European Journal of Applied Physiology

, Volume 105, Issue 1, pp 119–130

Estimating body fat in NCAA Division I female athletes: a five-compartment model validation of laboratory methods

Authors

  • Jordan R. Moon
    • Metabolic and Body Composition Laboratories, Department of Health and Exercise ScienceUniversity of Oklahoma
  • Joan M. Eckerson
    • Department of Exercise Science and Athletic TrainingCreighton University
  • Sarah E. Tobkin
    • Metabolic and Body Composition Laboratories, Department of Health and Exercise ScienceUniversity of Oklahoma
  • Abbie E. Smith
    • Metabolic and Body Composition Laboratories, Department of Health and Exercise ScienceUniversity of Oklahoma
  • Christopher M. Lockwood
    • Metabolic and Body Composition Laboratories, Department of Health and Exercise ScienceUniversity of Oklahoma
  • Ashley A. Walter
    • Biophysics Laboratory, Department of Health and Exercise ScienceUniversity of Oklahoma
  • Joel T. Cramer
    • Biophysics Laboratory, Department of Health and Exercise ScienceUniversity of Oklahoma
  • Travis W. Beck
    • Biophysics Laboratory, Department of Health and Exercise ScienceUniversity of Oklahoma
    • Metabolic and Body Composition Laboratories, Department of Health and Exercise ScienceUniversity of Oklahoma
Original Article

DOI: 10.1007/s00421-008-0881-9

Cite this article as:
Moon, J.R., Eckerson, J.M., Tobkin, S.E. et al. Eur J Appl Physiol (2009) 105: 119. doi:10.1007/s00421-008-0881-9

Abstract

The purpose of the present study was to determine the validity of various laboratory methods for estimating percent body fat (%fat) in NCAA Division I college female athletes (n = 29; 20 ± 1 year). Body composition was assessed via hydrostatic weighing (HW), air displacement plethysmography (ADP), and dual-energy X-ray absorptiometry (DXA), and estimates of %fat derived using 4-compartment (C), 3C, and 2C models were compared to a criterion 5C model that included bone mineral content, body volume (BV), total body water, and soft tissue mineral. The Wang-4C and the Siri-3C models produced nearly identical values compared to the 5C model (r > 0.99, total error (TE) < 0.40%fat). For the remaining laboratory methods, constant error values (CE) ranged from −0.04%fat (HW-Siri) to −3.71%fat (DXA); r values ranged from 0.89 (ADP-Siri, ADP-Brozek) to 0.93 (DXA); standard error of estimate values ranged from 1.78%fat (DXA) to 2.19%fat (ADP-Siri, ADP-Brozek); and TE values ranged from 2.22%fat (HW-Brozek) to 4.90%fat (DXA). The limits of agreement for DXA (−10.10 to 2.68%fat) were the largest with a significant trend of −0.43 (P < 0.05). With the exception of DXA, all of the equations resulted in acceptable TE values (<3.08%fat). However, the results for individual estimates of %fat using the Brozek equation indicated that the 2C models that derived BV from ADP and HW overestimated (5.38, 3.65%) and underestimated (5.19, 4.88%) %fat, respectively. The acceptable TE values for both HW and ADP suggest that these methods are valid for estimating %fat in college female athletes; however, the Wang-4C and Siri-3C models should be used to identify individual estimates of %fat in this population.

Keywords

Air displacement plethysmographyUnderwater weighingMulti-compartmentDual-energy X-ray absorptiometry

Copyright information

© Springer-Verlag 2008