Body mass index may not tell whole health story

Body mass index, the ratio of weight to height, is often used to classify people as normal weight, overweight or obese, but it has several limitations when used as a personal health indicator.

Kate Harding, a freelance writer who blogs about fat-acceptance issues, put together a body mass index (BMI) photo project to visually show the difference between what we think of as overweight and obese people and who they actually are. The BMI Project aims to illustrate the measurement's limitations to address the stigma of its labels.

"A lot of people hear overweight and think 'fat person,' where you look at these pictures of people in the overweight range and they look perfectly normal," Harding said. "Even the obese people look more like what most of us would think of as chubby."

The standard classifications for BMI - calculated as weight in kilograms divided by height in meters squared - place a person with a BMI less than 18.5 as underweight, from 18.5 to 24.9 as normal, from 25 to 29.9 as overweight, more than 30 as obese, and over 40 as morbidly obese. The high cut-off point for the normal-weight range was lowered from 27 to 24.9 in 1998.

In the 1980s, doctors began to use BMI as an individual health metric to determine if a person was a normal weight or overweight, but this use of BMI has several limitations.

The biggest limitation of BMI is that it doesn't distinguish fat mass from muscle mass and doesn't assess fat distribution, said Dr. James de Lemos, a researcher at UT Southwestern Medical Center who has compared BMI with other health measures.

Harding pointed to the classic example of the athlete with a lot of muscle and a low body fat percentage who has a BMI in the overweight range. The index also doesn't account for frame size, bone density, age or ethnicity, she said.

As well, BMI and cardiovascular health have a J-shaped relationship, de Lemos said, where the best BMI isn't the lowest. Those with BMIs under 23 actually show a slightly higher risk of heart disease, which is probably associated with other illnesses that reduce body and muscle mass and are related to poor cardiac outcomes.

"People who are normal or even slightly overweight might be even better from a cardiovascular standpoint," he said. The risk then increases again as BMI increases.

BMI does have some value outside of clinical settings, de Lemos said, because it allows people to track changes in body mass over time. However, there are other measurements that more accurately indicate health risk, he said.

In a study comparing BMI, waist circumference and waist-to-hip ratio as risk factors for plaque build-up in arteries of the heart and aorta, de Lemos and his research team found that the waist-to-hip ratio measurement remained associated with cardiovascular risk even when obesity-associated conditions like diabetes, hypertension and cardiovascular risk factors were accounted for.