ARS | Publication request: Race and Ethnic Effect of Estimating Dxa Percent Fat from Bmi: the Tiger Study

Submitted to: Medical Science Sports Exercise
Publication Type: Abstract Only
Publication Acceptance Date: February 24, 2005
Publication Date: May 15, 2005
Citation: Bray, M.S., Ellis, K.J., Bush, J.A., Turpin, J., Callie, M., Miller, F., Jackson, A. 2005. Race and ethnic effect of estimating DXA percent fat from BMI: The Tiger Study [abstract]. Medical Science Sports Exercise. 37(5 Supl):S303.

Technical Abstract: PURPOSE Body mass index (BMI) has become the accepted public health standard of determining overweight (BMI = 25 kg/m2) and obese (BMI = 30 kg/m2). This study examined the effect of race and sex on estimating percent fat (%fat) using BMI. METHODS The subjects were 85 women and 39 men who ranged in age from 17 to 35 y. The ethnic makeup of the sample was: non-Hispanic white, 33.3%; African-American (AA), 30.0%; Hispanic, 18.7%; and other, 15.4%. Percent body fat was determined with dual energy X-ray absorptiometry (DXA). The mean %fat of the men and women was 24.1 plus or minus 7.4 and 34.8 plus or minus 7.1, respectively. The mean BMI of the men (27.5 kg/m2) and women were nearly identical (27.4 kg/m2) with over 50% being overweight and 25% obese. General linear models (GLM) was used to examine the effect of sex and race on the BMI and DXA %fat relationship with race coded as non-AA versus AA. RESULTS Seventy-five% of DXA %fat variance was accounted for by variation in BMI with sex and race contributing significantly to the model. There were no significant interactions between race and either sex or BMI. For the same BMI, DXA %fat of females was 5.9% higher than males and the DXA %fat of AA individuals was 3.5% lower than non-AA of the same sex and BMI. CONCLUSIONS These results demonstrate that the BMI relationship with DXA %fat is not independent of race and gender. Supported by NIH Grant DK61679.