Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: March 1, 2007
Publication Date: April 1, 2007
Citation: Harris, G.K., Stote, K.S., Clevidence, B.A., Paul, D.R., Kramer, M.H., Baer, D.J. 2007. BodPod Approximates Corrected DEXA Values More Closely Than BIA in Overweight and Obese Adults [abstract]. Federation of American Societies for Experimental Biology, April 28-May 2, 2007, Washington, DC. 21:679.2. Technical Abstract: Dual-absorbance x-ray absorptiometry (DEXA) is a rapid method for body fat estimation that correlates well with more time-consuming, technically challenging “gold standard” methods, and is relatively free of bias due to subject anthropometry. Since radiation exposure limits the frequency of DEXA use, it is important to identify methods that correlate well with DEXA and can be used for frequent, repeated measures. We compared body fat estimates by leg-to-leg bioelectrical impedance (BIA) and air-displacement plethysmography (BodPod) to DEXA scans taken twice over a 6-month period in otherwise healthy overweight/obese subjects (36 women; 29 men). Independent sample t-tests comparing the slope and intercept of DEXA vs. BIA (DBIA) and DEXA vs. BodPod (DBod) regression lines for all study subjects found that the slope and intercept of the DBIA, but not the DBod, regression line differed significantly (P < 0.05 for slope and intercept comparisons) from the (DEXA vs. DEXA) line of identity. The same results were observed when data for male and female subjects were examined separately, and when the women’s group was subdivided into African American (n = 12) and non-African American women (n = 24). The men’s group was not subdivided by ethnicity because 27 of the 29 men were Caucasian. F-tests comparing the standard errors of the slope and intercept for each group found a significant (P < 0.01 for slope and intercept comparisons) difference in scatter between the DBIA and DBod data only when all subjects were grouped together. Our data indicate that BodPod, but not BIA, is a robust alternative to DEXA for body fat estimation, based on its similarity to DEXA values, and to an apparent lack of bias due to subject gender and ethnicity. This research was supported by the US Department of Agriculture.