Validation of multi-frequency bioelectrical impedance analysis versus dual-energy X-ray absorptiometry to measure body fat percentage in overweight/obese Colombian adults.

Abstract

To verify the validity of multi-frequency bioelectrical impedance analysis (mBCA) for predicting body fat percentage (BF%) in overweight/obese adults using dual-energy X-ray absorptiometry (DXA) as the reference method. Forty-eight adults participated (54% women, mean age = 41.0 ± 7.3 years old). The Pearson's correlation coefficient was used to evaluate the correlation between BIA and BF% assessed by DXA. The concordance between BF% measured by both methods was obtained with Lin's concordance correlation coefficient and Bland-Altman difference plots. Measures of BF% were estimated as 39.0 (SD = 6.1) and 38.3 (SD = 6.5) using DXA and mBCA, respectively. The Pearson's correlation coefficient reflected a strong correlation (r =.921, P = .001). The paired t-test showed a significant mean difference between these methods for obese men BF% of -0.6 [(SD 1.95; 95% CI = -4.0 to 3.0), P =.037]. Overall, the bias of the mBCA was -0.6 [(SD 2.2; 95% CI = -5.0 to 3.7), P =.041], which indicated that the mBCA method significantly underestimated BF% in comparison to the reference method. Finally, in both genders, Lin's concordance correlation coefficient showed a strong agreement. More specifically the DXA value was ρc = 0.943 (95% CI = 0.775 to 0.950) and the mBCA value was ρc = 0.948 (95% CI = 0.778 to 0.978). Our analysis showed a strong agreement between the two methods as reflected in the range of BF%. These results show that mBCA and DXA are comparable methods for measuring body composition with higher body fat percentages. However, due to broad limits of agreement, we can only recommend mBCA for groups of populations.