# Explaining the quirks of childhood BMI

## The body mass index is a flawed measure of your kids' health, but its odd math is proof of our obesity crisis

### Published June 18, 2010 5:19PM (EDT)

A friend of mine said she has a dumb obesity-related question for me: How can 15 percent of kids be above the 95th percentile?

Let me explain.

My friend’s girls -- like most kids nowadays -- are monitored for obesity as part of their health screening at school. My friend looked at the screening results -- in which BMI is expressed as percentile for age and gender (more on that below) -- gave some thought to the definitions of overweight and obesity, and realized that the math just doesn’t make sense. We’re hearing time and time again that obesity has reached epidemic proportions, and that a third of our kids are overweight, half of these obese. Yet, if overweight in kids is defined as having a BMI above the 85th percentile how can a third of U.S. kids be classified as overweight? If obesity in kids is defined as having a BMI above the 95th percentile how can 15 percent of the kids be obese? There can only be 5 percent above the 95th percentile, right?

It’s not a dumb question at all, so I’d like to devote this post to some really basic concepts in the diagnosis of childhood obesity, and solve this question for those of you wondering if -- as in the imaginary Lake Wobegon of Prairie Home Companion fame -- all kids can be above average.

BMI for kids: What is it and why bother with BMI percentiles

Body Mass Index (BMI) is a number calculated from a child's weight and height (weight in kilograms divided by height in meters squared). BMI is a useful and easy screening tool for body fatness for most kids -- it doesn’t measure body fat directly but studies have shown that BMI correlates quite well with body fat in most people.

In adults BMI on its own can assess obesity -- we consider an adult (man or woman) with a BMI above 25 to be overweight, and above 30 to be obese -- yet in kids a BMI on its own tells us very little. For example: A BMI of 21 -- a healthy weight for an adult -- is indicative of obesity in a 6-year-old boy, would categorize a 10-year-old boy as overweight, but would put a 16-year-old well within the range of a healthy weight.

Why isn’t kids’ BMI on its own informative? Kids' body shape and composition change with age. The amount of body fat, muscle and bone transform dramatically with age, and the amount of body fat differs quite greatly between boys and girls. That’s why pediatricians use BMI-for-age charts, in which they plot your kids’ BMI comparing it to kids of the same age and gender.

The BMI percentile allows medical professionals to categorize kids’ weight: A BMI below the 5th percentile indicates underweight. BMI percentiles between the 5th and 85th percentile are considered healthy weight. BMI percentiles between the 85th and 95th percentile are overweight and BMI above the 95th percentile indicates obesity.

Who decides what’s a normal weight?

To establish a percentile chart you’d need a reference population. As we all know, weight (unlike other body characteristics such as eye color, blood type or even height) changes remarkably as a population changes its diet and activity level. Indeed, our kids' weights as a collective have shifted dramatically in the past 30 years.

Therefore, in establishing the BMI-for-age charts the reference population is the kids of the past -- mostly of the '60s and '70s, before the obesity epidemic started. There are several charts used around the world, each developed referencing a different population, but they all carry a historic picture of past generations, when childhood obesity was much less prevalent. When you get your kids’ BMI-for-age he’s not compared to the kids of today, but rather to what experts feel is a "normal" population.

And that explains how 15 percent of kids can plot above the 95th percentile -- they’re plotted compared to BMIs of kids before the obesity epidemic began in full force.

Is a statistical norm a good indicator of health?

Why do we insist on seeing the weight of the past as the norm, and not accept our new dimensions as the new "normal"?

Obesity isn’t merely a statistical description of the extreme upper end of a bell curve. Obesity is a health issue, and we have plenty of evidence showing that kids with excessive body weight have higher blood pressure, higher cholesterol, are at risk of early onset of many diseases -- heart disease, type 2 diabetes, some types of cancer, obstructive sleep apnea, to name a few -- and overall have a shorter life span.

Unfortunately, when it comes to weight and BMI, too many kids are above average. Looking at population surveys over time, we see that kids’ weights, as a whole, have shifted up, and that kids are generally more than 10 pounds heavier than kids 30 years ago .

We all want a healthy future for our kids -- that’s why we need to get childhood obesity under control.