Effect of total body water estimates via bioimpedance on bod pod-based three-compartment body fat models

Previous research has compared 2- and 3-compartment (2C and 3C, respectively) models against criterion 4-compartment (4C) models while utilizing the same body density (D_b) method for all measures. This design induces an inherent bias and obscures the added benefit of a 3C model over the simpler 2-compartment (2C) models.

The purpose of this study was to determine the effect of total body water estimates via single-frequency (SF-BIA) and multi-frequency (MF-BIA) bioimpedance analysis on body fat estimates derived from air displacement plethysmography (ADP)-derived 3C models.

A sample of 95 females and 82 males (n = 177) participated in this study. Underwater weighing, dual energy X-ray absorptiometry, and bioimpedance spectroscopy were used to calculate percent fat (%Fat) via a criterion 4C model (4C_CRITERION). %Fat was predicted via 3C_MFBIA (ADP and MF-BIA), 3C_SFBIA (ADP and SF-BIA), and a stand-alone 2-compartment (2C) model, based upon ADP, when using Siri and Brozek body density conversion formulas (2C_SIRI and 2C_BROZEK. respectively).

The standard error of estimate (SEE) was lowest for 3C_SFBIA when evaluated in females and males (2.72% and 2.31%, respectively) and highest for 2C_SIRI (3.98% and 3.84%, respectively). Similarly, the total error (TE) for females and males was lowest for 3C_SFBIA (3.21% and 2.67%, respectively) and highest for 2C_SIRI (4.58% and 4.48%, respectively) and 2C_BROZEK (4.65% and 4.33%, respectively).

Results suggest that SF-BIA and MF-BIA can improve the estimation of %Fat, beyond simpler 2C models, when integrated with ADP in a more advanced 3C model. Furthermore, the present study revealed that 3C_SFBIA was the best overall prediction model based upon TE values. The current study results support the integration of ADP and bioimpedance technology as part of a 3C model for the improvement of %Fat estimates over simpler 2C models.