Hacker News Reader

Comment by learnstats2 | Hacker News Reader

> Yes, person A may eat all the burgers and twinkies, because he's going to the gym all the time

One major problem with the standard BMI definitions is that any athlete in a sport that vaguely requires muscles will be classed as 'overweight' or 'obese'.

Although there may be some justification for this categorisation, these are people we normally consider at the peak of fitness.

One of the best things you can do for long-term health outcomes is to follow a strength training programme. For an average person, that may result in weight gain via muscle. This is not a bad thing.

View on HN

> One major problem with the standard BMI definitions is that any athlete in a sport that vaguely requires muscles will be classed as 'overweight' or 'obese'.

1. This is not a "major" problem, because those people are a tiny minority.

2. This is not a problem with the BMI itself; it's a problem with the "healthy BMI range" interpretation of the BMI value. Obviously, this range does not apply to special populations, like muscular athletes or people with missing limbs.

The BMI concept is valuable because it expresses weight in a way that normalizes for height. The BMI value is only as useful as raw body mass, but at least generalizes to different heights.

A better tool than anything derived from body mass (such as BMI) is a tape measure around your waist: keep your waist circumference less than around 40% of your height. That's it. For instance if you're 68 inches tall, you should have a 27" or narrower waist.

For BMI, by the way, I use the improved BMI formula: 1.3 * mass / (height ^ 2.5). The standard BMI's power of two gives taller adults excessively high BMI values, and is more lenient on shorter people.

https://people.maths.ox.ac.uk/trefethen/bmi_calc.html

keep your waist circumference less than around 40% of your height. That's it. For instance if you're 68 inches tall, you should have a 27" or narrower waist.

I'm kinda skeptical about this. As it happens, I am 68 inches tall and I do have a 27" waist. But I'm really skinny, and even then I feel a bit thicker in the middle than I used to - party l from some additional muscle, partly age (I'm 45 now). I'm very small-boned and have a slightly overactive thyroid, so while I do live a pretty healthy lifestyle I don't think my physique is a reasonable model for the general population.

Well, 40% of height would be right around 30" for me. I'm not sure that I had a 30" waist back when I was running marathons in my twenties; 32", maybe. At this point, I don't think anyone this side of Procrustes could get me into trousers with a 30" waist; yet I think I'm reasonably fit for my age.

The USMC calculator for body fat uses neck and waist circumference both, and by that I was (last time I measured) in decent shape for body fat percentage.

> > > Yes, person A may eat all the burgers and twinkies, because he's going to the gym all the time

> One major problem with the standard BMI definitions is that any athlete in a sport that vaguely requires muscles will be classed as 'overweight' or 'obese'.

Oh, what a major problem it is for the vast majority of overweight people changing their diet and lifestyle. This really is the most important point they have to keep in mind: BMI don't apply to athletes who regularly train and practice. What a bummer.

My BMI is 30.1, which is obese. With a 36' waist, and usually fitting pretty comfortably in to an American size M t-shirt, I am a little overweight, but BMI is not a helpful measurement for any kind of diagnostic for me. I lift the occasional heavy thing, but an athlete I am not.

So you are overweight and your BMI correlates (and a single data point).

For what it's worth I used to wear EU M t-shirts a decade ago and I haven't put on weight. But the same M t-shirt sold these days is way slimmer than my old t-shirts.

BMI doesn't apply to all non-athletes either. My anecdotal evidence is that a few years back, my BMI was below average while I was clearly overweight.

But it's still anecdotal. BMI is a pretty good indicator of obesity for the general population.

You seem to have misunderstood my point.

The general rule that muscle strength is positive for health applies to non-athletes just as well.

I am rebutting the point that BMI is useless because it doesn't always work as intended.

This criticism of BMI is becoming boring, mostly in fitness Internet communities. Yeah, I get it: it's pretty cool to know that there are different kinds of body mass, and - of course - Internet gurus got to show that they know it.

The limitations of BMI are well known; if doctors and nutrition professionals keep on using BMI as an indicator, that's because it's an useful one.

We shouldn't assume other professionals are "dumb"; we shouldn't assume that we are a brilliant bunch, capable of seeing beyond an indicator, while doctors and nutrition professionals use them blindly.

> We shouldn't assume other professionals are "dumb"; we shouldn't assume that we are a brilliant bunch, capable of seeing beyond an indicator, while doctors and nutrition professionals use them blindly.

Um... what? If doctors and professionals are using BMI blindly, that absolutely classifies their opinion as 'dumb'.

The fact is, 'boring' or not, the calculations for BMI most definitely give you skewed results because it doesn't consider muscle mass. As has already been pointed out, this classifies athletes as obese very easily.

This isn't just a 'woe is me' from athletes. It could have real-world implications.

1) What happens when they want to get life insurance, the company blindly calculates their BMI based on height/weight, and classifies them as obese? Tada: Higher premiums.

2) What happens when a self-conscious teenage girl who's very fit and active realizes she's 'overweight' without realizing it's muscle mass causing it? Agree or not, this scenario is a real one and could cause her to eat less, thus depriving her body of the nutrition that she needs.

3) While on the subject of kids... blindly following BMI rules and enforcing standards based on them, schools basically assume that every kid of overweight. As a parent with a child who fluctuates between normal and underweight, it baffles me that he's actively encouraged to eat low calorie meals.

The criticism of BMI is there precisely because it doesn't apply in a large portion of cases. The 'obesity epidemic' doesn't excuse it. It's irresponsible from a healthcare standpoint, and these people are being called out.

> If doctors and professionals are using BMI blindly

They aren't.

> The fact is, 'boring' or not, the calculations for BMI most definitely give you skewed results because it doesn't consider muscle mass.

They most definitely don't give you skewed results. How can the results be skewed if they are straightforward math from the body mass (independently of it being mostly lean or fat)?

Wrong interpretation of the values are what's problematic, not the "values". But, as I said above, professionals aren't using it blindly

> What happens when they want to get life insurance, the company blindly calculates their BMI based on height/weight, and classifies them as obese? Tada: Higher premiums.

Yeah, but this is not a problem with the indicator, nor the . It's a problem with (lack of) honesty.

> the criticism of BMI is there precisely because it doesn't apply in a large portion of cases.

My criticism of the criticistm of BMI is exactly because this is somewhat wrong, depending on what you define as a large portion. BMI applies to most people, and people who really should look at BMI for their job (i.e. doctors an nutrition professionals) are aware of its shortcomings.

> > If doctors and professionals are using BMI blindly

> They aren't.

I see, I misread your original comment. The phrasing just threw me.

> How can the results be skewed if they are straightforward math from the body mass (independently of it being mostly lean or fat)?

That's splitting hairs. I think you knew I meant interpretation.

> this is not a problem with the indicator, nor the . It's a problem with (lack of) honesty.

In fact, it IS a problem with the indicator, because it's being used as basis for large decisions, and it has the backing of the medical industry in spite of it's limitations.

You conveniently skipped over the other issues I mentioned.

> people who really should look at BMI for their job (i.e. doctors an nutrition professionals) are aware of its shortcomings.

You yourself admit that it has shortcomings, but then defend it as what... 'better than nothing'? There are many examples if BMI being used incorrectly by those 'nutrition professionals', so it's hard to see your side of the argument.

> You yourself admit that it has shortcomings, but then defend it as what... 'better than nothing'?

No, I defend it as any other useful indicator (and there are publications on BMI usefulness), to be used by knowledgeable persons. You wouldn't blame "miles per gallon" as something bad because someone evaluated the SUV against the economy-class car standards, right?

A physician looking at a patient might decide to look at BMI, if the thinks it is adequate. A large institution (e.g. an university) might also use it to classify its population as a whole.

Yes, BMI is practical because it relies on two simple measurements, weight and height, and it is most useful when you're talking about populations of people rather than individuals. There are other ways to measure body fat percentage which will give you a more accurate measure for individuals but they're not practical for large populations. The "gold standard" for body fat measurement involves dunking the body into water to measure volume-- this not done at a routine physical exam !

Reply on news.ycombinator.com