Relative body fat and anthropometric prediction of body density of male athletes

Abstract

Two hundred and seven male members of South Australian representative squads in 18 sports (mean +/- s = 24.2 +/- 4.7 years) were tested in order to provide descriptive data on relative body fat (% BF), develop a population specific equation and cross-validate existing equations. Measurements were taken of 10 circumferences, 2 diameters and 8 skinfolds; body density (BD) was measured by underwater weighing with the residual volume (RV) being determined by He dilution. The overall mean BD was 1.0761 g X cm-3 (s = 0.0085 g X cm-3; range = 1.0465-1.0968 g X cm-3) which corresponded to 10.0% BF according to Siri (s = 3.7%; range = 1.3-23.0%). The games players (n = 129) registered an overall mean of 10.3% BF (s = 3.7%; range = 2.2-23.0%). There were significant differences (p less than 0.05) for % BF between the lacrosse players (mean = 12.3%) an both the Australian Rules footballers (mean = 8.0%) and track and field athletes (mean = 8.7%). A stepwise multiple regression on 185 subjects yielded the following equation, which had an R of 0.787: BD = 1.078865-0.000419 (sigma abdominal, medial calf, front thigh and juxta-nipple skinfolds in mm) +0.000948 (neck circumference in cm) -0.000266 (age in decimal years) -0.000564 (ankle circumference in cm). Only those predictors which resulted in a significantly increased correlation (p less than or equal to 0.05) were included. The standard error of estimate of 0.00537 g X cm-3 was equivalent to 2.3% BF at the mean. This equation was satisfactorily cross-validated against the BD of a separate sample (n = 22) from the same population. However, cross-validation of 11 previously published equations indicated that they have limited applicability to State representative sportsmen.

Use of anthropometric variables to predict relative body fat determined by a four-compartment body composition model

“…Despite the fact that a heterogeneous sample (10.0-36.6 % BF) was deliberately chosen, the SEEs (1.8-3.0 % BF) are lower and the correlations (0.92-0.97) are higher than those for most prediction equations in the literature (Norton, 1996). These previous predictions of body composition using anthropometric measurements have relied primarily on the twocompartment hydrodensitometric model as the criterion (Durnin & Womersley, 1974;Jackson & Pollock, 1978Withers et al, 1987). The present investigation is therefore unique because the criterion of % BF via the four-compartment model is more valid than hydrodensitometrically estimated % BF which erroneously assumes that the FFM density is 1.1000 g/cm 3 for all subjects.…”

“…On the other hand, the expedient field technique of recording skinfold thicknesses provides a reasonably accurate measurement of subcutaneous adipose tissue and furthermore has been shown to be the best anthropometric predictor of % BF (Jackson & Pollock, 1978;Jackson et al, 1980;Withers et al, 1987). Nevertheless, despite considerable face validity, their use in body composition research has been plagued with biologically significant prediction errors (SEEX2.3 % BF; Norton, 1996).…”

Assessment of oxidative stress biomarkers – neuroprostanes and dihomo-isoprostanes – in the urine of elite triathletes after two weeks of moderate-altitude training

“…The anthropometric measurements were made according to the International Society for the Advancement of Kinanthropometry (ISAK) and were performed by the same, internationally certified anthropometrist (level 2 ISAK) in order to decrease technical errors. The body composition was determined by GREC Kinanthropometry consensus [20], using a model consisting of: total fat by Withers´s formula [21]; lean weight by the procedure described by Leet et al [22]; and residual mass by the difference in the weight (Table 1).…”

Subject-specific computer simulation model for determining elbow loading in one-handed tennis backhand groundstrokes

“…Values for the percentage composition of fat and fat-free mass for the upperarm and forearm segments and whole body masses of cadavers were taken from the literature (Clarys, Martin and Drinkwater, 1984;Clarys and Marfell-Jones, 1986). To relate this information to the elite subject, an estimate of body fat as a percentage of whole body mass was calculated (Withers, Craig, Bourdon and Norton, 1987) based on skinfold measurements at seven sites on the body (Balady, Berra, Golding, Gordon, Mahler and Myers, 2000). The stiffness of the non-linear massless springdampers of the wobbling masses were chosen to match measured displacements of markers on the upper-arm and forearm with simulated wobbling mass displacements, across the one-handed backhand performances.…”