Sex and age specific prediction formulas for estimating body composition from bioelectrical impedance: a cross-validation study

Int J Obes. 1991 Jan;15(1):17-25.

Abstract

In 827 male and female subjects, with a large variation in body composition and an age range of 7-83 years, body composition was measured by densitometry, anthropometry and bioelectrical impedance. The relationship between densitometrically determined fat free mass (FFM) with body impedance (R), body weight (W) and body height (H) was analysed, taking age and sex into account. The intercept of the regression equation FFM = a x H2/R + b was found to be age, and (at older ages) sex dependent, increasing from age 7 to age 15, and slowly decreasing after age 16. Therefore the population was subdivided into two age categories, the one 15 years and younger, and the other 16 years and older. Each age category was randomly divided into two groups, A and B. In each age category the developed prediction formula for group A was cross-validated in group B, and vice versa. No statistically and biologically meaningful differences between predicted and measured FFM were observed in either group. Therefore the data of group A and B in each age category were combined. The best fitted prediction formula at ages less than or equal to 15 was: FFM = 0.406 x 10(4) x H2/R + 0.360 W + 5.58 H + 0.56 Sex - 6.48: n = 166, R2 = 0.97, SEE = 1.68 kg (cv% = 4.9 percent); and at ages greater than or equal to 16: FFM = 0.340 x 10(4) x H2/R + 15.34 H + 0.273 W - 0.127 age + 4.56 sex - 12.44: n = 661, R2 = 0.93, SEE = 2.63 kg (cv% = 5.0 percent).(ABSTRACT TRUNCATED AT 250 WORDS)

MeSH terms

  • Adolescent
  • Adult
  • Age Factors
  • Aged
  • Aged, 80 and over
  • Body Composition*
  • Child
  • Cross-Sectional Studies
  • Electric Conductivity
  • Female
  • Humans
  • Male
  • Mathematics
  • Middle Aged
  • Regression Analysis
  • Reproducibility of Results
  • Sex Factors