Comparison of hand-to-hand bioimpedance and anthropometry equations versus dual-energy X-ray absorptiometry for the assessment of body fat percentage in 17-18-year-old conscripts.

In the present study, hand-to-hand bioelectrical impedance analysis (BIA), skinfold (SKF) thickness and height-weight (body mass index, BMI)-based equations and dual-energy X-ray absorptiometry (DEXA), as a criterion method, were compared with each other in the assessment of body fat percentage (BF%) in 17-18-year-old Estonian conscripts (n = 32). The Omron BF body fat monitor estimated that BF% was lower than that of the criterion method DEXA. The difference between DEXA and Omron BF 300 (III) was higher (1.1 +/- 3.0%; P = 0.04) and that between DEXA and Omron BF 306 lower (0.2 +/- 3.0%; P>0.05). Omron BF 300 (I) and (II) (series 8) had intermediate difference (0.9 +/- 3.0 and 0.9 +/- 3.0; P>0.05) when compared with DEXA. Three anthropometric equations estimated a higher BF% than cthat of DEXA. The Durnin & Womersley SKF equation BF% (1.0 +/- 2.4; P = 0.03) was higher than that of the DEXA. Deurenberg et al. and Gallagher et al. BMI-based equations overestimation yielded 0.9 +/- 3.7 and 0.6 +/- 3.8 BF% (P>0.05). From the anthropometric equations, only the Deurenberg et al. SKF equation slightly underestimated 0.5 +/- 3.4 BF% (P>0.05). DEXA-assessed BF% had highest correlation with SKF equations (r = 0.93), less so with BIA (r = 0.88-0.89) and lowest with BMI equation-assessed BF% (r = 0.81-0.84). All values were significant at P<0.001. We can conclude that the Omron BF 306 body fat monitor and the anthropometric Deurenberg et al. SKF equation yielded results close to the DEXA BF%.