Skip to main content

Join Our Newsletter


Use of a three-component model for evaluation of density and hydration of fat-free mass and validation of body composition predicted from the same whole-body bio-electrical impedance measurement, in lean and obese women.

Fuller NJ  Sawyer MB  Elia M

Abstracts of Communications 1993 (66A)

The aims of the present study were twofold: (1) to evaluate fat-free mass (FFM) density and hydration (Dffm and Hffm respectively) using a three-component reference model (Fuller et al. 1992) in discrete groups of twelve lean (body mass index 20.9 (SD 2.1) kg/m2; body fat 24.4 (SD 3.1) % and fifteen obese (body mass index 42.8 (SD 8.8) kg/m2; body fat 48.0 (SD 6.8) %) women; and (2) to establish the extent of agreement between body composition estimates obtained with this model and eight different predictions of the same whole-body bio-electrical impedance (BI) measurement (see Fuller, 1993).

The extent of agreement between estimates of FFM (kg) and body fat (%) by the three-component model and BI predictions using bias (three-component model minus BI prediction) and 95% limits of agreement (95% LA) is shown in the Table. Dffm was found to be 1.097 (SD 0.006) kg/l and 1.104 (SD 0.006) kg/l, and Hffm was 73.0 (SD 1.6) % and 71.2 (SD 1.6) % for lean and obese women respectively.

No material difference was found between either value for Dffm and its traditionally-applied value (1.1 kg/l), despite significant differences between the lean and obese women for both Dffm (p<0.01) and Hffm (p<0.01).

Mean values for Dffm and Hffm obtained on a relatively small number of subjects indicate no radical need to alter those traditionally applied to FFM (although the accuracy of the three-component model is limited by the assumption of constant ratio of protein to mineral; Fuller et al. 1992). In general, agreement between methods was better for lean than for obese women (with few exceptions for each, in the obese women the bias was more positive for FFM and more negative for fat, and the 95% LA were considerably larger). The extent of variability from different predictions incorporating BI highlights a potential source of major error and confusion, especially in the obese. Thus, a coherent approach is recommended to ensure that appropriate predictions (derived against valid reference methods) can be applied with confidence to relevant populations

Download Paper - Use of a three-component model for evaluation of density and hydration of fat-free mass and validation of body composition predicted from the same whole-body bio-electrical impedance measurement, in lean and obese women.