generalized average is really "useful" for a sample size of 1 outlier, including metabolic rate, yet we find MFP useful and don't write off metabolic rate calculations are useless. If someone has a higher/lower metabolic rate than predicted, they just make adjustments to their intake. Similarly, if someone has a higher/lower body fat than predicted, they can choose to maintain at a lower/higher BMI. As a general guideline for most, BMI is useful. It's a large range, not a single number.

I think part of BMI hate is that some people don't like to be told they're overweight (they consider it a negative word that means more than it does). In many (not all) cases, those who consider themselves outliers actually aren't, they just like a weight/body look that doesn't align well with what they want to see on the BMI scale and don't want to be described as having a "character defect".

KrazyKrissyy

I was overweight by 25 pounds in 2016. Am a size 2 and was back then as well. I had a huge blockage in my intestines that had to be removed. Lost the weight immediately. BMI is not helpful.
Edit: Woo= denial. Lol. It's ok if you don't believe me. My medical records are stashed away anyways

livingleanlivingclean

I keep coming back in hope that I find 1000 words.... Where are they?

Orphia

I'm a sample size of 1, and I find the chart pretty useful actually, so thanks anyway.

GrumpyHeadmistress

Beware the lessons learned from small samples...

amusedmonkey

KrazyKrissyy wrote: »

I was overweight by 25 pounds in 2016. Am a size 2 and was back then as well. I had a huge blockage in my intestines that had to be removed. Lost the weight immediately. BMI is not helpful.
Edit: Woo= denial. Lol. It's ok if you don't believe me. My medical records are stashed away anyways

I did not woo, but I can speculate why someone would. It's not that they don't believe you, it's that this is a medical case so it's automatically not the norm, just like protein recommendations being helpful even when there are people who have medical conditions where they need to severely limit protein intake.

Out of curiosity, you're a size 2 and a 25 BMI right now? In that case, you're an extreme outlier even without the medical issues you had.

Nony_Mouse

livingleanlivingclean wrote: »

I keep coming back in hope that I find 1000 words.... Where are they?

I'm guessing they were going for 'a picture paints a thousand words'. 'Cept this one doesn't...

cdjs77

This graphic isn't actually all that useful except for trying to look shocking. How many sample points are there? There is so much overlap in the middle it's hard to tell if the people misidentified by BMI are 50% of the sample or 10%. Also, BMI is useful for exactly what it was intended for: making predictions about individuals and identifying individuals whose BMI warrants further investigation. Of course it misindetifies some individuals, all statistics do, even diagnostic tests do. When you have a diagnostic test that has an accuracy of 98%, that means 2% will not be diagnosed when they actually have the disease, it also means there is another percentage (unstated) of people who will be diagnosed who do not have the disease.
Another point is that you use 25 as the cutoff, and while 25 is generally considered overweight, the real concerns start with obesity or a BMI of 30, which, if you moved the lines, you would notice significantly reduces the error rate.

It's not the BMI is not useful, it's that people do not understand what BMI was meant to be used for. It's a predictive statistic meant to identify individuals at risk of obesity and other related conditions. Being at risk does not mean that you are definitely obese, or definitely going to suffer conditions associated with a high BMI, it does, however, mean the chances are higher. The only reason BMI is a discussion, is because human beings have difficulty understanding statistics, risk and probability, and therefore misinterpret and misapply it. (That, and probably also a bit of being in denial).

stevencloser

All I keep hearing in these kinds of threads is the equivalent to a heavy smoker going "your recommendations are useless because I don't have lung cancer".