BMI and an Anthropometry-Based Estimate of Fat Mass Percentage Are Both Valid Discriminators of Cardiometabolic Risk: A Comparison with DXA and Bioimpedance

Table 1

Anthropometry- and bioimpedance-analysis- (BIA-) based estimates of fat mass percentage (FM%) and their respective bias versus DXA measurements.

Predictor	Equation for estimating FM%	Men			Women			Combined
		𝑛	Mean bias^a	SD	𝑛	Mean bias^a	SD	𝑛	Mean bias^a
(1) Larsson et al. [14]	Women: FM% = (−24.18 + 1.181 * weight/height)/weight; Men: FM% = (−30.84 + 1.120 * weight/height)/weight	205	−1.3	4.7	388	0.1	4.1	593	−0.6
(2) Gallagher et al. [15]	FM% = 64.5 – 848 * (1/BMI) + 0.079 * age −16.4 * sex^b − 0.05 * sex^b* age + 39.0 * sex^b* (1/BMI)	205	−2.4	4.6	388	−1.0	4.1	593	−1.7
(3) Deurenberg et al. [12]	FM% = 1.2*BMI + 0.23*age −10.8*sex^b  − 5.4	205	1.7	4.9	388	2.3	4.8	593	2.0
(4) Tanita BC 418 MA	bioimpedance-based proprietary algorithm	82	−4.8	3.9	58	−3.6	3.2	140	−4.2
(5) InBody (720)	bioimpedance-based proprietary algorithm	181	−4.6	3.4	273	−4.7	3.0	454	−4.6
(6) Mills [13]	Women: FM% = (−2.28 + 1.268 (weight/height); + 0.058 * age)/weight; Men: FM% = (−7.99 + 1.286 (weight/height); + 0.018 * age)/weight	205	−11.7	5.3	388	−1.6	4.2	593	−6.7
(7) Deurenberg et al. [16]	FM% = −11.4 * sex^b + 0.2 * age + 1.294 * BMI − 8	205	5.0	6.4	388	16.5	8.9	593	10.8
(8) FM%-equation	Arithmetic mean of equations (1)–(3)	205	−0.7	4.6	388	0.5	4.2	593	−0.1

afat mass percentage.
bmale: 1, female: 0.