Skip to main content
. 2018 Dec 31;9(1):96–105. doi: 10.2478/joeb-2018-0014

Table 2.

Statistics of BIS vs SFBIA data in non-CKD BMI stratified subgroup

Underweight subjects (n=40)
Bland-Altman
95% limits of agreement
Data analysed BIS SFBIA ICC r Biasa (%)b 95%CI Biasc (%)d Lower Upper
R (ohm) 848±74 596±103 -0.02 -0.12 252+134 (35.5+18.9) 209 to 294 (29 to 42) -10 513
Xc (ohm) 86±9.3 63±9.5 -0.03 -0.12 23+14 (31.1+18.1) 19 to 28 (25 to 37) -4.6 51
PA () 5.8±0.5 6.1±0.6 0.12 0.09 -0.3+0.8 (-4.8+12.6) -0.5 to -0.05 (-8.9 to -0.7) -1.8 1.2
TBW (L) 27±3.9 30±4.1 0.38** 0.47** -3.4+4.1 (-11.7+14.3) -4.7 to -2 (-16 to -7.1) -11 4.7
ECW (L) 11±1.4 14±2.2 0.20* 0.47** -2.9+2 (-22.8+14.8) -3.6 to -2.3 (-28 to -18) -6.8 0.9
ICW (L) 16±2.6 16±1.9 0.45** 0.44** -0.4+2.5 (-3.1+14.8) -1.2 to 0.4 (-7.9 to 1.7) -5.2 4.4
FFM (kg) 34±6.3 43±4.8 0.26* 0.59** -9.1+5.1 (-24.4+13.9) -11 to -7.5 (-29 to -20) -19 1
FM (kg) 14±4 5±4.4 0.07* 0.27* 8.8+5.1 (81+202.4) 7.2 to 10 (15 to 147) -1.1 19
Normal weight subjects (n=120)
Bland-Altman
95% limits of agreement
Data analysed BIS SFBIA ICC r Biasa (%)b 95%CI Biasc (%)d Lower Upper
R (ohm) 718±99 621±97 -0.02 -0.05 97+141 (14.6+20.6) 71 to 122 (11 to 18) -180 374
Xc (ohm) 82±9.6 63±8.9 0.04 0.13 19+12 (25.8+16.9) 16 to 21 (23 to 29) -5.3 43
PA () 6.6±0.8 5.9±0.5 0.10 0.17 0.7+0.9 (11.1+14.1) 0.5 to 0.9 (8.6 to 14) -1 2.5
TBW (L) 35±7.3 32±4.4 0.51** 0.62** 2.1+5.7 (4.9+16.7) 1.0 to 3.1 (1.8 to 7.9) -9.1 13
ECW (L) 14±2.6 16±2.4 0.58** 0.66** -1.3+2.1 (-9.4+14.3) -1.7 to -0.9 (-12 to -6.8) -5.4 2.7
ICW (L) 20±4.8 17±2 0.27* 0.56** 3.4+4 (16.2+19.4) 2.7 to 4.1 (13 to 20) -4.4 11
FFM (kg) 43±11 47±5.6 0.52** 0.71*** -3.6+8.5 (-10.9+20) -5.2 to -2.1 (-15 to -7.2) -20 13
FM (kg) 21±7.1 18±6.5 0.21* 0.25* 3.6+8.3 (18.8+50.4) 2.1 to 5.1 (9.8 to 28) -13 20
Overweight subjects (n=118)
Bland-Altman
95% limits of agreement
Data analysed BIS SFBIA ICC r Biasa (%)b 95%CI Biasc (%)d Lower Upper
R (ohm) 650±78 547±102 0.01 0.02 102±128 (18±22) 79 to 126 (14 to 22) -148 352
Xc (ohm) 77±9 62±8.3 0.01 -0.02 15±12 (21±18) 12 to 17 (18 to 24) -9.6 39
PA () 6.8±0.8 6.6±0.9 0.01 0.01 0.2±1.2 (3±18) -0.03 to 0.4 (-0.4 to 6.3) -2.2 2.6
TBW (L) 38±7.7 37±6 0.52** 0.52** 0.8±6.8 (1.4±18) -0.4 to 2.1 (-1.8 to 4.7) -13 14
ECW (L) 16±2.8 17±2.8 0.65** 0.70*** -0.9±2.2 (-6.1±13) -1.4 to -0.6 (-8.4 to -3.7) -5.2 3.3
ICW (L) 22±5.1 20±3.5 0.30** 0.34** 1.8±5.1 (7.4±23) 0.9 to 2.8 (3.1 to 12) -8.1 12
FFM (kg) 45±13 55±8.9 0.31** 0.46** -9.5±12 (-21±24) -12 to -7.4 (-26 to -17) -32 13
FM (kg) 34±9.2 24±5.7 -0.07* -0.13* 9.6±11 (31±41) 7.5 to 12 (24 to 39) -13 32
Obese subjects (n=108)
Bland-Altman
95% limits of agreement
Data analysed BIS SFBIA ICC r Biasa (%)b 95% CI Biasc (%)d Lower Upper
R (ohm) 596±77 482±53 0.07 0.19 114±85 (20.8+14.8) 98 to 130 (17.9 to 23.6) -52 280
Xc (ohm) 71±9.3 61±6.9 -0.01 -0.01 9.8±12 (14.5+17.4) 8 to 12 (11.2 to 17.9) -13 33
PA () 6.8±0.8 7.2±0.6 0.01 -0.05 -0.4±1 (-5.7+15.3) -0.6 to -0.2 (-8.7 to -2.8) -2.5 1.7
TBW (L) 42±8.2 42±5.4 0.70*** 0.77*** -0.6±5.4 (-2.4+12.5) -1.6 to 0.4 (-4.8 to 0.01) -11 9.9
ECW (L) 18±3.1 19±2.9 0.79*** 0.82*** -0.8±1.9 (-4.5+9.7) -1.2 to -0.4 (-6.4 to -2.7) -4.4 2.8
ICW (L) 24±5.3 23±2.4 0.53** 0.69** 0.2±4 (-0.9+16.6) -0.5 to 1 (-4.1 to 2.3) -7.6 8.1
FFM (kg) 46±12.9 64±8.1 0.26* 0.71*** -18±9.2 (-36.4+20.5) -20 to -16 (-40.3 to -32.5) -36 -0.5
FM (kg) 52±11.5 33±6.7 0.16* 0.61** 19±9.1 (43.3+18.7) 17.2 to 20.7 (39.7 to 46.9) 1 37

BIS, multifrequency bioelectrical impedance spectroscopy; BMI, body mass index; ECW, extracellular water; FFM, fat free mass; FM, fat mass; ICC, Intraclass Correlation Coefficient; ICW, intracellular water; PA, phase angle; r, Pearson correlation coefficient; R, resistance; SFBIA, single-frequency bioelectrical impedance; TBW, total body water; Xc, reactance. Data presented as mean ± SD or minimum to maximum value. ICC, *p<0.05, **p<0.01 BIS vs SFBIA. r, *p<0.05, **p<0.01 BIS vs SFBIA. aMean error between BIS and SFBIA: BIS – SFBIA. bMean percentage error between BIS and SFBIA: Bias/[(BIS + SFBIA)/2] X 100. c95%CI of difference between BIS and SFBIA: BIS – SFBIA. d95%CI of difference between BIS and SFBIA (%): Bias/[(BIS + SFBIA)/2] X 100. 95%CI that include zero are unbiased.