By Jen Christensen, CNN

(CNN) -- When Kate Townsend took her kids for their back-to-school health checkups last month, she got a surprise.

The Western Springs, Illinois, mother of three wanted to make sure their vaccinations were up-to-date. The pediatrician did talk about shots, but also mentioned her daughter's body mass index.

"She said to my daughter, 'you are in this percentile, which is in total normal range,' but she made it sound like one more number, and that is too big," Townsend said.

The family was surprised. The girl is an active volleyball player, Townsend said, who grew 2½ inches this year. "My daughter's in great shape. And she was sensitive about this news. I tried to reassure her 'that is just a number' and to move on."

Townsend was right, many experts say, because a growing body of research shows that body mass index, or BMI, may not be the best measure for a growing girl -- or for adults.

An increasing number of scientists are calling BS on BMI.

A tool with a different purpose

BMI is an easy and straightforward tool used to screen for obesity and to gauge whether you should worry about your weight. Divide your weight (in pounds) by your height (in inches) squared. If the resulting number is 30 or higher, by American standards, you are considered obese. You're overweight if your BMI is between 25 and 29.9. If you're under 18.5, you are underweight. The "normal" range is 18.5 to 24.9.

BMI was invented in the 1800s for use in population studies. In 1972, a key study that tested for the best formula to measure body fat percentage in more than 7,400 people found that the BMI formula best measured body-fat percentage as compared to merely calculating weight and height. It was this study that gave BMI its official name -- and its what popularized its use in research. But that study also warned against using BMI to calculate an individual's levels of fat.

"BMI really was a measurement created for epidemiology to give data that was relative and could be used in research," said Dr. Francisco Lopez-Jimenez, an obesity expert at the Mayo Clinic in Minnesota. Up until the 1980s, he said, doctors and scientists had been using a variety of measures to track whether a person had gained so much weight that it could hurt their health. The variety in measurement made it hard to chart trends. And as doctors were noticing that people were getting bigger, they wanted to understand how big a problem it was.

A handful of national and international government agencies then started to recommend that doctors use BMI, as it would be a uniform standard and was seen as more accurate than the insurance company height/weight charts the doctors had been using.

In 1985, the US National Institutes of Health recommended BMI to measure obesity. In 1997, the World Health Organization jumped on the bandwagon. The US Preventive Services Task Force, which sets guidelines for American medical professionals, still includes it as a measure in its recommendations.

It's not only doctors who use it. As the childhood obesity epidemic grew, more than a dozen states started to require schools to chart students' BMIs, with some even recording the measurement on report cards. Life insurance companies may also ask for it to determine whether you should get a policy or are too big a risk.

Yet while the US Centers for Disease Control and Prevention calls BMI "a reasonable indicator of body fat," the agency does not recommend that doctors -- or anyone else, for that matter -- use it as a diagnostic tool.

"Over time, BMI has gained a clinical use, but that was not the original intention behind its creation," Mayo Clinic's Lopez-Jimenez said. "That's because it does have real limitations."

Why do they use it?

BMI is a free and easy way to get a gauge on obesity for research, but as doctors started using it for individuals, they noticed that it doesn't tell the whole story.

"Some people who are labeled overweight by BMI have a broader frame or more muscle," said Courtney Peterson, an assistant professor in the University of Alabama at Birmingham's Department of Nutrition Sciences.

Peterson said that the most exact ways to measure obesity can be expensive, so BMI became the "next best thing we use" for adults. But, it's important to think of it as only one of several screening tools that "helps show at a quick glance that this person's health may be hurt by their weight, so it is an important thing to measure." More measurements would be needed to tell the whole story, Peterson noted.

That's because BMI doesn't distinguish between fat and muscle, which could be a real problem for someone who is athletic. The average football player, for instance, has a BMI of 31.35, considered obese by these standards. Yet many of these world-class athletes are solid muscle, and the average player probably does not have a level of fat that will threaten their health.

BMI can also underestimate the threat for people who are "skinny fat," as some doctors call those who tend to be fit, but have big bellies. Studies show that this belly fat can be more dangerous for health than any other kind, and a study in the journal Frontiers in Public Health found that using BMI as the lone measure missed 50% of cases of people who had what was determined to be dangerous fat that could hurt their health.

Paul Laursen, co-author of that study, prefers the term "overfat" to "overweight," because it really is the fat, rather than the weight, that hurts your health, he said.

"We shouldn't be as much worried about weight," said Laursen, an adjunct professor and performance physiologist at the Sports Performance Research Institute in New Zealand. For example, you could have gone on a fad diet and lost three pounds, but that doesn't necessarily make you healthier. Those three pounds could merely have been water weight.

"What we should really be worried about is the fat part and where your fat is concentrated," Laursen said. BMI misses that level of detail.

Growing children can also be shortchanged. Peterson's latest study found that BMI was off 19% of the time in children. The BMI measurements used for children are weighted for age, but when boys go through puberty, they add muscle at a rapid rate; girls add fat. BMI tends to miss these nuances.

BMI can also miss subtle racial differences. Peterson explains that African-Americans tend to have more muscle than fat, compared with Caucasians. Doctors in some countries have had to adjust their BMI ratios because studies have shown that Asians are at a greater risk for heart disease at much lower BMI than non-Asians.

What should doctors use?

One of the most exact ways to measure obesity is an X-ray, which can distinguish between fat and other innards. But that is expensive, and regular radiation exposure is not advised, so hospitals and doctors are trying a variety of methods and tools.

Some hospitals use a highly accurate machine for a bioelectrical impedance analysis, which runs an electric current through body tissue to determine fat composition. The Mayo Clinic's Lopez-Jimenez uses a Bod Pod, which takes measurements based on body volume. He recommends against the scales or handheld devices available on the consumer market, as they are not precise.

Where kids are concerned, if you can't get a bioelectrical impedance analysis, Peterson suggests a measurement called TMI, or tri-ponderal body mass index. The calculation is weight divided by height cubed. Her study showed that this was a much better measure of obesity for kids between 8 and 17, particularly boys. TMI misclassified obesity only 8% of the time (versus more than 19% with BMI).

For adults, Laursen suggests getting out a tape measure. Measure your circumference at your belly button. If your waist circumference is half your height or less, you are at a healthy fat level, if you are over that number, your fat could put you at risk for ill health.

If you want something even simpler, look at your hip to waist ratio. It's something even a doctor could eyeball quickly. "If the waist is bigger than the hips, it tells me that the risk carried with that weight is much higher for that person for premature death," Lopez-Jimenez said.

None of these measurements alone is perfect, and medical associations like the American Society for Nutrition and others are looking for even better measurements than what currently exists. You will likely still see doctors and schools taking BMI measurements, at least as an initial screening, to try and give you a better understanding of how your fat may be impacting your health.