Body Composition Profiles of Elite American Heptathletes

Transcription

1 162 / Houtkooper et al. International Journal of Sport Nutrition and Exercise Metabolism, 2001, 11, Human Kinetics Publishers, Inc. Body Composition Profiles of Elite American Heptathletes Linda B. Houtkooper, Veronica A. Mullins, Scott B. Going, C. Harmon Brown, and Timothy G. Lohman This study characterized body composition profiles of elite American heptathletes and cross-validated skinfold (SKF) and bioelectrical impedance analysis (BIA) field method equations for estimation of percent body fat (%Fat) using dual energy x-ray absorptiometry (DXA) as the criterion. Weight, height, fat mass (FM), fat-free mass (FFM), bone mineral density (BMD), and %Fat were measured in 19 heptathletes using standard measurement protocols for DXA, SKFs and BIA. The ages, heights, and weights were respectively 25.5 ± 3.5 years, ± 6.6 cm, 67.3 ± 7.1 kg. DXA estimates of mean ± SD values for body composition variables were 57.2 ± 6.1 kg FFM, 10.1 ± 2.6 kg FM, 114 ± 7% BMD for age/racial reference group, and 15 ± 3.0 %Fat. Ranges of bias values for %Fat (DXA minus SKF or BIA) were, respectively, 0.5 to 1.6% and 5.5 to 1.2%. Ranges for standard errors of estimate and total errors were, respectively, SKF %, % and BIA 3.0%, %. Regression analyses of the field methods on DXA were significant (p <.05) for all SKF equations but not BIA equations. This study demonstrates that elite American heptathletes are lean, have high levels of BMD, and that SKF equations provide more accurate estimates of %Fat relative to DXA than estimates from BIA equations. Key Words: female athletes, body fat, fat free mass, cross-validation, bone mineral density Heptathletes are female athletes who compete in seven events performed in competitions over a period of 2 days. The events include 100-m hurdles, high jump, shot put, 200-m dash, long jump, javelin, and 800-m run. The USA Track and Field Heptathlete Development Project (HDP) supports training enhancement of elite heptathletes who attain a minimum score of 5,600 points at major domestic or international competitions. Assessment of body composition is a component of the profiles in the development program for these heptathletes. It is recognized that athletic performance cannot be accurately predicted based solely on body weight and composition. However, they are two of the many factors that influence an athleteʼs performance (19, 43). Obtaining accurate and reliable estimates of body weight and composition in laboratory and L.B. Houtkooper, V.A. Mullins, and S.B. Going are with Department of Nutritional Sciences at the University of Arizona, Tucson, AZ Going and T.G. Lohman are with the Department of Physiology at the University of Arizona. C.H. Brown is with USA Track and Field, 345 Bowfin Street, Foster City, CA

2 Elite American Heptathletes / 163 field settings can help characterize successful athletes within this sport. Dual energy x-ray absorptiometry (DXA) is a relatively new laboratory method for measuring body composition that provides measures of fat mass (FM), lean soft tissue mass (LTM), total body bone mineral content (BMC), and bone mineral density (BMD). Fat free mass (FFM) from DXA is the sum of LTM and BMC. DXA is a safe, convenient, and noninvasive method that involves only a small radiation dose and provides precise measurements of BMC and LTM (24, 33, 34, 39, 41, 51). It may also be a more sensitive method for tracking changes in body composition than other body composition criterion methods such as hydrodensitometry (underwater weighing) (13, 20). Typically, athletic coaches do not have access to a laboratory for body composition assessment of athletes using DXA and must rely on field methods such as skinfolds (SKF) and whole body bioelectrical impedance analysis (BIA). It is important to cross-validate prediction equations used in field methods for estimation of body composition in order to determine the accuracy of these equations relative to criterion methods. The purposes of this study were to characterize body composition (FFM, FM, BMD, and %Fat) of elite U.S. heptathletes using DXA and to cross-validate SKF and BIA prediction equations for estimation of %Fat in this particular group of athletes using DXA as the criterion model. Subjects Methods Nineteen elite American female heptathletes, invited by USA Track and Field to take part in the HDP, participated in this study between the years of 1990 and All athletes provided written informed consent prior to participation in this study. Procedures Body composition measurements were taken by trained technicians at the University of Arizona Body Composition Research Laboratory during the annual HDP Summit Meeting, which was held during the phase of the training cycles when the athletes were not competing and were training at minimal levels. The clinical measurements included height (Ht), body weight (Wt), and body composition components using DXA, SKF, and BIA. All measurements were taken after a 12-hour interval of no strenuous exercise and an overnight fast. Each athlete was encouraged to drink enough fluids to stay hydrated during the fasting period. Data from the most recent year of the HDP with complete body composition assessment measurements from DXA, SKF, and BIA were used in this analysis. Dual Energy X-ray Absorptiometry (DXA) Measurements of whole body lean soft tissue mass, arm soft tissue mass, leg soft tissue mass, whole body FFM, FM, and BMD were made with the whole body DPX-L scanner from Lunar Radiation Corporation using standard procedures and software version 1.3y, extended research analysis mode (Madison, WI) (13). Each whole body scan was performed at a medium speed and took 15 to 20 min to complete depending on the height of the athlete.

3 164 / Houtkooper et al. Bioelectrical Impedance Analysis (BIA) Whole body resistance and reactance were measured with the athletes lying supine using standard procedures for the Valhalla Scientific bioelectrical impedance analyzer with four surface self-adhesive spot electrodes and a standard current of 800 A and 50 khz (49). Measurements were taken in triplicate, and the mean was used in four published regression equations to estimate FFM (Table 2). Total Wt and FFM measurements were then used to calculate FM and %Fat. Height and Weight Anthropometry Standing height (Ht) was measured in centimeters to the nearest millimeter with the head in the Frankfort Plane while holding a deep inspiration. Weight was measured to the nearest tenth of a kilogram using a calibrated balance beam or digital scale, with the subjects wearing minimal clothing. Body mass index (BMI) [Wt(kg)/Ht 2 (m 2 )] and Fat Free Body Mass Index (FFBMI) [FFM (kg)/h 2 (m 2 )] values were calculated for each athlete. Skinfolds (SKF) Measurements of SKF variables were taken using the procedures set forth in the Anthropometric Standardization Reference Manual and Lange skinfold calipers (27). Triceps, suprailliac, abdomen, and thigh skinfold sites were measured in triplicate, and the mean was used in four published prediction equations (Table 2) for young adult women to estimate body density (D b ) or %Fat. If D b was predicted, %Fat was estimated using the Siri equation (45). Statistical Analysis Regression analysis was used to determine which of the BIA or SKF prediction equations provided the best prediction of %Fat compared to the %Fat values estimated by the DXA criterion method. Student t tests were used to compare mean differences between the criterion and BIA and SKF %Fat values. The p value was set at <.05. All statistical analyses were performed using Statistical Package for Social Sciences (SPSS; v. 9.0) (48). Body Composition Profiles Results Values for age and body composition variables for individual athletes and the group (mean ± SD) are reported in Table 1. Age, Ht, and Wt, ranged from years, cm, and kg, respectively. The range for BMI was kg/m 2. The ranges for DXA measurements of body composition variables were kg for FFM and kg for FM. The ranges were kg/m 2 for FFBMI, g/cm 2 for BMD, and % for %Fat. The BMD values expressed as a percentage of age and racial group standard values ranged from %.

4 Elite American Heptathletes / 165 Table 1 Age, Height, Weight, Body Mass Index (BMI), and Body Composition Profiles From Dual Energy X-Ray Absorptiometry (DXA) of Elite American Female Heptathletes (N = 19) Age Ht Wt BMI FFM FM FFBMI BMD BMD 1 Fat (years) (cm) (kg) (kg/m 2 ) (kg) (kg) (kg/m 2 ) (gm/cm 2 ) (%) (%) M ±SD Percent of age and race BMD reference values. Cross-Validation of SKF and BIA Equations The comparisons of %Fat estimates among the SKF and BIA equations with DXA are summarized in Tables 2. The ranges of bias (mean differences between %Fat values measured by DXA and estimated values from BIA and SKF equations) were larger for BIA equations, ( 1.2 to 5.5%) than for SKF equations ( 0.5 to 1.6%). On average, the %Fat estimates among the four SKF equations were similar, ranging from %, and were within 1.5% of the average %Fat DXA criterion measurement. In comparison, only two of the BIA prediction equations estimates of %Fat were within 2% of the DXA measurement of average %Fat values. The average estimates of %Fat for the other two BIA equations were 4.4% and 5.5% higher than the average DXA measurement. The results of regression analyses predicting DXA %Fat measurements from the four SKF and four BIA equations are also summarized in Table 2. The coefficients of determination (r 2 ) are lower than typically seen for BIA and SKF estimates of

5 166 / Houtkooper et al. Table 2 Regression Analysis of Measured Percent Body Fat of U.S. Elite Female Heptathletes From Dual Energy X-Ray Absorptiometry (DXA) Criterion Model With Percent Body Fat Estimated From Skinfold Equations (SKF) a and Bioelectrical Impedance Analysis (BIA) b Equations %Fat %Fat bias SEE TE Method Mean ± SD r 2 (Criterion-Field) (%) (%) p c DXA: 15.0 ± 3.0 SKF A 13.4 ± ± SKF B 15.5 ± ± SKF C 13.5 ± ± SKF D 13.5 ± ± BIA A 17.1 ± ± BIA B 16.8 ± ± BIA C 20.5 ± ± BIA D 19.4 ± ± Note. %Fat = FFM converted to % fat using: Equation [(Wt FFM)/Wt] 100, n = 19, r 2 = coefficient of determination, SEE = standard error of estimate TE total error:. Units of measurements for the equations are: skinfolds (mm), age (years), Ht (cm), resistance and reactance (ohms), Wt (kg), FFM (kg). a Equation A: Db = [ (Triceps + Thigh + Suprailium)] + [ (Triceps + Thigh + Suprailium) 2 )] ( age) (Jackson & Pollock, 1985, 3 site). Equation B: Same as equation A, using Triceps, Abdomen, and Suprailium sites. Equation C: Db = (Triceps + Thigh + Suprailium + Abdomen) (Triceps + Thigh + Suprailium + Abdomen) (age) (Jackson & Pollock, 1985, 4 site). Equation D: Db = (Triceps + Thigh + Suprailium + Abdomen) (Triceps + Thigh + Suprailium + Abdomen) (age) (Jackson & Pollock, 1985, 4 site). Db converted to % fat using Siri, 1961: %fat = [(4.95/Db) 4.50] 100. b Equation A: FFM = [.73 (Ht 2 /resistance)] + (0.16 Wt) (Lohman, 1992, p. 53). Equation B: FFM = [.666 (Ht 2 /resistance)] + (.217 reactance) + (.164 Wt) 8.78; Adjusted FFM =.18(45.0 FFM) + FFM (Lohman equation for active women, to correct for influence of FFM size, 1992, p. 127). Equation C: Valhalla Impedance Analyzer (Model 1990B). Equation D: FFM=.73 (Ht 2 /resistance) + (.116 Wt) + (.096 reactance) 4.03 (Lukaski & Bolonchuk college women, 1987). c p values indicate the probability that the difference between mean %fat values determined by the field method and DXA occurred by chance.

6 Elite American Heptathletes / 167 %Fat because of the homogeneity of the group. Estimates of %Fat from SKF equations were significant (p <.005) predictors of %Fat relative to the criterion model, while estimates of %Fat from the BIA equations were not significant predictors of DXA %Fat. The range of SEEs for %Fat values determined from SKF ( %) and BIA (3.0%) are considered very good (SEE = 3.0%) (28). The range of TE values for %Fat determined from SKF ( %) was smaller than the range for BIA values ( %). Discussion The unique aspects of this study are the characterization of lean soft tissue, FFM, FFBMI, %Fat, and BMD profiles of elite American heptathletes using dual energy x-ray absorptiometry (DXA) and cross-validation of SKF and BIA equations. The results indicated that most of these athletes have low relative body fat (%Fat) and relatively high FFM for their Ht, and all the athletes have high levels of BMD compared to standard BMD values for women of their age and race. These findings contrast with low BMD values reported for some long-distance female runners, who reported reproductive cycle abnormalities and were oligomenorrheic or amenorrheic (9, 38). All of the heptathletes in this study self-reported that they were eumennorheic. Accurate assessment of muscle mass is difficult in living humans. Since muscle represents the single largest fraction of FFM, the FFM is often measured as a surrogate for muscle mass. Athletes involved in activities such as throwing, pushing, and weight lifting typically have a greater FFM than athletes engaged in sports that require translocation of the body mass such as distance runners and jumpers (14, 19). Endurance athletes who have a higher FFM tend to be involved in sports where the body mass is not supported by the athlete directly, as in rowing and cycling. In such sports, not only the body mass but also an external mass (skiff or bicycle) must be displaced. Comparisons of FFM among athletes are confounded by differences in average Ht. Taller individuals may have a higher absolute FFM while also having a lower mass for Ht than their shorter counterparts. The FFBMI can be used to compare profiles of FFM relative to Ht for athletes in different sports (46). The lowest FFBMI values among elite athletes, listed in Table 3, are found in dancers and distance runners (15 16 kg/m 2 ) because in running and dancing, a low %Fat and a low FFM index are advantageous. Shot put (20 kg/m 2 ), discus throwers (19 kg/m 2 ), pentathletes (19 kg/m 2 ), and field throwers (19 20 kg/m 2 ), who must generate higher forces than distance runners, have the highest FFMBMI values. The ideal ranges of %Fat and FFM index levels for elite athletes are currently unknown. Table 3 summarizes the average values for age, Wt, Ht, %Fat, FFM, and FFM index of high caliber female athletes in various sports. The body composition values for these athletes were estimated by hydrodensitometry or SKFs using various prediction equations. Recent studies using multi-component models as the criterion method have demonstrated that DXA is a valid measure of body composition in physically active groups of men and women (20, 24, 34, 39, 51). Historically, the two-component chemical model has been the criterion for body composition assessment of athletes using hydrodensitometry (underwater weighing) and the Siri (44) or Brozek (6) equation to estimate %Fat values. A two-component model divides the body into FM and FFM and is based on the assumptions that each component has a constant density and a constant composition of constituents for all individuals in a

7 168 / Houtkooper et al. given population (45). The density of adult human fat is relatively constant within and among individuals at g/cm 3 (26). However, studies have demonstrated that the density and composition of FFM are not constant and that multicomponent models, such as DXA, provide more accurate estimates than two-component models, such as hydrodensitometry, for estimation of %Fat values (24, 26, 34). Kohrt (24) reported that average %Fat (24.2 ± 6.6%) estimated using DXA in a sample of nonathletic females years of age was higher than average estimates from hydrodensitometry (21.4 ± 5.1%) using the Brozek (6) equation. Estimates of average %Fat values in this sample of female heptathletes from DXA (15.0 ± 3.0%) were also higher than the average estimate from hydrodensitometry using the Siri (44) equation (11.8 ± 3.3%). DXA %Fat estimates reported by Kohrt (1998) were made using Hologic QDR 1000/W, enhanced whole body software version 5.6, and estimates in this study with heptathletes were made using the Lunar DPX-L. In a sample of men and women (n = 24), Modlesky et al. (35) demonstrated that there was close agreement between average estimates of %Fat from DPX-L measurements (20.3 ± 11.0%) and a four-component model (26) that includes DPX-L measurements of bone mineral content (21.4 ± 10.1%). This indicates that DPX-L systems that use current software provide a good criterion method for estimation of %Fat values for young adults and that DXA estimates are slightly higher than estimates from hydrodensitometry. The range of %Fat estimates from DXA for athletes in this study was 7.7 to 20.6% and is somewhat lower than the range of values (14 23%) reported for elite endurance athletes, including runners, cyclists, and triathletes (Table 3). Female athletes participating in endurance sports have lower %Fat levels than young, healthy women in the general population, for whom %Fat ranges from approximately 22 28% (28). These endurance athletes often have %Fat levels near the minimum level of 8 10% associated with good health and performance (28). Of the high-caliber endurance athletes listed in Table 3, rowers have the lowest %Fat at 14.0%, followed by triathletes (14.1%), cyclists (15.4%), and cross-country skiers (15.7%). Except for dancers, high-caliber female athletes participating in other types of sports have %Fat levels closer to those of the general population. The throwers had the highest %Fat at 29.2%, followed by shot-putters (28.0%), discus throwers (24.9%), and cyclists (23.3%). Because heptathletes participate in both running and field events, they cannot be characterized solely as track athletes or field athletes. Although the heptathletes and endurance athletes had similar ranges of %Fat values, the heptathletes had higher FFBMI (range = kg/m 2 ) than the endurance athletes (15 18 kg/m 2 ). This is expected since heptathletes compete in power events such as shot put and javelin, where a larger muscle mass is advantageous (4, 17). When compared to other high-caliber female field athletes, including discus throwers, shot-putters, and throwers, who have %Fat levels ranging from % and FFBMI values of 19 to 20 kg/m 2, heptathletes had lower %Fat levels and similar FFBMI values. This is also expected, since heptathletes perform endurance activities during training and compete in speed/endurance events such as the 800-m run and therefore benefit from a relatively low %Fat (4). Although levels of %Fat and FFM for athletes influence successful performance within a sport, athletic performance cannot be accurately predicted solely on the basis of body composition (1, 3, 4, 10, 19, 43). Body size and structure are also significant determinants of competitive success. Successful performance in activities

8 Elite American Heptathletes / 169 Table 3 Age, Height, Weight, and Body Composition of Elite Female Athletes From Selected Sports Age Ht Wt FFM FFM/ Competitive Sport N (years) (m) (kg) %Fat (kg) Ht Technique Level Reference Cyclists UWW a National Team 7 Cyclists Anthropometry b Trained 23 Dancers Anthropometry c Professional 8 Discus athletes Anthropometry a,d Collegiate 31 Distance runners UWW a Collegiate 2 Distance runners UWW e World class 15 Distance runners UWW e World class 50 Distance swimmers UWW e U.S. national team 50 Heptathletes DXA g Elite Gymnasts UWW a Collegiate 42 Jumpers/hurdlers Anthropometry a,d Collegiate 31 Pentathletes UWW a U.S. national rank 25 Rowers Skinfolds a,d Elite 16 Shot putters Anthropometry a,d Collegiate 31 Skiers (Cross Country) Skinfolds f National team 18 Sprinters and middle distance runners UWW e World class 50 Sprinters ( m) Anthropometry a,d Collegiate 31 Throwers UWW e National rank 5 Throwers Anthropometry b National rank 12 Triathletes UWW a World class 36 Volleyball players UWW a Collegiate 40 a %Fat estimated using Brozek equation (1963); b %Fat estimated using Durnin and Womersley equation (1974); c %Fat estimated using Jackson & Pollack 3-site equation (1980); d Body density estimated using equation by Sloan (1962); e %Fat estimated using Siri equation (1956); f %Fat estimated from Sloan equation (1962); g DXA, Lunar 1.3 years.

9 170 / Houtkooper et al. such as throwing, pushing, and weightlifting, each of which requires the application of force against external objects, is positively related to the absolute amount of FFM and therefore to body size (4, 17). In these sports, more massive individuals have an advantage over their lighter counterparts. A large FFM and body size may adversely affect performance requiring translocation of the body mass, such as in running, jumping, or rotation of the body about an axis (4, 32, 37). Cross-Validation The cross-validation criteria for an acceptable field method equation are a small bias (mean difference between %Fat values estimated by the criterion method minus the field method), a standard error of estimate (SEE) less than 4.5%, a low total error (TE), which is close to the SEE, and a high coefficient of determination (r 2 ) (28). The SKF prediction equations for estimation of %Fat in these elite female athletes are the preferred field method equations because compared to the average %Fat estimates from the DXA, the criterion method, these equations resulted in the smallest mean differences (bias), lowest SEEs and TEs, and highest r 2 values compared to the BIA equations. However, two of the BIA equations, Equations A and B, had bias and SEEs values similar to the SKF equations, but values for TE were larger, and coefficients of determination were low. Others studies have demonstrated that with carefully applied SKF measurement methods, it is possible to estimate %Fat values in groups of athletes with an error of approximately ±3 %Fat and to estimate FFM with an error of ±2.5 kg (28, 30). With inappropriate prediction equations and poor measurement technique, prediction errors will be much larger. The SKF and BIA methods provide quick and noninvasive means of estimating body composition in elite female heptathletes. Compared to the BIA method, the SKF method requires less expensive equipment but more technical skill in order to obtain accurate and reliable measurements. Accurate and reliable estimates of %Fat from SKFs depends on the technicianʼs skill in measurement of the skinfolds, type of skinfold caliper, and the selection of an appropriate prediction equations used to estimate %Fat values (29). Training in standardized measurement procedures and practice with experienced technicians are essential to develop the level of skill and proficiency needed to obtain accurate and precise skinfold measurements. The accuracy and reliability of the BIA method will be acceptable for estimating body composition of elite heptathletes only if measurements are made using standard techniques, and appropriate prediction equations for female athletes are used to estimate %Fat values (29). Summary and Conclusions Body composition and weight are two of the many factors that contribute to optimal athletic performance. Body weight and composition influence an athleteʼs strength, speed, endurance, power, agility, and appearance. The %Fat and FFBMI values for high caliber female athletes vary among sports. Athletes who strive to maintain inappropriately low Wt or %Fat levels may be at risk for health problems related to poor energy and nutrient intakes, or at risk for developing an eating disorder. The results of this study indicate that elite American females heptathletes, at the lowest phase of their training cycle, are lean with low %Fat levels similar to elite,

10 Elite American Heptathletes / 171 endurance athletes and yet have high levels of BMD. These heptathletes also have a large amount of FFM relative to their Ht, which is a similar profile to field athletes. The cross validation of the SKF and BIA prediction equations indicated that the SKF equations provided better estimates of %Fat than values predicted from BIA equations in this group of homogeneous lean elite American female heptathletes. References 1. The American Dietetics Association. Position of the American Dietetic Association and the Canadian Dietetic Association: Nutrition for physical fitness and athletic performance for adults. J. Am. Diet. Assoc. 93: , Bartlett, H.L., M.J. Mance, and E.R. Buskirk. Body composition and expiratory reserve volume in female gymnasts and runners. Med. Sports Exer. 16: , Beals, K.A., and M.M. Manore. Nutritional status of female athletes with subclinical eating disorders. J. Am. Diet. Assoc. 98: , Boileau, R.A., and T.G. Lohman. The measurement of human physique and its effect on physical performance. Orthop. Clin. North Am. 8: , Brown, C.H., and J.H. Wilmore. The effects of maximal resistance training on the strength and body composition of women athletes. Med. Sci. Sports 6: , Brozek, J., F. Grande, J.T. Anderson, and A. Keys. Densitometric analysis of body composition: revision of some quantitative assumptions. Ann. N.Y. Acad. Sci. 110: , Burke, E.R. Physiological characteristics of competitive cyclists. The Physician and Sportsmedicine 8(7):78-84, Cohen, J.L., L. Potosnak, O. Frank, and H. Baker. A nutritional and hematologic assessment of elite ballet dancers. The Physician and Sportsmedicine 13:43, Drinkwater, B.L. Exercise and bone: lessons learned from female athletes. Am. J. Sports Med. 24:S33-S35, Dueck, C.A., K.S. Matt, M.M. Manore, and J.S. Skinner. Treatment of athletic amenorrhea with a diet and training intervention program. Int. J. Sports Nutr. 6:1, 24-40, Durnin, J.V., and J. Womersley. Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 years. Br. J. Nutr. 32:1, 77-97, Faber, M., A.J. Spinnler-Benade, and A. Daubitzer. Dietary intake, anthropometric measurements, and plasma lipid levels in throwing field athletes. Inter. J. Sports Med.11: , Going, S.B., M.P. Massett, M.C. Hall, L.A. Bare, P.A. Root, D.P. Williams, and T.G. Lohman. Detection of small changes in body composition by dual-energy x-ray absorptiometry. Am. J. Clin. Nutr. 57: , Going, S.B., and V.A. Mullins. Body composition of the endurance performer. In: Endurance in Sport, International Olympic Committee Encyclopedia (vol. 2, chap. 24), R.J. Shepard and P.O. Astrand (Eds.). Oxford, UK: Blackwell Scientific, in press. 15. Graves, J.E., M.L. Pollock, and P.B. Sparling. Body composition of elite distance runners. Int. J. Sports Med. 8:96-102, Hagerman, F.C., F.R. Hagerman, and T.C. Mickelson. Physiological profiles of elite rowers. The Physician and Sportsmedicine 7:74-83, Harmon, E.A., and P.N. Frykman. The relationship of body size and composition to performance of physically demanding military tasks. In: Body Composition and Phys-

11 172 / Houtkooper et al. ical Performance for the Military Service, B.M. Mariott and J. Grumstrup-Scott (Eds.). Washington, DC: National Academy Press, 1992, pp Haymes, E.M., and A.L Dickinson. Characteristics of elite male and female ski racers. Med. Sci. Sports Exer. 12: , Houtkooper, L.B., and S.B. Going. Body composition: how should it be measured? Does it affect sport performance. Sports Science Exchange (Gatorade Sports Science Institute). 7(5), Houtkooper L.B., S.B. Going, J. Sproul, R.M. Blew, and T.G. Lohman. Comparison of methods for assessing body composition changes over 1 year in postmenopausal women. Am. J. Clin. Nutr., i72: , Jackson A.S., M.L. Pollock, and A.Ward. Generalized equations for predicting body density of women. Med. Sci. Sports Exer. 12: , Jackson, A.S., and M.L. Pollock. Practical assessment of body composition. The Physician and Sportsmedicine 13:77-90, Keith, R.E., K.A. Oʼkeeffe, L.A. Alt, and K.L. Young. Dietary status of trained female cyclists. J. Am. Diet. Assoc. 89:1620, Kohrt, W. Preliminary evidence that DEXA provides an accurate assessment of body composition. J. Appl. Physiol. 84: , Krahenbuhl, G.S., C.L. Wells, C.H. Brown, and P.E. Ward. Characteristics of national world class female pentathletes. Med. Sci. Sports 11(1):20-23, Lohman, T.G. Applicability of body composition techniques and constants for children and youths. Exer. Sports Sci. Rev. 14: , Lohman, T.G, A.F. Roche, and R. Martorell (Eds.). Anthropometric Standardization Reference Manual. Champaign, IL: Human Kinetics, Lohman, T.G. Advances in Body Composition Assessment. Champaign, IL: Human Kinetics, Lohman, T.G., L. Houtkooper, and S.B. Going. Body fat measurement goes high-tech: not all are created equal. ACSMʼS Health & Fitness Journal. 1:30-35, Lukaski H.C., and W.W. Bolonchuk. Theory and validation of the tetrapolor bioelectrical impedance method to assess human body composition. In: K.J. Ellis, S. Yasamura, and W.D. Morgan (Eds.) In Vivo Body Composition Studies. London: The Institute of Physical Sciences in Medicine, 1987, pp Malina, R.M., A.B. Harper, and H.H. Avent. Physique of female track and field athletes. Med. Sci. Sports Exerc. 3:32-38, Malina, R.M. Physique and body composition: Effects of performance and effects on training, semistarvation, and overtraining. In: Eating, Body Weight, and Performance in Athletes, K.D. Brownell, J. Rodin, & J.H. Wilmore (Eds.). Philadelphia: Lea & Febiger, 1992, pp Mazess, R.B., H.S. Barden, J.P. Bisek, and J. Hanson. Dual-energy x-ray absorptiometry for total-body and regional bone-mineral and soft-tissue composition. Am. J. Clin. Nutr. 51: , Modelesky, C.M., Cureton, K.J., Lewis, R.D., Prior, B.M., Sloniger, M.A., and D.A. Rowe. Density of fat-free mass and estimates of body composition in male weight trainers. J. Appl. Physiol. 80: , Modelesky, C.M., E.M. Evans, M.L. Millard-Stafford, M.A. Collins, R.D. Lewis, and K.J. Cureton. Impact of bone mineral estimates on percent fat estimates from a four-component model. Med. Sci. Sports Exer. 31: , OʼToole, M.L., D.B. Hiller, L.O. Crosby, and P.M. Douglas. The ultraendurance triathlete: a physiological profile. Med. Sci. Sports Exerc. 19(1):45-50, 1987.

12 Elite American Heptathletes / Pate, R.R., C.A. Slentz, and D.P. Katz. Relationships between skinfold thickness and performance of health related fitness test items. Res. Q. Exerc. Sport 60: , Petterson, U., B.M. Stalnacke, G.M. Ahlenius, K. Henriksson-Larsen, and R. Lorentzon. Low bone mineral density at multiple skeletal sites, including the appendicular skeleton in amenorrheic runners. Calcif. Tissue Int. 64: , Prior, B.M., K.J. Cureton, C.M. Modlesky, E.M. Evans, M.A. Sloniger, M. Saunders, and R.D. Lewis. In vivo validation of whole body composition estimates from dual-energy X-ray absorptiometry. J. Appl. Physiol. 83: , Puhl, J., S. Case, and S. Fleck. Physical and physiological characteristics of elite volleyball players. Res. Q. 53: , Russell-Aulet, M., J. Wang, J. Thorton, and R. Pierson. Comparison of dual-photon absorptiometry systems from total body bone and soft tissue composition. Am. J. Clin. Nutr. 6: , Sinning, W.E., and G.D. Lindberg. Physical characteristics of college age women gymnasts. Res. Q. 43: , Sinning, W.E. Body composition and athletic performance. In: Limits of Human Performance, D.H. Clark and H.M. Eckert (Eds.), 1985, pp Siri, W.E. The gross composition of the body. Advance Biology Medicine Physiology 4: , Siri, W.E. Body composition from fluid spaces and density: analysis of methods. In: Techniques for Measuring Body Composition, J. Brozek and A. Henschel (Eds.). Washington, DC: National Academy of Science, National Research Council, 1961, pp Slaughter, M.H., and T.G. Lohman. An objective method for measurement of musculoskeletal size to characterize body physique with application to the athletic population. Med. Sci. Sports Exer. 12: , Sloan, A.W., J.J. Burt, and C.S. Blyth. Estimations of body fat in young women. J. Appl. Physiol. 17: , Statistical Package for the Social Sciences, Inc., V Chicago, IL, Valhalla, Model 1990b, San Diego, CA: Valhalla Scientific. 50. Wilmore, J.H., C.H. Brown, and J.A. Davies. Body physique and composition of women distance runners. Ann. N.Y. Acad. Sci. 301: , Withers, R.T., D.A. Smith, B.E. Chatterton, C.G. Schultz, and R.D. Gaffney. A comparison of four methods of estimating body composition of male endurance athletes. Eur. J. Clin. Nutr. 46: , 1992.