Table 5 Association between percent body fat (measured and predicted) and high low-density lipoprotein (LDL) obtained from logistic regression and bias assessment, NHANES 2001–2006.

From: Development and validation of a prediction equation for body fat percentage from measured BMI: a supervised machine learning approach

Studies

NHANES 2001–2002

(n = 1383)

NHANES 2003–2006

(n = 2340)

Odds ratios

95% CI

Biasa

Odds ratios

95% CI

Biasa

Measured BF%

1.042

1.022–1.062

 

1.02

1.010–1.031

 

Gallagher18

1.034

1.015–1.054

−0.008

1.021

1.006–1.037

0.001

Gomez-Ambrosi19

1.025

1.010–1.041

−0.017

1.017

1.004–1.031

−0.003

Liu21

1.053

1.025–1.081

0.01

1.03

1.009–1.052

0.01

Model 1

1.037

1.014–1.060

−0.005

1.026

1.008–1.044

0.005

Model 2

1.037

1.014–1.060

−0.005

1.026

1.007–1.044

0.005

Model 3

1.037

1.014–1.060

−0.005

1.026

1.007–1.044

0.005

  1. The metrics for the top performing model(s) are in Bold.
  2. High LDL is defined as LDL >  = 130 mg/dL.
  3. High LDL was model as a function of BF%, age, race, gender, education, income.
  4. aBias = ln(OR from predicted BF%)—ln(OR from measured BF%).
Back to article page