TABLE 4

Predictive value of body composition assessment models

Predictors	Men			Women
	Adjusted R²	β	p-value	Adjusted R²	β	p-value
Muscle strength
One-compartment model
BMI	0.29	2.50	0.00	0.30	1.58	0.00
Two-compartment model
FFMI	0.32	6.28	0.00	0.31	3.41	0.00
FMI			ns		0.97	0.03
Three-compartment model
ASMI	0.47	17.01	0.00	0.51	11.74	0.00
BMC		0.01	0.00		0.02	0.00
A/G%FM			ns		16.68	0.00
Peak workload
One-compartment model
BMI	0.47	0.96	0.00	0.40	0.56	0.02
Two-compartment model
FFMI	0.48	2.90	0.00	0.41	1.92	0.01
FMI			ns			ns
Three-compartment model
ASMI	0.56	8.06	0.00	0.47	10.68	0.00
BMC		0.01	0.00			ns
A/G%FM			ns			ns
CET
One-compartment model
BMI	0.02		ns	0.02		ns
Two-compartment model
FFMI	0.02	17.53	0.01	0.02		ns
FMI						ns
Three-compartment model
ASMI	0.07	42.31	0.03	0.04		ns
BMC		0.08	0.03		0.10	0.02
A/G%FM			ns			ns
6MWD
One-compartment model
BMI	0.14		ns	0.26	−5.47	0.00
Two-compartment model
FFMI	0.16	7.19	0.04	0.27		ns
FMI		−5.43	0.01		−7.62	0.00
Three-compartment model
ASMI	0.16		ns	0.20		ns
BMC			ns			ns
A/G%FM		64.52	0.01		−60.45	0.04

Summary of backward multiple regression analysis, with muscle strength, peak workload, cycling endurance time (CET) and 6-min walk distance (6MWD) as dependent variables and body composition models (one-compartment: body mass index (BMI); two-compartment: fat-free mass index (FFMI) and fat mass index (FMI); or three-compartment: appendicular skeletal muscle index (ASMI), bone mineral content (BMC) and android/gynoid percentage fat mass (A/G%FM)) as predictors. All models are adjusted for age and forced expiratory volume in 1 s. ns: nonsignificant.