Table 4 Evaluation of the accuracy of DXA for body composition assessment in ED patients and controls, using the 4C model as the reference method

From: Body composition in young female eating-disorder patients with severe weight loss and controls: evidence from the four-component model and evaluation of DXA

 

Bias

Correlation

 

Mean

Limits of agreement

P-values

r

P-values

 

(kg)

(%)

(kg)

(%)

   

ED patients

 FM

0.4

4.6

±3.1

±36.4

0.4

−0.40

0.2

 FFM

−0.3

−0.9

±3.2

±9.4

0.5

0.19

0.5

 FMI

0.2

4.6

±1.2

±36.4

0.4

−0.36

0.2

 FFMI

−0.1

−0.9

±1.2

±9.4

0.5

0.25

0.4

Controls

 FM

1.1

7.4

±2.6

±23.2

<0.001

0.37

<0.001

 FFM

−1.2

−3.2

±2.5

±6.2

<0.001

−0.48

<0.001

 FMI

0.4

7.4

±1.0

±23.2

<0.001

0.23

0.011

 FFMI

−0.5

−3.2

±0.9

±6.2

<0.001

−0.41

<0.001

  1. Abbreviations: 4C, 4-component model; DXA, dual X-ray absorptiometry; FM, fat mass; FMI, FM index; FFM, fat-free mass; FFMI, FFM index.
  2. Bias calculated as DXA value−4C value. Limits of agreement calculated as twice the s.d. of the bias. Correlation between magnitude of bias and the mean value by both techniques.
Back to article page