Friday, December 7, 2007

BMI: Bad Math Idiots

There is nothing more widely misused then Body Mass Index (BMI). It is used by the government and insurance companies to determine your health and well-being, yet it is probably the most skewed use of data out there.

Basically, the BMI calculation is as follows:

703 x (Body Weight (lbs.)/square(height (in.))

Apparently, this scaled is used to determine if you are underweight, overweight, or just right. If you are less than 18, then you are underweight. Over 25 is considered overweight. I'm not sure where this standard came from, but that's a very small range considering the wide range of people out there. But there are other considerations here, just in the calculation, that need to be addressed:

1) If you'll notice the formula, you'll see that the height of the person has a much greater impact then the weight. If you have your height in inches and square that (or multiply by itself), you'll have a very large value to divide into the weight. The problem with this is that not all height measurements taken by doctors are all that accurate. Rulers can bend or wear down. The doctor could have trouble reading the height as well. In the past couple of months, I've been measured by two different health professionals and gotten two different heights: 6 feet even and 5 feet 10 inches. If you assume my weight remains constant at 245 lbs, then you'll notice that my BMI is 33 for the first and 35 for second. That's a two point different. Given the small range of numbers to shoot for, that's very significant. In comparison to weight, you'd have to lose at least 15 lbs in order to make up for that different, if my height was 5 feet 10 inches. That's a very significant difference, especially for those who can't lose too much more weight.

2) This doesn't take into account muscle mass, even among inactive individuals like myself (that is, individuals whose profession doesn't require frequent exercise). Given that I weigh 245 lbs (that's the maximum; it can be as low as 235 lbs, depending on how much I've eaten in the past month) and that I am 5 feet 11 inches tall (the average of the two above), I should have a nasty looking gut and a waist size of 46 inches. Would you believe me if I told it's about 38 inches right now? Also, I don't have much of a gut, though I could lose the stomach weight. The main thing is that I have thick, muscular legs with very little body fat on them. They support my upper body and I don't have aches from all that support. I am certain that while my body fat is probably higher then is should be, my muscle mass is what throws this whole measurement off for me.

3) An August 2004 article indicated that having body fat is not necessarily a bad thing. While it is a good idea to eat right and get some good physical activity in every day, this doesn't mean that you are going to die tomorrow if you are considered obese. Many health researchers are starting find out things like people are less likely to have a heart attack if they are overweight (greater than 25 and less than 30) in comparison to those who are "just right."

4) BMI doesn't take into account body type, gender, or girth. People have a funny way of interacting with a three dimensional environment. Your height only indicates one measurement of one dimension of space. Your width from side to side and your width front to back should also be indicators as it would probably more accurately indicate whether you are overweight due to excess body fat or a living, breathing Conan. Also, people with slender body types may be misconstrued as being underweight. The mere fact that is uses height squared shows just how flawed the calculation is. Also, it doesn't take into account that women tend to weight less than men in general. Instead, it just paints a broad brush for all people without even considering the biggest genetic difference in the human race. Whoever determined the range was clearly smoking some real good weed.

5) BMI doesn't factor in age. When you measure heart rate, age is a huge factor. But not your body mass index. Nevermind that babies have almost no muscle mass and the elderly tend to loss muscle mass as well. Young people have a different body type than those who are middle aged than those who are just hitting puberty. There are huge differences in the human body's composition as they age. So, BMI apparently not ignores two dimensions of space, it also ignores the fourth dimension we know as time.

This is why whenever a study comes out about obesity numbers in any society, we shouldn't pay attention. Chances are, they are using a sample (and these samples probably aren't that large anyway, but more on that in a later blog) of BMIs.

Instead, I think we should start measuring the volume and mass of the human body and determining the density. Body density is probably a much more valid indicator of being overweight and better at determining all kinds of health risks people face in relation to weight issues. Volume can easily be measured via height, width, and depth. A standard would have to be developed in order to determine the appropriate dimensions, since the human body is not geometrical. Mass can be determined using weight and removing the force of gravity from that measurement (since that's what weight is).

But I doubt there will be much effort to do this as long as insurance companies and government bureaucrats continue to use this farce of a measurement to evaluate public health. So in the mean time, I say that we refer to them all as simply BMIs, which is short for Bad Math Idiots.