COMPARISON OF DIFFERENT TECHNIQUES TO MEASURE BODY-COMPOSITION IN MODERATELY ACTIVE ADOLESCENTS

Objective-To evaluate the differences in the estimate of body fat perc entage (%FM) and the amount (kg) of fat free mass (FFM) by different m ethods in 26 moderately active adolescents very similar in age, body f atness, and training status. Methods-Mean (SD) age was 16.7 (0.9) year s, height was 177.0 (5.1) cm, and weight 68.0 (5.2) kg. %FM was assess ed using dual-energy x ray absorptiometry (DXA) and two skinfold predi ction equations: that of Slaughter et al (%FM Sla) and that of Deurenb erg et al (%FM Deu). In the same way, FFM was measured using DXA and d ifferent impedance equations: those of Suprasongsin et al (FFM Sup), S chaefer et al (FFM Sch), Houtkooper et al (FFM Hou), and Deurenberg et al (FFM Deu). To determine the interchangeability of the different me thods of measuring %FM and FFM, one way analysis of variance, standard error (SE), and coefficient of variation (CV%) ((SD/mean) x 100) were used. Results-On average, no significant statistical differences were observed between the values determined for %FM: DXA value, 11.7 (5.4) %; %FM Sla, 10.9 (4.0)%; %FM Deu, 11.5 (2.3)%. On the other hand, SE a nd CV% between each pair of the three methods used showed very large v ariability. With regard to the measurement or prediction of FFM, the m ean value measured by DXA was significantly higher than that predicted by the equation of Sch (+7.2 kg, p<0.001), Deu (+3.2 kg, p<0.001), an d Hou (+2.6 kg, p<0.001), whereas it was lower than that predicted by the equation of Sup (-1.6 kg, p<0.05). The Hou and Deu values were the only two that, on average, did not differ in a statistically signific ant way, although they showed the highest CV%. Conclusions-In our samp le of moderately active adolescents the estimated values for %FM and F FM appear to be highly dependent on method.