BMI scale not accurate measure for obesity

HOUSTON -- (April 6, 2009) -- Certain ethnic groups may not be getting accurate estimates of disease risk when they try to determine obesity using the traditional body mass index scale, say researchers at Baylor College of Medicine and the University of Houston in a study in the current edition of British Journal of Nutrition.

DEXA scans of two individuals with the same BMI

DXA images are of two adults with the same BMI. While their weight and height may be similar, resulting in the same BMI, the person on the right has twice as much body fat as the person on the left. DXA images allow doctors to analyze bone density, lean mass and fat mass. Click the image to see a larger version.

"Our research shows that the number used to indicate weight category does not reflect the same amount of body fat for some races compared to others," said Dr. Molly Bray, associate professor of pediatrics - nutrition at the USDA/ARS Children's Nutrition Research Center at BCM and Texas Children's Hospital. "The results are consistent with other studies that say BMI is inexact and should be tailored to help target those at risk."

BMI is a formula that estimates a person's body fat using only his/her weight and height. The result is then used to determine weight categories. For example, 18.5 and below is considered underweight and between 18.6 and 24.9 is considered healthy. A person is overweight when their number falls between 25 and 29. Thirty and above is considered obese.

"This scale was created years ago and is based on Caucasian men and women," said Bray, who is also senior author of the study, "It doesn't take into account differences in body composition between genders, race/ethnicity groups, and across the lifespan."

In the current study, rather than using other potentially biased methods employed in the past as "gold standards" to examine body composition, researchers used dual-energy x-ray absorptiometry, which is a low dose x-ray known as DXA, to determine percent fat. DXA can be used to estimate bone density, lean mass and fat mass.

"It's important to break down the BMI numbers," Bray said. "Just because you weigh a certain amount doesn't necessarily mean you are overweight. Take, for example, an athlete who is very muscular and still weighs up to 300 pounds."

When the two results were compared, researchers found that the DXA estimate of percent fat of African American women was 1.76 percent lower for the same BMI compared to non-Hispanic white women. Since BMI is assumed to represent body fatness, an African American woman would not be considered overweight or obese until she reached a higher number than what is indicated by the current BMI standards. The opposite is the case for Hispanic, Asian and Asian-Indian woman. Their percent fat is higher by 1.65 percent, 2.65 percent and 5.98 percent, respectively. So they would be considered overweight or obese at amounts lower than what the BMI standards indicates. The results for men were similar.

"Right now non-Hispanic white women are not considered obese until they have a BMI of 30 or above. Based on our data in young adults, for Hispanic women the number would be around 28," said Bray. "For African American women the number to cross is around 32."

Bone mineral content, hydration state, and the density of lean mass found in different ethnic groups are some factors that account for the differences.

"These results demonstrate that body composition measures obtained from DXA can be used to more accurately define risk factors and provide better recommendations for the treatment and care for patients," said Bray.

The study participants were recruited from the TIGER (Training Intervention and Genetics of Exercise Response) study. It is designed to investigate how variation in DNA sequence may influence levels of body fatness and fitness.

Other researchers who took part in the study include Drs. Andrew S. Jackson, lead author of the study, and Brian K. McFarlin, both of the Department of Health and Human Performance at the University of Houston; Drs. Kenneth J. Ellis and Mary H. Sailors, both of the USDA/ARS Children's Nutrition Research Center at BCM and TCH; Dr. John Foreyt, professor of medicine, psychiatry and behavioral sciences at BCM.

Funding came from a grant from the National Institute of Health.

The paper can be found at http://journals.cambridge.org/action/displayJournal?jid=BJN.