ArticlePDF Available

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

Authors:
R T Withers
R T Withers
R T Withers
  • This person is not on ResearchGate, or hasn't claimed this research yet.
N O Whittingham
N O Whittingham
N O Whittingham
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Kevin Norton at UNSW Sydney
Kevin Norton
J La Forgia
J La Forgia
J La Forgia
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Show all 6 authorsHide
Download full-text PDF
Read full-text
Download citation

Abstract

Ninety-one percent (n=182) of the female members of South Australian representative squads in 14 sports volunteered to act as subjects. Twenty-seven percent of them had represented Australia. The underwater weighing method together with the measurement of residual volume (RV) by helium dilution were used to determine body density (BD); the percent body fat (% BF) was then computed according to Siri. A stepwise multiple regression analysis yielded a correlation coefficient (R) of 0.863 between the criterion (BD) and the best weighted sum of predictors (anthropometric variables): BD (g·cm−3)=1.14075−0.04959 (log10 ∑ triceps, subscapular, supraspinale and calf skinfolds in mm)+0.00044 (age in decimal years)−0.000612 (waist girth in cm)+0.000284 (height in cm)−0.000505 (gluteal girth in cm)+0.000331 (breast girth in cm). Only those predictors which resulted in a statistically significant increase inR (p⩽0.05) were included. The standard error of estimate of 0.00597 g · cm−3 was equivalent to 2.7% BF at the mean. This equation was shown to be largely population specific. There was a range of 7.6–35.8% of BF and the overall mean of 18.5% was significantly lower (p
Content uploaded by Alan Crockett
Author content
All content in this area was uploaded by Alan Crockett on Nov 07, 2015
Content may be subject to copyright.
Eur J Appl Physiol
(1987) 56:615
European Journal of
Applied
Physiology
and Occupational Physiology
9 Springer-Verlag 1987
Erratum
Relative body fat and anthropometric prediction of body density
of female athletes*
R. T. Withers, N. O. Whittingham, K. I. Norton, J. La Forgia, M. W. Ellis, and A. Crockett
The Exercise Physiology Laboratory, School of Education, The Flinders University of South Australia, Bedford
Park,
South Australia
5942
Eur J Appl Physiol
(1987) 56:169--180
During the final printing process the following two corrections were unfortunately not carried out:
1. On page 169, line 27 of the Summary, the first "of" should have been deleted.
2. On page 177, line 8 of the right-hand column, "for an equation" should have been deleted.
... Somatotype scores were calculated as previously described [27]. Estimation of fat-free mass was derived from the Withers regression model [28]. ...
In-Group Nutrition Interventions with Elite Weight-Sensitive Athletes Are Associated with an Increase in Nutrition Knowledge and Intake of Carbohydrate-Rich Foods, but Not a 400+ Increase in Daily Caloric Intake
Preprint
Full-text available
  • Apr 2025
Background: Nutrition knowledge is positively associated with energy intake in athletes, and therefore, improving nutrition knowledge may offer a cost-effective approach to prevent negative performance and health outcomes associated with low energy availability (LEA) described in the Relative Energy Deficiency in Sports (REDs) syndrome model. The aim was to assess the short-term influence of a 60-min group-based nutrition education intervention on sports nutrition knowledge and dietary intake in elite athletes. Method: Twenty-five elite Canadian athletes were enrolled into a 3-week prospective single blinded randomized pragmatic trial. Participants were randomly assigned to 1 of 2 group-based nutrition education interventions. Education content was similar between both groups with 1 group assigned an additional 5-10 mins to review athlete testimonials on the negative impacts of REDs. Participants were assessed, before and about 10 days after the intervention, using anthropometry, Platform to Evaluate Athlete Knowledge of Sports Nutrition, Athlete Food Choice, Eating Disorder Examination 6.0, LEA/Triad, Athlete Diet Index questionnaires, and consecutive 5-day food record. Results: The education intervention that was supplemented with athlete testimonials performed similarly to the intervention without (p>0.05). Overall, nutrition knowledge (score from 69.8, to 72.8) and the intake of carbohydrate-rich foods (12.4 to 14.3) increased while disordered eating/eating symptoms decreased (0.788 to 0.642) after the education intervention (all p< 0.05). No changes in daily energy and carbohydrate intake were observed after the education intervention (p>0.05). Conclusion: The nutrition education sessions were associated with an in-crease in nutrition knowledge and other factors associated with food choice leading to increased consumption of carbohydrate-rich foods and reducing disordered eating/eating disorder symptoms. More research is required on methods to influence athlete eating behaviours to optimize energy intake to prevent associated negative health and performance outcomes.
View
... BMI was calculated as body mass in kilograms divided by the square of stature in meters (kg/m 2 ) [33]. Body density (BD) was estimated by using specific equations for M [34] and F [35] athletes. BD was transformed into body fat (BF) percentage using equations specific for each sex published by the ACSM [31]. ...
Relationship of Body Composition and Somatotype with Physical Activity Level and Nutrition Knowledge in Elite and Non-Elite Orienteering Athletes
Article
Full-text available
  • Feb 2025
Objectives: The primary aim of this single cross-sectional study was to identify the physical characteristics (anthropometric, somatotype, body composition) of orienteering athletes (OAs) and to compare them with nutrition knowledge (NK) and physical activity level (PAL). Methods: Data were collected from 58 subjects of seven countries, including Angola (n = 1), Brazil (n = 5), Poland (n = 1), Portugal (n = 26), South Africa (n = 1), Spain (n = 22) and Sweden (n = 2). The subjects included 10 elite (E) female (F) OAs [age: 25.5 ± 6.4 years, body mass: 59.5 ± 7.7 kg, stature: 168.1 ± 6.5 cm, body mass index (BMI): 21.0 ± 1.9 kg/m²], 13 E male (M) OAs (age: 24.3 ± 5.0 years, body mass: 65.0 ± 5.5 kg, stature: 175.1 ± 6.0 cm, BMI: 21.3 ± 2.2 kg/m²), 18 non-elite (NE) FOAs (age: 41.7 ± 10.3 years, body mass: 60.6 ± 8.5 kg, stature: 161.3 ± 11.7 cm, BMI: 23.4 ± 3.7 kg/m²), and 17 NEMOAs (age: 37.2 ± 14.6 years, body mass: 71.5 ± 14.2 kg, stature: 174.0 ± 8.8 cm, BMI: 23.6 ± 4.1 kg/m²). The participants were selected to ensure a diverse and representative sample of international-level orienteering athletes. Measurements were taken at two IOF world ranking events, the “Portugal “O” Meeting (POM)” and the “35° Trofeo Internacional Murcia Costa Cálida”, where only top-ranked orienteers compete. The selected participants from these seven countries were among the registered athletes in these international competitions. The OAs were measured according to the guidelines of the International Society for the Advancement of Kinanthropometry (ISAK). NK was evaluated using the updated Abridged Nutrition for Sport Knowledge Questionnaire (A—NSKQ). PAL was assessed using the short version of the self-reported International Physical Activity Questionnaire—Short Form (IPAQ—SF). Results: The percentage of body fat (p < 0.01) in MOAs was significantly lower than in FOAs. Endomorphy (p = 0.037) and mesomorphy (p = 0.025) in EOAs were significantly lower than in NEOAs, but ectomorphy (p = 0.038) was significantly higher. EMOAs are ectomorphic mesomorphs, while NEMOAs are balanced mesomorphs, EFOAs are central, and NEFOAs are endomorphic mesomorphs. Significant differences (p < 0.01) were also observed in sports nutrition knowledge (SNK) among EOAs and NEOAs, with the former group achieving a higher percentage of correct responses. In the case of total nutritional knowledge (TNK), EOAs of both sexes scored significantly higher (p = 0.043) than their NEOA counterparts. A significant negative correlation was also observed between percentage of body fat (%BF) and metabolic equivalent (MET) in minutes per week (min/week) (r = −0.39, p = 0.038), bone mass (BM) and MET-min/week (r = −0.40, p = 0.033), and endomorphy and SNK (r = −0.38, p = 0.045) in FOAs. Among MOAs, the most significant findings included a negative correlation between age and METmin/week (r = −0.49, p = 0.010), kilocalorie (kcal) per week (r = −0.46, p = 0.016), and SNK (r = −0.40, p = 0.029). Conclusions: The key findings indicate that EOAs have lower BF percentages and higher NK scores compared to NEOAs. These results on the physical characteristics of OAs and the score of PAL and classification of NK can be useful to coaches and sports scientists to improve orienteer’s performance.
View
... Body Mass Index (BMI) was calculated as weight (kg) to height squared (m 2 ) quotient [19]. Body density (BD) was estimated by using specific equations for male [20] and female [21] athletes. BD was transformed into body fat (BF) percentage using equations specific for each sex published by the ACSM [18]. ...
Relationship of Body Composition and Somatotype with Physical Activity Level and Nutrition Knowledge in Elite and Non-Elite Orienteering Athletes
Preprint
Full-text available
  • Jan 2025
This study investigates the physical characteristics, including anthropometry, somatotype, and body composition, of elite and non-elite orienteering athletes. Additionally, it explores the relationship between these physical characteristics and the athletes' nutrition knowledge and physical activity levels. Our findings indicate significant differences in body fat percentage, mesomorphy, ectomorphy, and nutrition knowledge scores between elite and non-elite athletes. These insights can be valuable for coaches and sports scientists aiming to enhance the performance of orienteering athletes. The key contributions of our study include: - Identification of physique characteristics specific to orienteering athletes. - Comparison of these characteristics between elite and non-elite athletes. - Analysis of the relationship between body composition, somatotype, physical activity levels, and nutrition knowledge.
View
... Body composition was determined using the equations described in the consensus document of the Spanish Group of Kinanthropometry of the Spanish Federation of Sports Medicine [6], following the four-component model (muscle mass (MM), fat mass (FM), bone mass (BM), and residual mass (RM)). The following equations were used: the Withers [24], Slaugther [25], Carter [26], Faulkner [27], Durnin [28], and Jackson and Pollock [29] equations to calculate FM expressed in percentage (%) and kilograms (kg); the Poortmans [30] and Lee [31] equations to calculate MM expressed in kg and %; and Rocha's equation [32] to calculate BM expressed in kg and %. The sum of 3, 6, and 8 skinfold measurements were also calculated. ...
Anthropometric and Body Composition Changes during Pre-Season of Spanish Professional Female Soccer Players According to Playing Position
Article
Full-text available
  • Aug 2024
Background: In professional soccer, body composition analysis is crucial to assess preparation and optimize performance. Different playing positions have different physical demands, which can lead to variations in body composition. However, there are few studies on women’s soccer that consider the playing position. This study aims to fill that gap by examining position-specific differences in anthropometric and body composition characteristics among Spanish professional female soccer players at the beginning and end of pre-season. Furthermore, it investigates the possible changes during the pre-season period between positions and correlates the data obtained from anthropometric equations with bioimpedance (BIA) measurements. Methods: Thirty-four female soccer players: 8 midfielders, 12 defenders, 11 forwards, and 3 goalkeepers (age: 23.06 ± 4.29 years, height: 164.15 ± 5.84 cm, weight: 58.39 ± 6.62 kg, and ∑6 skinfolds: 74.57 ± 18.48 mm) completed the study that lasted 4 weeks (pre-season) where they were measured anthropometrically and by bioimpedance twice. Results: Goalkeepers showed greater wingspan (176.60 ± 7.06 p < 0.05) compared to other positions. Regarding differences during pre-season, midfielders had the greatest decrease in ∑6 skinfolds compared to other positions (∆ −12.10 ± 5.69 p < 0.05). There was a correlation of % fat between Faulkner’s equation and BIA (Pearson’s r = 0.817). Conclusions: It seems that there are no significant differences in terms of positions and body composition, except for the wingspan and ankle diameter. During pre-season, midfielders are the ones who improve their body composition the greatest. The anthropometric equation for body fat that shows the highest correlation with BIA is Faulkner’s equation, followed by Durnin’s equation.
View
... Body Mass Index (BMI) was calculated as weight (kg) to height squared (m 2 ) quotient 14 . Body density (BD) was estimated by using specific equations for male 15 and female 16 athletes. BD was transformed into body fat (BF) percentage using equations specific for each sex 13 . ...
Anthropometric characteristics and physical activity level of Orienteers: a pilot study
Article
Full-text available
  • Nov 2023
The present study aimed to describe a study protocol for Orienteers’ anthropometric characteristics and physical activity level. This is an individualized, observation and cross-sectional pilot study, that has encompassed three Orienteers (2 Portuguese men and 1 Spanish woman) with mean age 22.6 ± 3.78 years and 8.0 ± 2.64 consecutive years of Orientation practice. In total, 26 anthropometric variables were assessed: four basic measurements, nine skinfolds, nine girths and four breadths; Body Mass Index was calculated, as well as body composition and somatotype (according to the Heath-Carter method). The International Physical Activity Questionnaire – Short Form was used to obtain the score expressed as metabolic equivalent and the energy expenditure in kilocalories, both during the reference week. Mean somatotype recorded for Orienteering athletes could be defined as balanced mesomorph. Scores recorded for different physical activity/ inactivity expressed domains have shown high-energy expenditure. Sports Science Professionals can use the herein proposed protocol to improve Orienteers’ training.
View
... En esta línea, las primeras publicaciones de prestigio que aportaron información sobre aspectos tales como la altura de las jugadoras, su peso corporal o el porcentaje de grasa se encuentran a finales de esa década. (Colquhoun, y Chad, 1986;Withers et al., 1987) Según Kirkendall (2007), el número de publicaciones científicas en referencia al fútbol femenino fue muy bajo durante las primeras décadas de desarrollo y crecimiento. Asimismo, la gran parte de ellas tenían una temática orientada a la fisiología y las lesiones. ...
Investigación en fútbol femenino: antecedentes, progreso y futuros horizontes
Article
Full-text available
  • May 2023
El fútbol femenino ha experimentado un crecimiento exponencial en la última década. Patrocinadores y clubes finalmente han tomado la decisión de apostar por el desarrollo de una modalidad históricamente menospreciada. La última Copa Mundial Femenina de la FIFA Francia 2019 supuso un récord de audiencia televisiva: 1,12 billones de personas se conectaron en alguno de los 52 partidos disputados en este campeonato. En cambio, en el ámbito de la investigación, sólo 1 de cada 4 artículos publicados sobre fútbol, está basado en el fútbol femenino. Este aspecto supone una problemática a la hora de poner en prácticas los conocimientos científicos del fútbol femenino. Por ello, el objetivo de este estudio fue analizar los artículos publicados sobre fútbol femenino así cómo establecer una guía de actuación para los próximos años. Si bien el progreso ha sido bueno en la última década, se encuentra en un punto de inflexión que puede permitir desarrollar de forma exponencial el conocimiento científico sobre el fútbol femenino en los próximos años, mejorando el rendimiento colectivo e individual en un futuro cercano.
View
Current Resting Metabolic Rate Prediction Equations Lack Sensitivity and Specificity to Indicate Relative Energy Deficiency in Sport: A Large Cohort Study in Elite Athletes
Article
  • Apr 2025
  • INT J SPORT NUTR EXE
Objectives: Measured resting metabolic rate (RMR) was compared to predicted RMR equations (RMRratio) to see whether a low RMRratio relates to the Relative Energy Deficiency in Sport (REDs) Clinical Assessment Tool 2 (CAT2) severity/risk score. Methods: Female (n = 127) and male (n = 53) athletes (performance Tiers 3–5) were assigned green/yellow/orange/red light according to CAT2. RMR and submaximal exercise energy expenditure (via cycle ergometer) were assessed fasted on the same morning via indirect calorimetry. Low RMR was defined as RMRratio < 0.90, with 11 RMR prediction equations tested for sensitivity, specificity, and predictive validity against the CAT2. Results: RMRratio (Cunningham) was only lower in red versus green light (0.90 ± 0.07 vs. 0.99 ± 0.10; p = .023; but RMRratio was only low in 44% of red light athletes). The prevalence of low RMRratio ranged from 1% (Owen equation) to 68% (van Hooren equation) despite the overall prevalence of REDs being 46%. As a diagnostic (no REDs [green] vs. REDs [yellow + orange + red]), Cunningham equation reported sensitivity (true positives) of 0.77 at RMRratio of 1.00 and specificity (true negatives) of 1.00 at RMRratio of 0.70. Exercise energy expenditure was significantly lower in green versus orange (0.131 ± 0.013 vs. 0.142 ± 0.008 kcal·kg fat-free mass−1·min−1; p < .001) but was greater in red (0.127 ± 0.011) versus orange. Conclusion: Interpretation of RMRratio is radically impacted by choice of prediction equation. Although there may be some utility in cross-sectionally detecting extreme REDs cases (red light) via RMRratio, more research with a focus on sport/phenotype-specific prediction equations and varying risk thresholds is required to strengthen the validity and reliability of RMR as a part of REDs diagnostics.
View
View
View
Critical power, W′ and W′ reconstitution in women and men
Article
Full-text available
  • Jun 2023
  • EUR J APPL PHYSIOL
PurposeThe aim of this study was to compare critical power (CP) and work capacity W′, and W′ reconstitution (W′REC) following repeated maximal exercise between women and men.Methods Twelve women (V˙\dot{V}O2PEAK: 2.53 ± 0.37 L·min−1) and 12 men (V˙\dot{V}O2PEAK: 4.26 ± 0.30 L·min−1) performed a minimum of 3 constant workload tests, to determine CP and W′, and 1 maximal exercise repetition test with three work bouts (WB) to failure, to quantify W′REC during 2 recovery periods, i.e., W′REC1 and W′REC2. An independent samples t test was used to compare CP and W′ values between women and men, and a repeated-measures ANOVA was used to compare W′REC as fraction of W′ expended during the first WB, absolute W′REC, and normalized to lean body mass (LBM).ResultsCP normalized to LBM was not different between women and men, respectively, 3.7 ± 0.5 vs. 4.1 ± 0.4 W·kgLBM−1, while W′ normalized to LBM was lower in women 256 ± 29 vs. 305 ± 45 J·kgLBM−1. Fractional W′REC1 was higher in women than in men, respectively, 74.0 ± 12.0% vs. 56.8 ± 9.5%. Women reconstituted less W′ than men in absolute terms (8.7 ± 1.2 vs. 10.9 ± 2.0 kJ) during W′REC1, while normalized to LBM no difference was observed between women and men (174 ± 23 vs. 167 ± 31 J·kgLBM−1). W′REC2 was lower than W′REC1 both in women and men.Conclusion Sex differences in W′REC (absolute women < men; fractional women > men) are eliminated when LBM is accounted for. Prediction models of W′REC might benefit from including LBM as a biological variable in the equation. This study confirms the occurrence of a slowing of W′REC during repeated maximal exercise.
View
Generalized equations for predicting body density
Article
Full-text available
  • Dec 1978
1. Skinfold thickness, body circumferences and body density were measured in samples of 308 and ninety-five adult men ranging in age from 18 to 61 years. 2. Using the sample of 308 men, multiple regression equations were calculated to estimate body density using either the quadratic or log form of the sum of skinfolds, in combination with age, waist and forearm circumference. 3. The multiple correlations for the equations exceeded 0.90 with standard errors of approximately ±0.0073 g/ml. 4. The regression equations were cross validated on the second sample of ninety-five men. The correlations between predicted and laboratory-determined body density exceeded 0.90 with standard errors of approximately 0.0077 g/ml. 5. The regression equations were shown to be valid for adult men varying in age and fatness.
View
View
View
View
Validity of estimating body fat of females: Effect of age and fitness
Article
  • Jan 1978
The influence of age and fitness factors on the validity of prediction equations for estimating percent body fat (%BF) of females was investigated for 60 females, ages 12 to 78 years. Regression equations specific to age and fitness subgroups were derived from a series of skinfold, girth, and diameter measurements of these subjects. The combination of anthropometric variables in these equations providing the highest multiple correlation coefficient varied according to both the aerobic power and age of the subjects. The validity coefficients and standard errors of estimate from previously published equations also varied from group to group, demonstrating again that equations to estimate %BF from anthropometric measurements are specific to the population from which they were derived. As cross-validation of the equation for young adult females in this study demonstrated, prediction errors are of such magnitude that the use of such equations for research or clinical application is highly questionable.
View
View
Anthropometric estimation of body density, fat, and lean body weight in women gymnasts
Article
  • Feb 1978
  • Med Sci Sports
Measurements from 44 highly skilled women gymnasts were used to evaluate previously derived equations for estimating body composition and to develop new equations by step-wise linear regression. True composition values were computed from body density (DB). Measures were taken of skeletal diameters (D), circumferences (C) and skin-folds (SF). Equations evaluated were those by Sloan et al. (23), F. Katch and Michael (15), Wilmore and Behnke (28), F. Katch and McArdle (18) and Weltman and V. Katch (26). With exception of the Weltman-Katch equation, correlations between true and estimated LBW's were high (.91 to .95) but underestimated true LBW 1.13 to 3.88 kg. Regression analysis suggested adjustment of ordinate values would give these equations acceptable accuracy for use with women gymnasts. New equations were derived from variables that had been selected by using factor analysis to minimize shared variance. Equations derived were DB - 1.024620 + .002024 Neck C--.001435 Supra-iliac SF--.001039 Juxta-um-bilicus SF (R.806, +/- .0061); Fat = .3798--.5809 Neck C + .3253 Abdominal C + .3007 Supra-iliac SF + .2047 Juxta-umbilicus SF, (R .891, +/- 1.29) and LBW = .3973 Height + .8357 Neck C + .4317 Shoulder C--87.203 (R .923, +/- 1.75).
View
Influence of muscular development, obesity, and age on the fat free mass of adults
Article
  • Sep 1976
  • J Appl Physiol
Body fat and the fat-free mass (FFM) were estimated in 36 men and 43 women deliberately chosen to represent a variety of physical types; these were 1) young sedentary, 2) "muscular," 3) younger obese, 4) older obese, and 5) older nonobese individuals of both sexes. The body fat and the FFM were estimated from measurements of body density (by total immersion in water, measurement being made of the residual volume of air present in the lungs at immersion) and from measurements of total body potassium (using a whole-body monitor to assess the natural 40K isotope present in the body). The muscular men and women and the younger obese men and women had a considerably greater FFM and thus had greater quantities of potassium than the corresponding sedentary groups. There were significantly different estimates of the FFM calculated from density and from total body K in three groups, the sedentary young men, the muscular, and the younger obese women. The density and the potassium content of the FFM appear to decline with obesity and aging. Muscular development is associated with a decrease in the density but an increase in the potassium content of the FFM.
View
Prediction of Body Density in Young and Middle-Aged Women
Article
  • May 1975
  • J Appl Physiol
The purpose of this investigation was to predict body density of young and middle-aged women and to determine if the use of a greater variety of variables, particularly those for fat in the bust and hip regions, increases the predictability of body density. Body density determined by the hydrostatic technique (dependent variable) was obtained from 83 volunteer young women and 60 middle-aged women ranging from 18 to 22 and 33 to 50 yr of age, respectively. Independent variables included 8 skinfold, 13 girth, and 7 diameter measures; age; height; weight; and bra and cup sizes. Mean body density for young women was 1.043 g/ml (SD plus or minus 0.014) and percent fat, 24.8 (SD plus or minus 6.4); 1.031 g/ml (SD plus or minus 0.015) and 29.8% (SD plus or minus 6.7) for middle aged subjects. Percent fat was calculated by the formula of Siri. Factor analysis was used to examine the dimensions measured by the independent variables as a function of age. A multiple regression model was used to develop predictions of body density from the independent variables. The best combination of four variables for predicting body density was skinfold thigh, skinfold suprailiac, cup size, skinfold suprailiac, girth waist, and skinfold thigh (R = 0.89) for middle-aged women. The data showed that the highest predictons were found by using combinations of skinfold, girth, and diameter variables; cup size also supported the need for different regression equations for different age groups.
View
The relative body fat and anthropometric prediction of body density of South Australian females aged 17?35 years
Article
  • Feb 1987
  • Eur J Appl Physiol Occup Physiol
One hundred and thirty-five females were tested in order to: produce some normative percentage body fat (% BF) data on an Australian sample which represented a cross-section of physical activity patterns, cross-validate existing multiple regression equations which predict body density (BD) from anthropometric measurements, and if necessary develop population specific equations. Measurements were taken of 10 girths, 3 widths and 7 skinfolds. Body density was measured by underwater weighing with the residual volume (RV) being determined by helium dilution. The Siri equation was then used to convert BD to % BF. The % BF scores had an overall mean of 23.4 (range 10.8–49.2). The very active group (n=45) had a significantly lower (p<0.05) relative body fat (X=20.6% BF) than either the active (n=45; 23.5% BF) or sedentary groups (n=45; 26.2% BF). Previously published equations were found to have limited applicability to Australian subjects. A stepwise multiple regression was therefore used to develop the following equation (R = 0.893): BD(g·cm−3) = 1.16957-0.06447 (log10 ∑ triceps, subscapular, supraspinale, front thigh, abdominal and calf skinfolds in mm)-0.00081 (gluteal girth in cm)+ 0.0017 (forearm girth in cm) + 0.00606 (biepicondylar humerus breadth in cm). Only those predictors which resulted in a statistically significant increase inr (p⩽0.05) were included. The standard error of estimate of 0.00568 g · cm−3 was equivalent to 2.6% BF at the mean.
View