Consistency of fat mass--fat-free mass relationship across ethnicity and sex groups

Stephanie T. Broyles, Pennington Biomedical Research Center, Louisiana State University System, 6400 Perkins Road, Baton Rouge, LA 70808-4124, USA. stephanie.broyles@pbrc.edu
Claude Bouchard
George A. Bray
Frank L. Greenway
William D. Johnson
Robert L. Newton
Eric Ravussin
Donna H. Ryan

Abstract

The model developed by Forbes (1987) of how body fat mass (FM) and fat-free mass (FFM) change during periods of weight loss or gain (Δ body weight (BW)) assumed that they change in relationship to a constant C = 10·4, where ΔFFM/ΔBW = 10·4/(10·4+FM). Forbes derived C based on aggregated, cross-sectional data from a small sample of women. The objective of the present study was to reanalyse the relationship described by Forbes and to explore whether this relationship is consistent across ethnicity and sex groups using cross-sectional data from a large sample of white and African-American men and women. Baseline data from white and African-American men and women aged 18-60 years, who participated in a clinical study at the Pennington Biomedical Research Center since 2001 and who underwent dual-energy X-ray absorptiometry scans, were available for analysis. To overcome differences in BMI distributions among the ethnicity-by-sex groups, a stratified random sample of participants was selected within each group such that numbers in each BMI category ( < 25, 25-29·9, 30-34·9, 35-39·9, 40+ kg/m2) were proportional to those within the group with the smallest sample size, yielding a sample of 1953 individuals. Linear regression models assessed the FM-FFM relationship across the four ethnicity-by-sex groups. The FM-FFM relationship varied little by ethnicity (P = 0·57) or by sex (P = 0·26). The constant describing the FM-FFM relationship was estimated to be 9·7 (95 % CI 9·0, 10·3). In conclusion, results from our large, biethnic sample of men and women found a FM-FFM relationship very close to that originally described by Forbes, absent of significant variability by ethnicity or sex.