BMI Can Be Misleading

Posted Sept 25, 2013

By Marie McCullough, The Philadelphia Inquirer

Do you know your Quetelet Index?

You may, but perhaps you call it Body Mass Index.

The nifty ratio -- weight divided by the square of height -- was invented in the mid-1800s by Belgian social scientist Adolphe Quetelet. But its use as an indicator of obesity-related health risks dates to a 1972 journal article by American scientist Ancel Keys, the man who renamed it.

"Ancel Keys' emphasis was on cardiovascular disease," said physician Rexford Ahima of the University of Pennsylvania's Institute for Diabetes, Obesity and Metabolism. "He showed a tight association between BMI and the risk for cardiovascular disease. But he used it for population studies. He never meant it to be used as an individual risk assessment."

Alas, that's exactly how BMI is widely used by doctors, public health experts, nutritionists, and other medical professionals.

What's wrong with that? As Ahima discusses in a recent editorial in the journal Science, BMI is a misleading measure of healthy weight on an individual basis.

"BMI does not accurately measure fat content, reflect the proportions of muscle and fat, or account for sex and racial differences in distribution" of abdominal fat, Ahima wrote.

He is hardly the first to fault the BMI. In a 2009 column, "Do You Believe in Fairies, Unicorns, or the BMI?" Stanford University mathematician Keith Devlin pointed out that, based on the BMI, Kobe Bryant, George Clooney, Matt Damon, Brad Pitt, and Will Smith qualify as "overweight." And when the U.S. government brought U.S. definitions into line with World Health Organization guidelines, 29 million Americans went overnight from a healthy BMI to overweight.

Despite controversy, BMI has become such an integral part of the battle against the girth of America that other proposed measures have not caught on. Ahima cites one of the also-rans: the Body Shape Index, which quantifies waist circumference relative to BMI.

Recently, a growing body of contrarian studies have found that a subset of overweight and mildly obese people have a lower risk of premature death than supposedly healthy- weight people.

Ahima sees this "obesity-mortality paradox" as another example of the limitations of BMI. An estimated 24 percent of U.S. adults with normal BMI are "metabolically unhealthy" -- bad readings on blood pressure, insulin, cholesterol. Conversely, 10 percent with obese BMI are metabolically healthy.

The bottom line, he said, is that "health is more than just your BMI."

"We need to broaden our minds," he said. "Thirty percent of adults are obese. That's a huge number. How do we apportion [health-care] resources? We have to risk-stratify even among the obese. We need better measures to determine how fit a person is."


19-24 -- Normal BMI

25-29 -- Overweight

30-39 -- Obese

40-54 -- Extreme obesity

BMI calculator (cdc website):

Contact Marie McCullough at 215-854-2720 or


©2013 The Philadelphia Inquirer

Printable Version     E-mail a Friend
Back Back

    Verisol Bioactive Collagen Peptides