It’s time for some incredibly riveting talk about numbers, kids. If you stick around, I promise I’ll say something funny at the end of the post.

Let’s get started: I’m really fascinated by this question of how we quantify health, and it’s a topic I’ll be coming back to frequently on this blog. Today we’re going to start with the basics, though.

If you read much about fitness or health (which I’m guessing you do, if you’re reading this), you’ve inevitably heard of Body Mass Index (BMI). You’ve possibly even heard that it’s not a great measure of overall health, despite the fact that the medical community continues to use it to gauge individuals’ likeliness towards illness. The simple version of the problem is that the number, which is derived from a ratio involving a person’s weight divided by her height squared, can be skewed to say someone’s unhealthy if she carries a lot of muscle on her frame, or to say someone’s healthy even if she is generally slender, but too much of her composition is fat.

(Thus, the concept of “skinny fat.” No, it’s not just another way for chicks to talk about their frenemies behind their backs. Although it makes you wonder about this general idea of catty remarks overlapping with actual health issues. What disease are badly done hair extensions a sign of?)

Going back in time to the early days of BMI’s inception, you’ll find some interesting tidbits. Like: the formula was developed 200 years ago, by a mathematician who didn’t actually have any medical background. He figured out the formula for BMI to help study large populations of sedentary people—so, in other words, yes, the general population of people with BMIs between 25 and 30 should have a less healthy average than those with BMIs under 25. That is, if the formula were accurately derived.

See, the mathematician in question was working with a group of people that he already had the results on, so he jerry-rigged a formula so it grouped the people in the way he wanted them grouped. Ever read *Hitchhiker’s Guide to the Galaxy *trilogy*?* Most of the series, the characters know that the answer to the question of life, the universe, and everything is 42. They set out to find the question, and in the process start considering what questions might lead to that answer, like, “How many roads must a man walk down?” Eventually they find the true question **25-year-old-spoiler-alert**: “What do you get if you multiply six by nine?”

In other words, the problem with developing an equation that leads to a specific answer is that many formulas can come up with the same answer. (And, like in the case of *HGG, *there can be miscalculations that mean you don’t even get the numbers you want). This conundrum is actually the issue with BMI: the formula’s just plain wrong.

It works for the majority of the population of sedentary individuals, but as you study taller groups, the equation becomes increasingly inaccurate. And not just because “they carry more muscle,” like we read in our health magazines, either. This is true for your skinny-fat tall people as well. Here’s the explanation from Wikipedia, because I honestly don’t know how to explain it any more simply than they have it:

“For a given height, BMI is proportional to mass. However, for a given mass, BMI is inversely proportional to the *square* of the height. So, if all body dimensions double, and mass scales naturally with the cube of the height, then BMI doubles instead of remaining the same. This results in taller people having a reported BMI that is uncharacteristically high compared to their actual body fat levels.” (Link)

Basically, as you begin looking at taller people, their mass/weight actually increases at a steeper rate than BMI allows for—the weight ranges you see on BMI charts should be skewed a little higher. And the people being studied back in 1840 or 1850 when the BMI formula was created? *At least* on average four inches shorter than today.

Of course, we know these problems. Researchers have been aware of these problems for quite some time as well. Yet we continue to apply BMI to individuals—when it’s completely bunk in that situation—and use it for statistical measurements when we know that there are some major flaws in the equation. It makes a lot more sense why we have this whole debate over “fit and fat”—we’re trying to figure out the point at which a person’s size is going to affect her health, but we’re plugging our numbers into an equation that isn’t working. And we see it not working, with the number of contradictory studies we have gracing the headlines. Still, BMI remains one of the most-used variables in research.

To study health, though, we need a means of quantification to compare groups. Some say body fat is the answer—but do we need to consider what this article in the *New York Times *suggests: that even fat may differ in sedentary versus non-sedentary individuals? Or should we consider cholesterol, VO^{2} max, white blood cell count? Turning human beings into numbers ends up being a philosophical quandary as much as a statistical one.

Oh, crap. I’m supposed to say something funny now. No pressure, right? How ‘bout I give you a rain check on that one? Go check out Hyperbole and a Half in the meantime.

Quetelet was primarily an astronomer, is my favorite fact about him.

Because Obama’s health-care legislation didn’t have any meaningful cost control, BMI is going to get more prevalent as a measurement, not less. That’s precisely because of its vagueness and inaccuracy – since it says more people are outside the statistical norm than actually match the statistical norm (with that number getting bigger all the time as we get fatter and, as you pointed out, taller) it’s going to mean more people are ‘leading unsustainable and unhealthy lifestyles,’ or whatever language insurance companies want to use to jack up our premiums and whatever language doctors want to use to terrify us into buying statins or whatever other drug has the cutest rep with contacts at the best steakhouse.

Very true– which is why it’s important that the general public and the media take the time to acknowledge the bias so we have a better chance of understanding the outcomes of research.