Yt. Wu et al., CROSS-VALIDATION OF BIOELECTRICAL IMPEDANCE ANALYSIS OF BODY-COMPOSITION IN CHILDREN AND ADOLESCENTS, Physical therapy, 73(5), 1993, pp. 320-328
Background and Purpose. The reliability and validity of measurements o
btained with two bioelectrical impedance analyzers (BIAs), an RJL Syst
ems model BIA-103 and a Berkeley Medical Research BMR-2000, were inves
tigated using the manufacturers' prediction equations for the assessme
nt of fat-free mass (FFM) (in kilograms) in children and adolescents S
ubjects. Forty-seven healthy children and adolescents (23 male, 24 fem
ale), ranging in age from 8 to 20 years (XBAR = 12. 1, SD = 2.3), part
icipated. Methods. In the context of a repeated-measures design, the d
ata were analyzed according to gender and maturation (Tanner staging).
Hydrostatic weighing (HYDRO) and Lohman's Sin age-adjusted body densi
ty prediction equation served as the criteria for validating the BIA-o
btained measurements Results. High intraclass correlation coefficients
(ICC greater-than-or-equal-to .987) demonstrated good test-retest (be
tween-week) measurement reliability for HYDRO and both BIA methods. Be
tween-method (HYDRO versus BIA) correlation coefficients were high for
both boys and girls (r greater-than-or-equal-to .97). The standard er
rors of estimate (SEEs) for FFM were slightly larger for boys than for
girls and were consistently smaller for the RJL system than for the B
MR system (RJL SEE = 1. 8 kg for boys, 1. 3 kg for girls, BMR SEE = 2.
4 kg for boys, 1.9 kg for girls). The coefficients of determination we
re high for both BIA methods (r2 greater-than-or-equal-to .929). Total
prediction errors (TEs) for FFM showed similar between-method trends
(RJL TE = 2.1 kg for boys, 1.5 kg for girls; BMR TE = 4.4 kg for boys,
1.9 kg for girls). Discussion and Conclusion. This study demonstrated
that the RJL BIA with the manufacturer's prediction equations can be
used to reliably and accurately assess FFM in 8- to 20-year-old childr
en and adolescents. The prediction of FFM by the BMR system was accept
able for girls, but significant overprediction of FFM for boys was not
ed.