Clin Sports Med 26 (2007) 37–68

CLINICS IN SPORTS MEDICINE

Body Composition in Athletes:

Assessment and Estimated Fatness
Robert M. Malina, PhD*
Tarleton State University, Stephenville, TX, USA

T
he study of body composition attempts to partition and quantify body
weight or mass into its basic components. Body weight is a gross measure
of the mass of the body, which can be studied at several levels from basic
chemical elements and specific tissues to the entire body. Body composition is
a factor that can influence athletic performance and as such is of considerable
interest to athletes and coaches. This article provides an overview of models
and methods used for studying body composition, changes in body composi-
tion during adolescence and the transition into adulthood, and applications
to adolescent and young adult athletes.

LEVELS OF BODY COMPOSITION

The study of body composition historically has been driven by the availability of
methods to measure or, more appropriately, to estimate it. Since the early 1980s,
considerable progress has been made in the development and refinement of tech-
niques to estimate the composition of the body, so that virtually all components of
the body can now be measured. This progress has resulted in the modification of
the models that provide the framework for studying body composition.
Body composition can be approached at a variety of levels. The five-level
approach provides a sound guide: atomic, molecular, cellular, tissue, and
whole body [1,2]. The multilevel view provides a framework within which
the lure and difficulty inherent in the study of body composition can be
appreciated.
Basic chemical elements compose the atomic level. There are 106 elements in
nature. About 50 are found in the human body, and with more recent techno-
logic advances, all 50 can be measured in vivo. Oxygen, carbon, hydrogen, and
nitrogen account for greater than 95% of body mass, and the addition of seven
other elements—sodium, potassium, phosphorus, chloride, calcium, magne-
sium, and sulfur—accounts for 99.5% of body mass [2].
The molecular level of body composition focuses on four of five components:
water, lipid (fat), protein, minerals, and carbohydrate. The last component,

*10735 FM 2668 Bay City, TX 77414. E-mail address: [email protected]

0278-5919/07/$ – see front matter ª 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.csm.2006.11.004 sportsmed.theclinics.com
38 MALINA

carbohydrate, occurs in small amounts in the form of glycogen, largely in the

liver and skeletal muscle, and is not usually considered in estimates of body
composition. The following equation is used:

Body mass ¼ water þ protein þ mineral þ fat

Most mineral is located in bone with a small fraction in other tissues. Histor-
ically, the relative contribution of each of the four components to body mass
was derived from chemical analyses of human cadavers, although each can
now be measured in vivo.
At the cellular level, body mass is viewed as composed of cells and substances
outside of cells. The body cell mass (BCM) is defined by intracellular fluids and
intracellular solids and is the metabolically active component of the body. Pres-
ently available methods do not permit measurement of cell solids in vivo. Ex-
tracellular fluids (ECF) and extracellular solids (ECS) compose the substances
outside of the cells. The primary ECF are bone minerals and other components
of connective tissues. Adipocytes are fat cells; they store lipids and comprise fat
mass (FM). The equation is as follows:

Body mass ¼ BCM þ ECF þ ECS þ FM

At the tissue level, the study of body composition focuses on the contribution
of specific tissues to body mass: skeletal muscle, adipose, bone, blood, viscera,
and brain. Skeletal muscle, adipose, and bone tissues historically have been
a primary focus in studies using traditional technologies, such as anthropome-
try and radiography. New technologies permit more refined assessment of
these primary tissues (eg, the mineral content of bone tissue or subcutaneous
versus internal adipose tissue).
The fifth level of body composition is the whole body, its size, shape, physique,
and proportions. Anthropometry is the basic tool for estimating body size and
configuration, although photographic techniques also have been used, espe-
cially for the study of shape and physique. The body mass index (BMI) (weight
[kg]/height2 [m]) and skinfold thicknesses are perhaps the most widely used an-
thropometric indicators at this level of body composition. Two other properties
of the whole body are crucial in the study of body composition—volume and
density.

MODELS OF BODY COMPOSITION

The two-component model historically has been the model of choice for parti-
tioning body mass into meaningful components or compartments. This tradi-
tional approach has evolved into more complex models that include three,
four, or more compartments.
BODY COMPOSITION IN ATHLETES 39

Two Components
The two-component model partitions body mass into its lean (fat-free mass
[FFM]) and fat (FM) compartments. The equation follows:

Body mass ¼ FFM þ FM

The term lean body mass is occasionally used, but FFM is more appropriate.
Lean body mass is a more anatomic concept that includes some essential lipids,
whereas FFM is a biochemical concept. This model has had the widest applica-
tion in the study of body composition, including many studies of athletes. FM
is the more labile of the two compartments; it is readily influenced by diet and
training. A shortcoming of the two-component model is the heterogeneous
composition of FFM; it includes water, protein, mineral (bone and soft tissue
mineral), and glycogen.

Three Components
The three-component model includes FM and partitions FFM into total body
water (TBW) and fat-free dry mass (FFDM). The equation is as follows:

Body mass ¼ TBW þ FFDM þ FM

Water is the largest component of body mass, and most is located in lean
tissues. FFDM includes protein, glycogen, and mineral in bone and soft tissues.

Four Components
With the development of techniques to measure bone mineral, the four-compo-
nent model is a logical extension of the preceding model. FFDM is partitioned
in bone mineral (BM) and the residual. The following equation is used:

Body mass ¼ TBW þ BM þ FM þ residual

Overview
All models include FM. It is the aspect of body composition that has received
and continues to receive most attention. Excess FM can have a negative influ-
ence on physical performance and is often viewed by coaches and trainers as
a major limiting factor in athletic achievements. Excessive fatness also is a major
independent risk factor for several degenerative diseases.
FM, although often treated as a singular component of body composition, is
heterogeneous. Fat, or more appropriately lipid, is physiologically divided into
essential and nonessential lipids. Essential lipids are vital components of cells
and are basic for a variety of physiologic functions; they constitute about
10% of total lipids in the body. The remaining lipids, 90% of total body lipids,
are nonessential. They are triglycerides, which provide a storage form of avail-
able energy and perhaps thermal insulation [2]. The small amount of essential
lipids in the body usually are not considered in estimates of body composition
40 MALINA

and usually are grouped with the residual component or with FM, depending
on the model and method of assessment.
FFM is highly correlated with overall body size. Partitioning FFM into frac-
tions has several problems. Error is inherent in the measurement of each com-
ponent, and the more components included in a model increases the chances of
error. The techniques available for the measurement of TBW, potassium, cal-
cium, and sodium each have associated error. When measured, these proper-
ties must be converted to the body composition component of interest. The
transformation of the measured property to the component is essentially math-
ematical and includes a variety of assumptions.
Multicomponent models are additive; it is assumed that the separately mea-
sured properties can be summed to provide an estimate of the whole. Thus, the
measurement of body composition is essentially an estimate of body composition.
The different models of body composition have been largely developed on
adults. They also assume that during periods of stable body mass, the various
components exist in a steady state, which means that they are constant, or the
relationships among components are constant. The assumption of constant re-
lationships has permitted the development of procedures to estimate the differ-
ent fractions of body mass in adults. The application of these procedures to
children and adolescents, to adults in different stages of the life cycle, and to
elite athletes requires care in interpretation of various estimates. The propor-
tions of each component and the relationships among components change dur-
ing growth and maturation and with aging. Systematic training for sport is an
additional factor that influences body composition.
METHODS FOR ESTIMATING BODY COMPOSITION
Methods of estimating body composition in vivo are numerous and often quite
complex (Table 1). The methods are sufficiently different in technique that one
may inquire whether they provide reasonably similar estimates of body compo-
sition. The formulas for estimating FFM or FM, or components of FFM, and
the assumptions underlying the procedures are based primarily on adults in
the general population (ie, nonathletes). Their application to growing and ma-
turing children and adolescents and to athletes may result in spuriously high or
low estimates.
Several commonly used methods are briefly described. Three have been
used regularly for some time—the measurement of body density (Db), TBW,
and potassium concentration. Two more recent methods—dual-energy x-ray
absorptiometry (DXA) and bioelectrical impediance analysis (BIA)—also are
described. The specific protocols for each of these methods and their limitations
are discussed in detail in Roche et al [3] and Heymsfield et al [4]. Anthropomet-
ric approaches also are briefly summarized.
Body Density—Densitometry
Density is mass per unit volume. The density of specific body tissues varies.
The density of lean tissues (1.100 g/cm3) is greater than the density of water
(1 g/cm3) and fat (0.9007 g/cm3). Density is inversely related to body fat
BODY COMPOSITION IN ATHLETES 41

Table 1
Summary of methods used to estimate body composition
Underwater weighing, gas displacement Estimates body density, which is converted
to % Fat
Isotope dilution Estimates total body water, which is
converted to FFM; compartments of total
body water also can be estimated
40
K whole-body counting Estimates potassium content of body,
which is converted to FFM
Dual-energy x-ray absorptiometry (DXA) Estimates bone mineral, also lean and fat
tissues
Bioelectrical impedance Estimates FFM
Neutron activation analysis Uses isotopes of nitrogen and calcium to
estimate lean tissue and mineral
Uptake of fat-soluble gases Estimates FM
24-hour urinary creatinine excretion Estimates muscle mass
3-Methylhistidine excretion Estimates muscle mass
MRI Estimates of fat, muscle, and bone without
ionizing radiation, plus chemical
composition
CT Estimates of bone, muscle, and fat
Ultrasound Estimates of fat, muscle, and bone
Radiography Estimates of fat, muscle, and bone
Anthropometry Estimates of subcutaneous fat and
predictions of FM and FFM
Data from Malina RM, Bouchard C, Bar-Or O. Growth, maturation, and physical activity. 2nd edition.
Champaign (IL): Human kinetics; 2004.

content: The greater the proportion of fat, the lower the Db. A measure of Db
permits an estimate of the percentage of body mass that is fat (% Fat).
The most common method for measuring Db is underwater (hydrostatic)
weighing, but air or helium displacement techniques also are used. Two formu-
las are used most often to convert Db to % Fat:

% Fat ¼ 4:570=Db  4:142 ½5

% Fat ¼ 4:950=Db  4:500 ½6

The formulas and their underlying assumptions are derived from adults.
The two equations give reasonably similar estimates of % Fat except for the
very lean and very obese [7].
The estimate of % Fat is based on the assumption that the densities of the fat
and fat-free components are known and are constant, and that adults are iden-
tical in composition except for variability in the proportion of fat. The propor-
tions and chemical composition of the various components of FFM change with
growth and maturation, in addition to interindividual differences in composi-
tion of FFM.
42 MALINA

An estimate of FM is derived by multiplication and subtraction:

FM ¼ body mass  % Fat

FFM ¼ body mass  FM

Total Body Water—Hydrometry

Water is the largest component of the body, varying between 55% and 65% of
body mass in normally hydrated young men, with lower values for women.
The TBW of 70-kg young men can range from 38 to 45 kg of water. Emphasis
is on normal hydration. TBW varies during the course of a day, depending on
fluid intake and physical activity level, especially strenuous exercise. It also
varies with severe protein-energy undernutrition and extreme obesity.
Most of the water in the body is in lean tissues. Water constitutes approxi-
mately 72% to 74% of FFM in normally hydrated adults, although the esti-
mated water content of FFM has been reported to vary between 67% and
74%. In contrast, adipose tissue is relatively nonaqueous and contains a small
proportion of water, about 20%.
The measurement of TBW provides an estimate of FFM. The process is
based on two principles of isotope dilution: Certain substances distribute them-
selves evenly throughout a fluid space or compartment of the body, and the
dilution of a known amount of substance, an isotope tracer, administered
into an unknown volume or mass enables the calculation of the unknown vol-
ume or mass. The protocol consists of administering a known amount of a sta-
ble isotope tracer, allowing it sufficient time to dilute or mix, and then
measuring its concentration after dilution and after correcting for the amount
of the tracer lost from the body by excretion or exhalation. Three isotopes gen-
erally are used to measure TBW: deuterated water (2H, deuterium), tritiated
water (3H, tritium), and 18O-labeled water (heavy isotope of oxygen).
TBW usually is measured in the morning after an overnight fast. The iso-
tope is administered to the subject based on body mass. Time is permitted for
its equilibration with body water, usually 2 to 4 hours depending on the iso-
tope. The concentration of the isotope in serum, urine, or saliva is measured.
Assuming that the percentage of water in FFM is constant in adults, FFM is
estimated as follows:

FFM ¼ TBW=0:732

FM ¼ body mass  FFM

TBW can be subdivided into water that is intracellular (ICW) and extracel-
lular (ECW). Estimates of ICW and ECW in young men are 57% and 43% [8].
BODY COMPOSITION IN ATHLETES 43

ECW usually is measured with the same isotope dilution principles as TBW
with either chloride or bromide as the isotope. ECW is composed mainly of
water in support and transport tissues: plasma, dense connective tissue (tendon,
cartilage, bone), interstitial lymph, and transcellular fluids (cerebrospinal fluid,
joint fluids). ICW corresponds closely to skeletal muscle mass, the work-
producing tissue of the body, but is not exclusively composed of it. After esti-
mating ECW, ICW usually is derived by subtraction:

ICW ¼ TBW  ECW

Body Potassium—Whole Body Counting

Potassium occurs primarily in cells and especially in skeletal muscle tissue.
Measurement of the concentration of potassium in the body can provide an es-
timate of FFM. This is done by measuring the amount of potassium-40 (40K),
a naturally occurring isotope of potassium that accounts for 0.0118% of the nat-
urally occurring potassium in the human body [9]. The concentration of 40K is
measured with highly sensitive detection instruments, whole body counters,
which count the gamma emissions of the naturally occurring potassium. A con-
stant proportion of potassium in FFM is assumed, but there is a sex difference
[10,11]. FFM is estimated as follows:

FFM ¼ mEq K=68:1 in males

FFM ¼ mEq K=64:2 in females

FM ¼ body mass  FFM

More recent studies indicate variation in the potassium content of FFM, spe-
cifically values lower than the proportions indicated here [9]. Most of the avail-
able data for estimating body composition from measures of total body
potassium use the constants reported by Forbes [10,11]. Total body potassium
per unit FFM tends to decline with age in adulthood and shows differences
between American blacks and whites [12,13].

Dual-Energy X-Ray Absorptiometry

DXA is used to measure bone mineral and soft tissue composition of the body.
It provides estimates for the total body and of specific regions in the form of
bone mineral, fat-free soft tissues (sometimes called bone-free lean tissue),
and fat. The method requires a low radiation exposure, 0.05 to 1.5 mrem, de-
pending on the machine and how quickly the total body scan is done [14]. The
DXA unit measures the attenuation of the low-dose x-ray beam as it passes
through different tissues of the body. How much of each photon beam is
44 MALINA

absorbed by the atoms in bone mineral and soft tissues of the body is recorded
during the scan and converted to estimates of bone mineral and soft tissues
[15]. The DXA instrument must be linked with appropriate computer algo-
rithms to derive estimates of bone mineral, fat-free soft tissue, and fat tissue
content of the total body. The algorithms also permit division of the body
into anatomic segments—arms, legs, trunk, and head—to permit estimates of
regional body composition.
The derivation of fat and fat-free soft tissue from DXA scans is based on the
ratio of soft tissue attenuation of the low-energy and high-energy photon beams
as they pass through the body. The attenuation of the low-energy and high-
energy soft tissues is known based on scans of pure fat and fat-free soft tissues
and theoretical calculations. It is assumed that the attenuation of fat and fat-free
soft tissues is constant. The attenuation for fat is lower than that for fat-free soft
tissues. Using these constants and the scans from the DXA unit, the amount of
fat and fat-free soft tissue is calculated.
The derivation of bone mineral requires adjustment for the soft tissue over-
lying bone. DXA technology provides an estimate of total body bone mineral
content (g) and total bone area (cm3). The ratio of total body bone mineral to
total bone area is used to estimate bone mineral density (g/cm3). DXA basically
measures the cross-sectional area of a scan (total bone area) and not bone vol-
ume; expressing bone mineral relative to bone area is only an approximation of
bone mineral density.
Several types of commercially available DXA instruments are presently in
use. Each type of unit has its own computer algorithms for deriving estimates
of body composition, and there are inter-instrument differences. There is con-
cern for the comparability of measurements, especially of soft tissue, from ma-
chines produced by different manufacturers. All of them assume that the
attenuation characteristics of bone, fat-free soft tissue, and fat are known and
constant [14].

Bioelectrical Impedance Analysis

The method of estimating body composition from BIA is based on the fact that
lean tissue has a greater electrolyte and water content than fat. This difference
in electrolyte content permits an estimate of FFM from the magnitude of the
body’s electrical conductivity or from the body’s impedance to an electrical cur-
rent as it flows from the source (usually the ankle) to the sink (usually the wrist)
electrodes. FFM has low impedance and high conductivity, whereas FM, which
has a relatively low water and electrolyte content, has high impedance and low
conductivity.
BIA uses an imperceptible electrical current, which is introduced into the
body via electrodes placed on the ankle. The injected current passes through
the body, and the voltage that is produced is measured in voltage-sensing elec-
trodes placed on the wrist [16,17]. The ratio of voltage to the current is imped-
ance. Impedance to the flow of the current is related to the shape, volume, and
length of the body, which is the conductor of the current. Because impedance is
BODY COMPOSITION IN ATHLETES 45

proportional to the geometry of the conductor, variation in body shape may be

a factor in the application of BIA.
BIA measures the voltage for the path from the ankle to the wrist. It yields
a measure of resistance, which is the major component of impedance. Resis-
tance usually is converted to TBW, which is transformed into an estimate of
FFM as described earlier. The equation used to convert resistance to TBW
usually includes stature.
Several types of commercially available BIA units are presently in use, and
there are differences among the units. They are portable (the size of a briefcase)
and relatively inexpensive. BIA also is convenient, rapid, and noninvasive. As
such, it is finding increased application for estimates of body composition. As
with other methods, BIA has many underlying assumptions, which need to
be verified.

Anthropometry
The use of anthropometric dimensions to estimate body composition has
a long history [18,19]. Skinfold measurements are used most often to predict
Db, which is converted to an estimate of % Fat. A variety of prediction equa-
tions incorporating skinfold measurements and other dimensions (height,
limb and trunk circumferences, skeletal widths) are available. Most are based
on samples of nonathlete adults, and several are based on children and ado-
lescents, although equations for athletes also are available [19–23]. Prediction
equations are sample specific and should be cross-validated. ‘‘Generalized’’
equations that adjust for age and the curvilinear relationship between skin-
fold thicknesses and Db also have been developed [24,25]. Equations based
on advances in body composition methodology and multicomponent models
also are available for the general population [26–28] and for female athletes
[29].
Some degree of error is inherent in the measurement of body dimensions
and body composition. Error associated with the measurements per se and
with prediction equations should be noted. The standard error of estimate as-
sociated with available equations to predict body density generally ranges be-
tween 2% and 5%. For a discussion of error associated with anthropometry
and the accuracy of body composition prediction equations, see Malina [30]
and Sun and Chumlea [31].
More recently, the BMI has found increasing use as an alternative method for
rapid assessment of body composition of athletes. The BMI is widely used in epi-
demiologic surveys of the weight status in adults: underweight (BMI <18.5 kg/m2),
overweight (BMI 25 < 30 kg/m2), and obesity (BMI 30 kg/m2) [32]. It also is
used as a screening device of overweight and obesity in children and adolescents
[33,34]. The BMI is reasonably well correlated with total and percentage body
fat in large and heterogeneous samples, although it has limitations especially
with youth. Correlations between the BMI and FFM and FM are reasonably
similar among youth, which would suggest that the BMI is an indicator of heavi-
ness and indirectly of body fatness [35]. At the extremes of heaviness, the BMI is
46 MALINA

probably a reasonable indicator of fatness in the general population. Given the

relatively large body size (height and mass) and relative leanness of athletes,
the BMI has limitations with athletes.

Applications
The methods described provide an estimate of FFM and FM. Db is converted
to % Fat; TBW, 40K, and BIA (resistance) yield estimates of FFM. The other
half of the two-component model is derived by subtraction. The three-compo-
nent model involves the simultaneous measurement of Db and TBW to derive
an estimate of % Fat, whereas the four-component model includes Db, TBW,
and total body bone mineral to estimate % Fat. Multicomponent models pro-
vide the advantage of greater accuracy of estimates [36]. The cost and technical
constraints of the required methodology often limit their applicability outside of
the clinical or laboratory setting, however.
Most of the available body composition data for athletes are derived from
the two-component model using Db. Data are less extensive for estimates de-
rived from TBW, 40K, multicomponent models, DXA, and BIA. The assump-
tions underlying the methods are based on nonathletic adults, and limitations
of applications to youth and adult athletes need to be recognized. Fat estimates
from densitometry are based on the assumption that the density of fat and lean
tissues is constant. FFM estimates from TBW and 40K are based on the as-
sumption that the water and potassium contents of the FFM are constant.
They also assume that the density of fat and lean tissues and the water and po-
tassium contents of the FFM are the same in children, adolescents, and adults,
which is not the case. An important issue in growing and maturing individuals
is the age at which adult density, water concentration, and potassium concen-
tration of FFM are achieved (see later).
BIA is finding increased application. The method is useful to describe the
body composition of groups, but estimates have large errors within individuals,
which limits its application. BIA is influenced by nutritional and hydration sta-
tus and is not sensitive to acute changes in electrolytes and fluids. There also is
significant variation between BIA machines produced by different manufac-
turers. The resistance (R) function of impedance is used most often with stature
(length of the conductor) to estimate FFM, but there is uncertainty about the
appropriate hydration factor to use in converting R to FFM. Other equations
use R and stature, in conjunction with body mass, circumferences, and skinfold
measurements to estimate FFM.
DXA is used most often to estimate bone mineral content and density. DXA
measures of total body bone mineral content also are used in the four-compo-
nent model to increase the accuracy of body composition estimates. The pro-
cedures require the measurement of Db, TBW, and bone mineral, and it is
unclear whether the time and expense involved markedly improve the accu-
racy of the body composition estimates.
DXA also is being used more often to estimate FFM, FM, and % Fat. The
accuracy of these estimates needs further verification relative to estimates
BODY COMPOSITION IN ATHLETES 47

derived from the more established methods of body composition assessment,

specifically densitometry and hydrometry.
An important issue in all body composition studies is validation. How accu-
rately does a given technique estimate the specific components of body compo-
sition? What is the appropriate criterion against which to compare estimates of
body composition derived from the different methods and models? Do the dif-
ferent methods and models provide the same estimates of FFM, FM, and %
Fat? The assessment of bone mineral content increases the accuracy of the
four-component model over the three-component model. The instrument to
measure bone mineral content is expensive, however: Is this expense justifiable
given the small increase in the accuracy of the body composition estimates?
These and other questions need to be considered in evaluating the application
of new technologies and multicomponent models.

CHEMICAL MATURITY
If the principles and methods for estimating body composition are to be accu-
rately applied to youth, including young athletes, it is important to determine
when during growth are adult values for the primary components of FFM at-
tained. When adult values are reached, chemical maturity is said to be attained.
The concept of chemical maturity was introduced by Moulton [37]: ‘‘The point
at which the concentration of water, proteins, and salts [minerals] becomes
comparatively constant in the fat-free cell is named the point of chemical matu-
rity of the cell.’’
During the interval of growth and maturation, approximately the first 2 de-
cades of life, the relative contribution of water to body mass decreases, and cor-
responding contributions of solids—protein, mineral, and fat—to body mass
increase. Growth is an accretive process, adding or accumulating solids at
the expense of fluids. The relative contributions of protein and mineral to
FFM also increase, whereas the relative contribution of water to FFM
decreases.
The estimated composition of FFM from late childhood through adolescence
into young adulthood is summarized in Table 2. With growth and maturation
during adolescence, the relative contribution of solids (protein and mineral) to
FFM increases, and that of water decreases. Sex differences are apparent. FFM
in males contains relatively less water and relatively more protein and mineral
compared with females. The sex difference also is reflected in the estimated po-
tassium content and density of the FFM, which are greater in males than in fe-
males. The difference reflects the sex difference in muscle mass and bone
mineral and persists into young adulthood. The gain in bone mineral between
age 10 years and young adulthood reflects, to a large extent, the growth and
maturation of the skeleton during the adolescent growth spurt. The relative
mineral content of FFM in males increases from 5.4% at about 10 years to
6.6% at 17 to 20 years, a small increment (1.2%) that is, however, about
22% of the initial value at age 10. The corresponding increase in the relative
mineral content of FFM in females is less, 5.2% to 6.1%, a relative increase
48 MALINA

Table 2
Estimated composition of the Fat-free mass during the transition into adolescence, adoles-
cence, and young adulthood

Compartments of the FFM (%)

Potassium Density
Age (y) Water Protein Mineral (g/kg) (g/cm3)
Males
9–11 76.2 18.4 5.4 2.45 1.084
11–13 75.4 18.9 5.7 2.52 1.087
13–15 74.7 19.1 6.2 2.56 1.094
15–17 74.2 19.3 6.5 2.61 1.096
17–20 74.0 19.4 6.6 2.63 1.099
Females
9–11 77.0 17.8 5.2 2.34 1.082
11–13 76.6 17.9 5.5 2.36 1.086
13–15 75.5 18.6 5.9 2.38 1.092
15–17 75.0 18.9 6.1 2.40 1.094
17–20 74.8 19.2 6.0 2.41 1.095
The protein content of the FFM in the estimates is derived by subtraction: 100  water  mineral ¼ protein.
Data from Lohman TG. Applicability of body composition techniques and constants for children and
youths. Exerc Sport Sci Rev 1986;14:325–57.

of about 16% of the initial value in late childhood. It is apparent that chemical
maturity of FFM is not attained until after the adolescent growth spurt, prob-
ably about 16 to 18 years in girls and 18 to 20 years in boys.
The information summarized in Table 2 represents estimates of the chemical
composition of the FFM. The estimates are derived in part from limited biochem-
ical cadaver analyses and from in vivo estimates of TBW, potassium, nitrogen,
calcium, and bone mineral. Estimates also vary from laboratory to laboratory,
which is not unexpected because different data, assumptions, and methods are
used in their derivation. There are continued efforts to arrive at the more accurate
estimates of the chemical composition of the FFM. Nevertheless, an important
conclusion to be derived from these estimates and ongoing studies is that the
chemical maturity of FFM changes during growth and maturation and is not at-
tained until late adolescence or young adulthood. The equations and constants
based on adult values and earlier studies are often adjusted for the chemical im-
maturity of the FFM in growing and maturing individuals.

CHANGES IN BODY DENSITY, TOTAL BODY WATER,

AND TOTAL BODY POTASSIUM DURING GROWTH
The basic properties used for estimating body composition change with growth
and maturation [35,38,39]. TBW and total body potassium, both of which are
found primarily in lean tissue, follow a growth pattern similar to that of height
and body mass with a clear adolescent spurt during which growth in TBW and
total body potassium is greater in males than in females. Both reach a plateau at
about 15 to 17 years in females and increase into the early 20s in males. Db has
BODY COMPOSITION IN ATHLETES 49

a different growth pattern. It declines in males from about 8 to 10 years, but

then increases more or less linearly to about 16 to 17 years of age. In females,
Db decreases from about 8 to 11 years, then increases only slightly, and finally
reaches a plateau by about 14 years. Both sexes also show a slight decline in Db
in late adolescence and young adulthood. Db is inversely related to body fat
content, although not linearly. Males have greater Db than females at all
ages and have a correspondingly lower % Fat.

GROWTH IN FAT-FREE MASS, FAT MASS, AND FAT

PERCENTAGE OF BODY MASS
The growth patterns of FFM, FM, and % Fat from late childhood through ad-
olescence into young adulthood have been described [35]. FFM follows
a growth pattern similar to that of height and weight. Sex differences are small
during childhood and become established during the adolescent growth spurt.
Young adult values of FFM are reached earlier in females, about 15 to 16 years,
compared with 19 to 20 years in males. In late adolescence and young adult-
hood, males have, on average, an FFM that is about 1.5 times larger than
that of females. The average FFM of young women is about 70% of the
mean value for men. The difference reflects the male adolescent spurt in muscle
mass and the sex difference in adult height.
When FFM is expressed per unit height (kg/cm), the sex difference is rela-
tively small in late childhood and early adolescence, but after 14 years of
age, males have more FFM for the same height as females. The sex difference
increases with age so that young men have an estimated 0.36 kg FFM/cm
height, whereas young women have an estimated 0.26 kg FFM/cm height.
Estimated FM increases more rapidly in females than in males from late
childhood through adolescence, but seems to reach a plateau or to change
only slightly near the time of the adolescent growth spurt in boys (about 13–
15 years). In contrast to FFM, females have, on average, about 1.5 times the
FM of males in late adolescence and young adulthood.
Estimated % Fat is greater in females than in males from late childhood
through adolescence into young adulthood. Percent Fat increases gradually
through adolescence in the same manner as FM in females; it also increases
gradually in males until just before the adolescent spurt (about 11–12 years)
and then gradually declines. Percent Fat reaches its lowest point at about 16
to 17 years in males and then gradually increases into young adulthood. In con-
trast to estimates of FM, % Fat declines during male adolescence. The decline is
due to the rapid growth of FFM and slower accumulation of FM at this time.
As such, fat contributes a lesser percentage to body mass in male adolescence.
To illustrate the impact of adolescence on body composition, differences in
estimates of body composition between early and late adolescence are summa-
rized in Table 3. Corresponding data for a mixed-longitudinal sample of youth
from the Fels Longitudinal Study are included for comparison [40]. The Fels
data are derived from subjects who had at least six serial measurements
of Db between 8 and 23 years. A multicomponent model that included
50 MALINA

Table 3
Estimated differences in densitometric estimates of body composition from early to late
adolescence
Males Females
a
Composite sample
FFM 32.5 kg 17.3 kg
FM 3.2 kg 7.1 kg
% Fat 2.7% þ5.0%
Fels sampleb
FFM 31.3 kg 14.0 kg
FM 3.4 kg 7.4 kg
% Fat 3.5% þ3.8%
a
Adapted from Malina et al. [39] and Malina [38], compiled from the literature. Estimates are differences
between 10–11 and 18–19 years.
b
Estimated from Guo et al. [40], mixed-longitudinal data. Estimates are the differences between two age
groups, 10–12 and 18–20 years.

age-specific and sex-specific estimates of the density and major components of

FFM was used to derive % Fat, FM, and FFM [40]. Allowing for methodologic
variation, the estimates are reasonably similar. Males gain almost twice as much
FFM as females over adolescence, and females gain about twice as much FM
as males. The net result is a decline in relative fatness in males and an increase
in relative fatness in females.
A major portion of changes in body composition over adolescence is concen-
trated in the interval of rapid growth in height. Peak height velocity (PHV) oc-
curs, on average, at 12 and 14 years of age in a sample of North American and
European girls and boys [35]. Assuming 1 year each side of PHV, the interval
between 11 and 13 years in girls and 13 and 15 years in boys provides an es-
timate of changes in body composition at this time (Table 4). The interval of
maximal growth in height accounts for about 40% of the total adolescent
gain in FFM and FM. Although these estimates are based on cross-sectional
data from the literature, they correspond reasonably well with limited data
from longitudinal studies. Czech boys, many of whom were athletes, gained

Table 4
Estimated changes in FFM, FM and % Fat during the interval of maximal growth in height dur-
ing the adolescent spurta
Total gain/loss Annual gain/loss
Females Males Females Males
FFM 7.1 kg 14.3 kg 3.5 kg/yr 7.2 kg/yr
FM 2.8 kg 1.5 kg 1.4 kg/yr 0.7 kg/yr
% Fat 1.7% 1.1% 0.9%/yr 0.5%/yr
Based on the composite data summarized in Table 2.
a
Adolescent spurt: 11–13 years in girls and 13–15 years in boys.
BODY COMPOSITION IN ATHLETES 51

about 7.5 kg/y in FFM and 0.8 kg/y in FM and declined about 0.4%/y in % Fat
near the time of PHV [41]. The decline in relative fatness is an effect of the
larger increase in FFM, so that FM, although increasing slightly, contributes
a small percentage of body mass at this time. In the mixed-longitudinal sample
from the Fels study (see Table 2), peak gain in FFM was about 7 kg/y in boys,
whereas no clear spurt in FFM was evident in girls [40]. In a separate analysis
of adolescent changes in total body bone mineral content, peak gains occurred,
on average, a bit more than one half of a year after PHV in both sexes. Esti-
mated peak gain in BMC was, on average, greater in boys, 407  93 g/y,
than in girls, 325  67 g/y [42]. Peak gain in bone mineral content occurred
closer to the age of menarche, suggesting that adolescent growth in bone min-
eral is closely related to sexual maturation in girls [42].

BODY COMPOSITION OF ATHLETES

Why Estimate the Body Composition of Athletes?
It is important to know how the different components of body composition
vary with age, sex, and maturity status, especially during adolescence. Like-
wise, it is important to understand the influence of systematic training for
sport on body composition. Many discussions of body composition in ath-
letes focus on relative fatness because of the potentially negative influence
of excessive fatness on performance. Focus of such discussions is often on
female athletes, usually in the context of low levels of fatness, late sexual
maturation, menstrual irregularities, and disordered eating. Weight control
and in many instances reduction are issues with many athletes. Associated
concerns are potential metabolic complications with long-term weight reduc-
tion [23] and weight control behaviors that may lead to complications of dis-
ordered eating, particularly among female athletes. Among male athletes,
there is concern for low levels of fatness of males in weight category sports,
specifically wrestling.

Youth Athletes
The body composition of young athletes is influenced by their growth and ma-
turity status. With few exceptions, young athletes of both sexes tend to be at or
above median reference values in height and mass; exceptions are gymnasts of
both sexes and female figure skaters. Elite young male athletes tend to be, on
average, advanced in maturity status, although there is variation among sports.
Earlier maturation in males is associated with larger size and FFM, greater
strength and power, and a lower % Fat compared with average (‘‘on time’’)
and later maturing peers of the same chronologic age. The size, strength,
and power associated with earlier maturation in males are an advantage in
many sports. Elite young female athletes tend to be, on average, average and
later in maturity status compared with peers of the same chronologic age. Later
maturation in females is associated with smaller body size, a more linear phy-
sique, lower % Fat, and generally better performances compared with early
52 MALINA

maturing peers [35]. Discussions of the body composition of young athletes

should consider individual differences in maturity status; studies rarely con-
sider maturity-association variation among young athletes. Variation associated
with individual differences in biologic maturation is a potential confounding
factor in evaluating the body composition of young athletes.
Because FFM follows a growth pattern similar to that for height and
mass, and FFM is highly correlated with height and mass, most studies of
body composition of young athletes emphasize % Fat. As noted, excessive
fatness tends to exert a negative influence on performances, especially per-
formances that require the movement or projection of the body through
space (ie, running, jumping, vaulting), in contrast to those that require pro-
jection of objects (ie, shot put, discus throw). Coaches of young athletes of-
ten focus on weight control and relative fatness. Middle school, high school,
and collegiate wrestling currently set minimum weight or % Fat require-
ments (see later).
Estimates of the relative fatness of adolescent athletes in a variety of sports
are shown relative to data for nonathletes in Fig. 1. The estimates are means
based on densitometry with one exception; data for a sample of female gym-
nasts based on DXA also are included. Allowing for variation among samples
and in methodology, athletes as a rule have a lower % Fat than nonathletes of
the same chronologic age. Male athletes and nonathletes show a decrease in %
Fat during adolescence; athletes have less relative fatness at most ages, but
there is considerable overlap (Fig. 1A). In contrast, % Fat in female athletes
tends to be reasonably stable across adolescence, whereas that for nonathletes
increases with age (Fig. 1B). The difference in % Fat between female athletes

A MALE ATHLETES B FEMALE ATHLETES

25 25
% FAT % FAT

20 20
%

15 15

10 10

5 5
8 10 12 14 16 18 20 8 10 12 14 16 18 20
Age, years Age, years

Fig. 1. Estimates of % Fat in samples of youth athletes. (A) Males. (B) Females. Male athletes
include cyclists, wrestlers, gymnasts, runners, jumpers, and volleyball, ice hockey, and Amer-
ican football players. Female athletes include swimmers, runners, jumpers, gymnasts, and
speed skaters. (See ref. [35] for sources of data. Data for the nonathlete reference from Malina
et al [35,39].)
BODY COMPOSITION IN ATHLETES 53

and nonathletes is greater than that between male athletes and nonathletes. Al-
lowing for the sports represented, there seems to be more variation in % Fat
among female than male athletes 14 to 18 years old.

Young Adult Athletes

Estimated % Fat in young adult male and female athletes in numerous sports is
summarized in Tables 5 and 6. Most data are based on body density derived
from hydrostatic weighing; estimates based on other methods are limited. Data
for youth and adult track and field athletes by discipline are considered sepa-
rately (see later).
The data summarized in Tables 5 and 6 provide an overview of estimated
relative fatness and associated variation in young adult athletes in a variety
of sports. Samples sizes are generally small, and many sports are not well rep-
resented. Corresponding estimates for % Fat predicted from skinfold measure-
ments in national-level Polish athletes [43] and American intercollegiate female
athletes in numerous sports are summarized in Tables 7 and 8, respectively.
Within corresponding sports, % Fat of female athletes is, on average, greater
than that of male athletes. Variation in % Fat by position or event within a sport
is apparent for American football players (see Table 5), but is not marked
among female basketball, volleyball, and track and field athletes, with the ex-
ception of the throwing events (Table 8).
The ethnic composition of samples of athletes is not usually reported; this
is relevant given ethnic variation in body proportions and composition [13].
Ethnic variation in bone mineral content between American blacks and
whites has been long documented [13] and influences estimates of Db via
hydrostatic weighing [44]. Observations using more recent technology indi-
cate that black women have a larger appendicular skeletal muscle mass than
white women matched for age, body size, and menstrual status [45,46]. The
ethnic difference is more marked in the upper than the lower extremity [45].
Limited data for American black and white males show less ethnic variation
in limb musculature, but after adjustment for age, height, and mass, skeletal
muscle mass of the legs is significantly larger in American blacks of both
sexes. The ethnic difference also persists after adjusting for variation in
relative leg length. Arm skeletal muscle mass also is significantly larger
in American blacks of both sexes after adjusting for age and mass and
for relative arm length [47].
The preceding observations on ethnic variation are based on nonathletes.
Corresponding body composition data for athletes are limited. Estimated fat-
ness (hydrostatic weighing) of collegiate football players is lower in black
(14.7  5.6%, n ¼ 55) than white (19.0  7.1%, n ¼ 35) players [48]. Relative
fatness estimated from skinfold measurements gives similar estimates, but BIA
tends to overestimate % Fat based on Db, and near-infrared spectrophotometry
tends to underestimate % Fat based on Db in black and white players [48].
Among female collegiate athletes, differences in estimated % Fat predicted
from skinfold measurements between black and white athletes are, on average,
54 MALINA

Table 5
Relative fatness (% Fat) in samples of male athletes in several sports
Age (y) % Fat
Sport n Mean SD Method Mean SD Reference
Badminton 7 24.5 3.6 HW 12.8 3.1 [21]
Baseball 10 20.8 9.9 TBW 14.2 6.7 [69]
Basketball 10 20.9 1.3 HW 10.5 3.8 [62]
Basketball 11 25.7 3.1 HW 9.7 3.1 [21]
Canoeing/kayaking 19 21.1 7.1 HW 13.0 2.5 [65]
Cycling 11 22.2 3.6 HW 10.5 2.4 [21]
Cycling 11 21.7 1.7 TBW 13.7 2.3 [66]
Cycling 13 24.1 3.1 HW 11.2 3.3 [67]
Cycling 63 21.9 3.2 HW 11.8 3.3 [65]
Field hockey 14 23.7 3.6 HW 10.3 4.4 [21]
Football by modality
40
American football 21 19.9 K 9.5 [68]
American football 16 20.3 0.9 TBW 13.8 6.7 [69]
American football 65 17–23 HW 15.0 5.8 [70]
Defensive back 15 HW 11.5 2.7
Offensive back, 15 HW 12.4 5.3
receiver
Defensive lineman 15 HW 18.5 4.4
Defensive 7 HW 13.4 4.1
linebacker
Offensive lineman 13 HW 19.1 7.0
American football [71]
Defensive back 26 24.5 3.2 HW 9.6 4.2
Offensive back, 40 24.7 3.0 HW 9.4 4.0
receiver
Quarterback, 16 24.1 2.7 HW 14.4 6.5
kicker
Defensive lineman 32 25.7 3.4 HW 18.2 5.4
Defensive 28 24.2 2.4 HW 14.0 4.6
linebacker
Offensive lineman 38 24.7 3.2 HW 15.6 3.8
American football, 55 19.4 1.2 HW 14.7 5.6 [48]
blacks
American football, 35 19.7 1.5 HW 19.0 7.1 [48]
whites
Australian rules 23 24.5 4.3 HW 8.0 3.0 [21]
Rugby union 16 24.2 3.3 HW 10.3 3.2 [21]
Soccer 9 24.8 1.9 TBW 6.2 1.9 [72]
Soccer 18 26.0 — HW 9.6 — [73]
Soccer 22 24.5 3.5 HW 6.9 3.3 [74]
Soccer 12 25.3 4.0 HW 9.7 3.0 [21]
Gymnastics 7 20.3 0.9 TBW 4.6 3.3 [69]
Gymnastics 8 20.2 2.7 HW 7.9 1.4 [21]
Ice hockey 27 24.9 3.6 HW 9.2 4.6 [75]
Lacrosse 26 26.7 4.2 HW 12.3 4.3 [21]
Rowing 8 24.7 3.2 TBW 7.3 1.3 [72]
Rowing 7 24.7 1.9 HW 11.2 1.4 [21]
(continued on next page)
BODY COMPOSITION IN ATHLETES 55

Table 5
(continued )
Age (y) % Fat
Sport n Mean SD Method Mean SD Reference
Skiing 9 25.9 2.9 HW 6.3 1.9 [76]
Skiing, cross-country 11 22.8 1.9 HW 7.2 1.9 [75]
Skiing, cross-country 11 24.0 4.5 HW 12.3 4.6 [65]
Speed skating 33 18.4 2.9 HW 11.2 2.8 [65]
Speed skating 6 22.2 4.1 HW 7.4 2.5 [78]
Squash 9 22.6 6.8 HW 11.2 3.7 [21]
Swimming 7 20.6 1.2 TBW 5.0 4.5 [69]
Swimming 13 21.8 2.2 HW 8.5 2.9 [76]
Swimming 14 19.9 2.3 TBW 7.5 3.0 [72]
Swimming 39 19.1 4.5 HW 12.3 4.6 [65]
Volleyball 19 23.8 3.2 HW 11.2 2.8 [65]
Volleyball 11 20.9 3.7 HW 9.8 2.9 [21]
Water polo 10 25.8 4.6 TBW 8.8 2.6 [72]
40
Abbreviations: HW, hydrostatic weighing; TBW, total body water; K, potassium 40.

quite small (see Table 8) and within the range of error associated with the pre-
diction equation [49].
Adolescent and Young Adult Track and Field Athletes
Data on the size, physique, and body composition of track and field athletes in
specific events within the sport are more extensive compared with other sports
and span early adolescence through young adulthood. The literature dealing
with track and field athletes is diverse and can be summarized in the frame-
work of four general themes: (1) talent identification and selection; (2) interest
in the growth, body composition, and functional characteristics of elite young
athletes in a variety of sports; (3) increased popularity of distance running for
children and adolescents; and (4) interest in the comparative morphology of
athletes in general. The data often include estimates of % Fat based primarily
on measured Db and predicted Db based on skinfold measurements; estimates
based on other methods are limited [50]. The data permit evaluation of varia-
tion in % Fat by event. Estimates of mean % Fat are summarized in Figs. 2, 3,
and 4 for sprinters and hurdlers, middle and long distance runners, and
jumpers and throwers. Corresponding estimates for specific jumping and
throwing disciplines, pole vaulters, race walkers, and decathletes are not exten-
sive. Data for adolescents and young adults from the general population are in-
cluded for comparison.
Most samples of male sprinters and hurdlers tend to be below the reference
in % Fat from early adolescence into young adulthood, although there is some
overlap during adolescence (Fig. 2A). In contrast, % Fat of female sprinters and
hurdlers is well below the reference (Fig. 2B).
Estimates of % Fat among middle and long distance runners show consider-
able overlap within sex, although there seems to be more variation among
56 MALINA

Table 6
Relative fatness (% Fat) in samples of female athletes in several sports
Age (y) % Fat
Sport n Mean SD Method Mean SD Reference
Badminton 6 23.0 5.3 HW 21.0 2.1 [22]
Basketball 18 22.9 2.6 HW 20.1 4.0 [22]
Canoeing/kayaking 21 21.2 3.7 HW 22.2 4.6 [65]
Field hockey 13 19.8 1.4 HW 21.3 7.1 [79]
Field hockey 17 22.6 2.3 HW 20.2 6.0 [22]
Field hockey 10 19.8 1.2 DXA 18.3 2.7 [80]
Gymnastics 5 19.0 3.8 TBW 12.9 1.4 [81]
Gymnastics 44 19.4 1.1 HW 15.3 4.0 [82]
Gymnastics 15 19.8 1.0 DXA 19.1 2.2 [80]
Gymnastics, rhythmic 7 20.7 2.7 HW 15.6 5.1 [65]
Handball, team 17 23.2 1.9 HW 19.0 3.7 [65]
Lacrosse 17 24.4 4.5 HW 19.3 5.7 [22]
Netball 7 23.7 4.2 HW 17.8 3.8 [22]
Rowing 19 23.6 3.9 HW 18.4 3.9 [65]
Rowing 22 20.4 1.9 DXA 21.9 2.3 [80]
Rowing, lightweight 5 19.4 7.5 HW 20.7 3.1 [22]
Rowing, heavyweight 7 20.5 3.4 HW 24.2 4.2 [22]
Skiing, cross country 5 23.5 4.7 HW 16.1 1.6 [77]
Soccer 10 24.4 4.5 HW 20.8 4.7 [83]
Soccer 11 22.1 4.1 HW 22.0 6.8 [22]
Soccer 10 19.8 0.9 DXA 21.8 2.7 [80]
Softball 14 22.6 4.0 HW 19.1 5.0 [22]
Softball 17 20.4 1.4 DXA 20.9 3.9 [80]
Speed skating 9 19.7 3.0 HW 16.5 4.1 [78]
Squash 6 27.4 5.6 HW 16.0 4.9 [22]
Swimming 19 19.2 0.8 HW 16.1 3.7 [20]
Tennis 7 21.3 0.9 HW 22.4 2.0 [79]
Volleyball 36 21.7 2.5 HW 15.8 4.8 [65]
Volleyball 13 23.0 2.6 HW 11.7 3.7 [84]
Volleyball 13 21.5 0.7 HW 18.3 3.4 [84]
Volleyball 11 22.8 3.4 HW 17.0 3.3 [22]
Abbreviations: HW, hydrostatic weighing; TBW, total body water.

females than males (Fig. 3). Most estimates for males are below the reference dur-
ing adolescence into young adulthood (Fig. 3A). The reported estimate for a sam-
ple of 17-year-old male distance runners (2.6%) is spuriously low. As noted, % Fat
of female middle distance and distance runners varies (Fig. 3B). Most values are
below the reference value, but several approach the reference in later adoles-
cence. In young adulthood, however, estimates of % Fat are considerably lower,
especially in distance runners.
Estimated % Fat of field athletes in jumping and throwing events shows two
distributions with little overlap (Fig. 4). The relative fatness of male jumpers is
consistently below the reference from adolescence into adulthood. In contrast,
% Fat of male adolescent throwers is consistently above the reference, reflecting
BODY COMPOSITION IN ATHLETES 57

Table 7
Anthropometric estimates of predicted relative fatness (% Fat) in national level Polish athletes
19–26 years old
Males % Fat Females % Fat
Sport n Mean SD n Mean SD
Basketball 24 10.4 2.5 31 20.1 4.8
Biathlon 10 10.8 2.0 10 17.2 3.9
Cycling 29 11.6 1.4 — — —
Fencing 21 12.0 3.4 15 21.9 4.5
Handball, 23 12.7 3.1 29 21.9 4.7
team
Kayaking 29 9.6 2.7 26 16.3 4.3
Rowing 31 10.7 3.0 22 17.7 3.8
Skiing 13 9.6 1.5 10 17.0 1.8
Soccer 30 9.7 2.1 — — —
Swimming — — — 13 15.5 2.5
Tennis — — — 7 17.7 1.5
Volleyball 26 10.8 1.7 — — —
Relative fatness was estimated from predicted Db using three skinfolds after Piechaczek [85].
Data from Krawczyk B, Sklad M, Majle B. Body components of male and female athletes representing
various sports. Biol Sport 1995;12:243–50.

in part their massiveness. Percent Fat tends to decline with age, however,
among samples of throwers. Among young adult throwers, estimates for sev-
eral samples are lower than corresponding values for adolescent throwers
and fluctuate above and below the reference values (Fig. 4A).
Female jumpers have relative fatness levels consistently below the reference,
and % Fat in young adult jumpers tends to be lower than that for adolescents
(Fig. 4B). Percent Fat of female throwers varies above and below the reference
in adolescence and young adulthood, which is in contrast to the trend in male
throwers. In female jumpers and throwers, variation in % Fat is greater among
young adults than among adolescents.
Estimates of % Fat in male track and field athletes except for throwers are
generally below the reference. There is considerable overlap among mean esti-
mates for sprinters and hurdlers, middle distance and distance runners, and
jumpers, and all have % Fat that is generally lower than in samples of throwers.
Trends in % Fat among adolescent track and field athletes should be viewed in
two contexts: first, the decline in % Fat that accompanies the growth spurt and
sexual maturation in males, and second, the decline in % Fat associated with
systematic training [35].
In female track and field athletes, there is considerable overlap in mean esti-
mates of % Fat among for runners of all disciplines and jumpers, and all are, on
average, generally below the reference. Relative fatness of female throwers
tends to be higher, but estimates fall above and below the reference.
Comparison of % Fat of male and female athletes within the same track and
field events suggests several trends: (1) Percent Fat in male sprinters and
58 MALINA

Table 8
Anthropometric estimates of predicted relative fatness (% Fat) in female university athletes by
sport, event/position, and ethnicity

Age (y) % Fat

Sport n Mean SD Mean SD
Swimming 87 18.8 0.9 16.5 1.6
Freestyle 37 18.8 0.8 16.3 1.6
Backstroke 18 18.9 1.0 16.8 1.9
Breaststroke 17 18.7 0.7 16.4 1.3
Butterfly 9 19.1 1.0 16.4 1.9
Medley 6 18.9 1.4 17.5 0.8
Diving 19 19.5 1.6 17.4 2.6
Tennis 29 19.0 0.9 17.7 2.4
Golf 32 19.0 0.9 19.1 2.2
Basketball 57 19.5 1.2 16.4 2.2
Guard, white 13 19.4 0.9 16.9 2.1
Guard, black 7 19.6 1.2 14.9 1.2
Wing, white 6 19.3 0.9 16.1 1.4
Wing, black 11 19.8 1.5 15.5 1.7
Post, white 11 19.6 1.4 17.4 3.0
Post, black 9 19.6 1.6 17.1 2.5
Volleyball 47 19.1 0.9 16.9 2.7
Outside hitter 24 19.2 1.0 16.9 3.0
Middle blocker 12 18.7 0.7 17.8 2.6
Setter 6 19.2 0.8 17.0 1.9
Back row 5 19.2 0.5 14.9 1.3
Track and field 116 19.4 1.2 15.4 3.8
Sprint, white 5 21.1 1.4 14.1 0.3
Sprint, black 38 19.3 1.4 14.1 1.4
Middle distance, 6 18.9 0.6 13.8 1.6
white
Middle distance, 5 18.6 0.8 13.6 0.7
black
Distance, white 29 19.3 1.2 14.2 1.6
Jumps, white 9 19.3 0.9 15.1 1.8
Jumps, black 7 20.0 1.3 14.6 1.4
Throws, white 11 19.5 1.1 22.4 4.1
Throws, black 5 18.7 0.5 23.9 7.8
Relative fatness was estimated from predicted Db from the sum of four skinfolds skinfolds after Meleski et al
[20]; see text.
Data from Malina RM, Battista RA, Siegel SR. Anthropometry of adult athletes: concepts, methods and
applications. In: Driskell JA, Wolinsky I, editors. Nutritional assessment of athletes. Boca Raton (FL): CRC
Press; 2002. p. 135–75.

hurdlers is only slightly lower than and overlaps the male reference, whereas
corresponding estimates of % Fat in female athletes in these events are well be-
low the female reference. (2) Percent Fat varies from early adolescence through
adolescence into young adulthood in male and female middle distance and dis-
tance runners; it is considerably lower than the reference among female run-
ners, whereas % Fat in male runners approximates the male reference at
BODY COMPOSITION IN ATHLETES 59

A MALES B FEMALES
SPRINTS AND HURDLES SPRINTS AND HURDLES
PERCENTAGE FAT PERCENTAGE FAT
27 30
24 27
21 24

Percentage Fat
Percentage Fat

18
21
15
18
12
15
9
6 12

3 9
0 6
10 12 14 16 18 20 22 24 10 12 14 16 18 20 22 24
Age, years Age, years

Fig. 2. Estimates of % Fat in samples of adolescent and young adult sprinters and hurdlers. (A)
Males. (B) Females. (See ref. [50] for sources of data. Data for the nonathlete reference from
Malina et al [35,39].)

many ages (3). The same general trend is apparent in jumpers and throwers.
Percent Fat in female jumpers is well below the female reference, whereas it
is quite close to the reference in male jumpers. Percent Fat in female throwers
straddles the reference, but is above the reference in male throwers, especially
in adolescence. Relative fatness tends to be lower in young adult compared
with adolescent male throwers; such variation with age is lacking in female
throwers.

A MALES B FEMALES
DISTANCE EVENTS DISTANCE EVENTS
PERCENTAGE FAT PERCENTAGE FAT
27 30
Middle Distance Middle Distance
24 Distance 27 Distance
21 24
Percentage Fat
Percentage Fat

18 21
15
18
12
15
9
12
6
3 9
0 6
10 12 14 16 18 20 22 24 10 12 14 16 18 20 22 24
Age, years Age, years

Fig. 3. Estimates of % Fat in samples of adolescent and young adult middle distance and dis-
tance runners. (A) Males. (B) Females. (See ref. [50] for sources of data. Data for the nonath-
lete reference from Malina et al [35,39].)
60 MALINA

A MALES B FEMALES
JUMPS, THROWS JUMPS, THROWS
PERCENTAGE FAT PERCENTAGE FAT
27 30
Jumpers Jumpers
24 Throwers 27 Throwers
21 24

Percentage Fat
Percentage Fat

18
21
15
18
12
15
9
6 12

3 9
0 6
10 12 14 16 18 20 22 24 10 12 14 16 18 20 22 24
Age, years Age, years

Fig. 4. Estimates of % Fat in samples of adolescent and young adult field athletes in jumping
and throwing events. (A) Males. (B) Females. (See ref. [50] for sources of data. Data for the
nonathlete reference from Malina et al [35,39].)

Wrestling
Weight management and specifically weight reduction are concerns in wres-
tling. Organizations associated with the sport have developed guidelines for
minimum weight, defined as the lowest weight that an individual can maintain
without comprising health [23]. Wrestling rules of the National Federation of
State High School Associations [51] mandates the following:
Each state association shall develop and use a weight-management pro-
gram that includes a specific gravity not to exceed 1.025; a body fat assess
no lower than seven percent for males/12 percent for females; and a mon-
itored weekly weight loss plan not to exceed 1.5 percent a week.

The American College of Sports Medicine [52] suggests the following for
young wrestlers:
Assess the body composition of each wrestler before the season using valid
methods for this population. Males 16 years old and younger with body fat
below 7 percent or those over 16 with a body fat below 5 percent need
medical clearance before being allowed to compete. Female wrestlers
need minimal body fat of 12–14 percent.

The National Collegiate Athletic Association uses 5% Fat as the minimum

allowable in weight certification formulas for wrestlers; however, body compo-
sition guidelines emphasize that estimates of body composition focus on a range
of variation [53].
A key issue in putting the proposed guidelines into action relates to methods
of estimating body composition and associated errors. Estimates derived from
different methods may not be directly comparable. Some high school
BODY COMPOSITION IN ATHLETES 61

associations and school districts specify the type of estimate (DXA, BIA, skin-
fold measurements) at times with associated costs.
In one of the few detailed studies, body composition was assessed via hydro-
static weighing in the preseason, late season, and postseason in a sample of 9
male wrestlers 15.4  0.3 years old [54,55]. Percent Fat at the three measuring
points was 9.9  0.5%, 8.0  0.7% and 12.3  0.8%. Corresponding estimates
of FFM were 54.3  3.1 kg, 53.2  3 kg, and 56.2  3.1 kg. The changes are
relatively small and need to be viewed in the context of growth and maturation.
Although body mass declined from the preseason to late season and increased
into the postseason, linear growth, skeletal maturation, and growth-related
hormones were not affected.
Among collegiate wrestlers competing at national tournaments (top 5–10%
of wrestlers within each division), % Fat based on skinfold measurements
was compared from the preseason to the competition [56]. Body mass and %
Fat declined from the preseason to the tournament, 74.0  11.1 kg to 71.5
 10.4 kg for mass and 12.3  3.4% to 9.5  1.8%. Mean minimal weights
established for each wrestler in the preseason did not differ at the tournament.

TRAINING AND BODY COMPOSITION

Body composition is responsive to the demands of systematic training, as is evi-
dent in the observations on wrestlers. It is essential to consider separately obser-
vations in growing and maturing youth and in mature adults. Of relevance, are
changes in body composition associated with training greater than those associ-
ated with normal growth and maturation? The increase in FFM (hydrostatic
weighing) observed in youth regularly active in sport from 11 to 17 years old
would seem to suggest an increase greater than that expected with normal growth
and maturation [57,58]. This group of boys was larger in size and advanced in bi-
ologic maturation, however, compared with their age peers not engaged in sport.
Similarly, among boys 11 to 13 years old undergoing a 5-month endurance train-
ing program, larger FFM (40K) were observed in boys advanced in sexual matu-
ration [59], suggesting perhaps that the boys were in their growth spurts [35]. The
two studies highlight the difficulties in attempting to partition changes in body
composition associated with training from 11 to 17 years old or with a short-
term training program from the changes that accompany normal growth and mat-
uration during male adolescence. Much of the variation in body composition in
both studies was associated with a reduction in fatness.
Corresponding observations for girls are limited. A program of 30 minutes
of high-impact aerobic and strength training activities three times per week
for10 months was associated with increases in FFM (2.2  1.1 kg) and FM
(0.5  0.8 kg) in 9- to 10-year-old girls [60]. Girls of the same age, body size,
composition, and stage of sexual maturity who followed their normal pattern
of activity for 10 months also increased in FFM (1.4  1.4 kg) and FM (1.0 
0.8 kg). Both groups gained in FFM and FM, and there was, on average,
no decrease in fatness. There also was considerable overlap between the
trained and normal activity groups. Although the results are suggestive, they
62 MALINA

indicate difficulties inherent in attempting to partition growth-related from

training-related changes in estimated body composition in adolescents [35].
Studies comparing changes in body composition associated with training in
young adults generally compare pretraining and post-training means for FFM
and % Fat, but duration of training programs and frequency of training vary. In
addition, the samples do not comprise athletes. Using data reported by Wil-
more [61], differences between pretraining and post-training means in nine
studies of males 18 to 23 years old range from 0.2 kg to þ1.4 kg for FFM
(overall mean þ0.8 kg) and 0.4% to 3.0% for relative fatness (overall
mean 1.7%). Corresponding differences for 10 studies of females 18 to 22
years old range from 1.7 kg to þ1.5 kg for FFM (overall mean þ0.3 kg)
and 2.1% to þ3.1% for relative fatness (overall mean 0.4%). The persis-
tence of changes associated with training are not usually considered; a relevant
question is the following: How much training is needed to maintain changes in
body composition induced by systematic training?

SEASONAL VARIATION IN BODY COMPOSITION IN ATHLETES

An issue related to the influence of training on body composition is changes in
body composition of athletes during the course of a season. Studies commonly
compare the body composition of athletes before and at the close of a compet-
itive season [62,63]. Of potential relevance is variation in body composition as-
sociated with preseason or early season and in-season training protocols. This
variation was considered in a sample of 15 elite university-level female swim-
mers who had their body composition estimated via densitometry at three
points during a competitive season: October, December, and March [64].
Weight training with emphasis on high-repetition and low-resistance activities
typically preceded swim training early in the season. Decreases in body mass
(1.3  1.8 kg), FM (2.4  1.2 kg) and % Fat (3.8  1.9%) and an increase
in FFM (1.1  1.8 kg) characterized the early part of the season between Oc-
tober and December. These changes were generally maintained during the sec-
ond part of the season, December to March, as the swimmers tapered in
preparation for the national championships. Over this interval, changes in
body mass (0.8  1.2 kg), FM (0.8  1.5 kg) and % Fat (1.2  2%) were small,
and FFM, on average, did not change (0  1.1 kg). The results suggest that
changes in body composition over the course of a season were concentrated
primarily in the early part of the season. Corresponding changes in estimated
body composition using anthropometric indicators have been described for
female collegiate distance runners and basketball players [49].

BODY MASS INDEX AND ATHLETES

Many programs monitor the heights and weights of athletes in the form of the
BMI. As noted, the BMI is reasonably well correlated with FM and % Fat in
large and heterogeneous samples, but correlations between the BMI and
FFM and FM are reasonably similar among youth [35]. A question of interest
is the association between the BMI and indicators of body composition at the
 BODY COMPOSITION IN ATHLETES
Table 9
Correlations between the body mass index and estimates of fatness and muscularity in female university athletes and nonathletes grouped by category of
body mass index
Athletes Nonathletes
Estimated Muscle Circumference Estimated Muscle Circumference
Arm Calf Arm Calf
BMI (kg/m2)
P
Category n 8 skf % Fat TþB T only MþL M only n % Fat T only M only
<18.5 14 0.68 0.64 0.10 0.08 0.10 0.05 19 0.61 0.27 0.39
18.5 < 25.0 350 0.39 0.42 0.57 0.50 0.44 0.37 79 0.44 0.34 0.15
25.0 23 0.70 0.82 0.88 0.83 0.66 0.52 13 0.88 0.36 0.50
P
8 skf, sum of the triceps, biceps, forearm, subscapular, suprailiac, midthigh, medial calf, and lateral calf skinfold thicknesses; % Fat, based on predicted Db using an equation
developed on elite female swimmers [20]; TþB, arm circumference corrected for the thickness of the triceps and biceps skinfolds; T only, arm circumference corrected for the
thickness of the triceps skinfold only; M þ L, calf circumference corrected for the thickness of the medial and lateral calf skinfolds; M only, calf circumference corrected for
the thickness of the medial calf skinfold only [30].
Data from Malina RM, Battista RA, Siegel SR. Anthropometry of adult athletes: concepts, methods and applications. In: Driskell JA, Wolinsky I, editors. Nutritional assessment
of athletes. Boca Raton (FL): CRC Press; 2002. p. 135–75.

63
64 MALINA

extremes of the BMI distribution. This was addressed in a large sample of fe-
male athletes using anthropometric indicators of body composition (Table 9).
Although numbers of athletes with low (<18.5 kg/m2) and high (25 kg/m2)
BMI are small, the trends in correlations suggest variable relationships. Among
athletes with a BMI in the normal range (18.5 < 25 kg/m2), correlations be-
tween estimates of fatness and muscularity are moderate and reasonably simi-
lar, ranging from 0.37 to 0.57. Correlations for % Fat and arm muscle are
similar for nonathlete female college students with a BMI in the normal range,
0.44 and 0.34, although that for estimated calf muscle is lower, 0.15. Among
those with a low BMI, correlations for fatness are higher and similar in athletes
and nonathletes, 0.61 to 0.68. Correlations for estimated limb musculature and
the BMI are low in athletes, 0.08 to 0.10, however, and higher in nonathletes,
0.27 and 0.39. Among those with a high BMI, correlations for fatness and es-
timated limb muscle circumferences are higher in athletes, 0.52 to 0.88, com-
pared with athletes in the other two BMI categories. A similar high
correlation, 0.88, is evident for % Fat in nonathletes with a high BMI, but cor-
relations for limb musculature in nonathletes are negative. Overall, the data
suggest variation in the association of the BMI and indirect estimates of fatness
and muscularity at the extremes of the BMI distribution and indicate a need for
further evaluation of the relationship between the BMI and body composition
in athletes and nonathletes.

References
[1] Wang ZM, Pierson RN, Heymsfield SB. The five-level model: a new approach to organizing
body composition research. Am J Clin Nutr 1992;56:19–28.
[2] Wang ZM, Shen W, Withers RT, et al. Multicomponent molecular-level models of body com-
position analysis. In: Heymsfield SB, Lohman TG, Wang ZM, editors. Human body compo-
sition. 2nd edition. Champaign (IL): Human Kinetics; 2005. p. 163–76.
[3] Roche AF, Heymsfield SB, Lohman TG, editors. Human body composition. Champaign (IL):
Human Kinetics; 1996.
[4] Heymsfield SB, Lohman TG, Wang ZM, et al, editors. Human body composition. 2nd
edition. Champaign (IL): Human Kinetics; 2005.
[5] Brozek J, Grande F, Anderson JT, et al. Densitometric analysis of body composition: revision
of some quantitative assumptions. Ann N Y Acad Sci 1963;110:113–40.
[6] Siri WE. The gross composition of the body. Adv Biol Med Phys 1956;4:239–80.
[7] Going SB. Hydrodensitometry and air displacement plethysmography. In: Heymsfield SB,
Lohman TG, Wang ZM, et al, editors. Human body composition. 2nd edition. Champaign
(IL): Human Kinetics; 2005. p. 17–33.
[8] Schoeller DA. Hydrometry. In: Heymsfield SB, Lohman TG, Wang ZM, et al, editors. Hu-
man body composition. 2nd edition. Champaign (IL): Human Kinetics; 2005. p.
35–49.
[9] Ellis KJ. Whole-body counting and neutron activation analysis. In: Heymsfield SB,
Lohman TG, Wang ZM, et al, editors. Human body composition. 2nd edition. Champaign
(IL): Human Kinetics; 2005. p. 51–62.
[10] Forbes GB. Body composition in adolescence. In: Falkner F, Tanner JM, editors. Human
growth, vol. 2. Postnatal growth, neurobiology. New York: Plenum; 1986. p. 119–45.
[11] Forbes GB. Human body composition: growth, aging, nutrition, and activity. New York:
Springer-Verlag; 1987.
BODY COMPOSITION IN ATHLETES 65

[12] Ellis KJ, Shypailo RJ, Abrams SA, et al. The reference child and adolescent models of body
composition: a contemporary comparison. Ann N Y Acad Sci 2000;904:374–82.
[13] Malina RM. Variation in body composition associated with sex and ethnicity. In:
Heymsfield SB, Lohman TG, Wang ZM, et al, editors. Human body composition. 2nd
edition. Champaign (IL): Human Kinetics; 2005. p. 271–98.
[14] Lohman TG, Chen Z. Dual energy X-ray absorptiometry. In: Heymsfield SB, Lohman TG,
Wang ZM, et al, editors. Human body composition. 2nd edition. Champaign (IL): Human
Kinetics; 2005. p. 63–77.
[15] Goran MI. Energy expenditure, body composition, and disease risk in children and adoles-
cents. Proc Nutr Soc 1997;56:195–209.
[16] National Institutes of Health. Bioelectrical impedance analysis in body composition mea-
surement. Bethesda: NIH Technology Assessment Statement; 1994. p. 1–35.
[17] Chumlea WC, Sun SS. Bioelectrical impedance analysis. In: Heymsfield SB, Lohman TG,
Wang ZM, et al, editors. Human body composition. 2nd edition. Champaign (IL): Human
Kinetics; 2005. p. 79–88.
[18] Malina RM. Quantification of fat, muscle and bone in man. Clin Orthop 1969;65:9–38.
[19] Malina RM. The measurement of body composition. In: Johnston FE, Susanne C, Roche AF,
editors. Human physical growth and maturation: methodologies and factors. New York:
Plenum; 1980. p. 35–59.
[20] Meleski BW, Shoup RF, Malina RM. Size, physique and body composition of competitive
female swimmers 11 through 20 years of age. Hum Biol 1982;54(3):609–25.
[21] Withers RT, Craig NP, Bourdon PC, et al. Relative body fat and anthropometric prediction
of body density of male athletes. Eur J Appl Physiol 1987;56:191–200.
[22] Withers RT, Whittingham NO, Norton KI, et al. Relative body fat and anthropometric pre-
diction of body density of female athletes. Eur J Appl Physiol 1987;56:169–80.
[23] Sinning WE. Body composition in athletes. In: Roche AF, Heymsfield SB, Lohman TG,
editors. Human body composition. Champaign (IL): Human Kinetics; 1996. p.
257–73.
[24] Jackson AS, Pollock ML. Generalized equations for predicting body density of men. Br J
Nutr 1978;40:497–504.
[25] Jackson AS, Pollock ML, Ward A. Generalized equations for predicting body density of
women. Med Sci Sports Exerc 1980;12:175–82.
[26] Chumlea WC, Guo SS, Zeller CM, et al. Total body water reference values and predictions
equations for adults. Kidney Int 2001;59:2250–8.
[27] Guo SS, Roche AF, Houtkooper LH. Fat-free mass in children and young adults from
bioelectrical impedance and anthropometry variables. Am J Clin Nutr 1989;50:
435–43.
[28] Sun SS, Chumlea WC, Heymsfield SB, et al. Development of bioelectrical impedance anal-
ysis prediction equations for body composition with the use of a multicomponent model for
use in epidemiological surveys. Am J Clin Nutr 2003;77:331–40.
[29] Warner ER, Fornetti WC, Jallo JJ, et al. A skinfold model to predict fat-free mass in female
athletes. J Athl Train 2004;39:259–62.
[30] Malina RM. Anthropometry. In: Maud PJ, Foster C, editors. Physiological assessment of
human fitness. Champaign (IL): Human Kinetics; 1995. p. 205–19.
[31] Sun SS, Chumlea WC. Statistical methods. In: Heymsfield SB, Lohman TG, Wang ZM, et al,
editors. Human body composition. 2nd edition. Champaign (IL): Human Kinetics; 2005.
p. 151–60.
[32] World Health Organization. Obesity: preventing and managing the global epidemic.
Report of a WHO Consultation on Obesity. Geneva: WHO; 1998.
[33] Cole TJ, Bellizzi MC, Flegal KM, et al. Establishing a standard definition for child over-
weight and obesity worldwide: international survey. BMJ 2000;320:1240–3.
[34] Hedley AA, Ogden CL, Johnson CL, et al. Prevalence of overweight and obesity among US
children, adolescents, and adults, 1999–2002. JAMA 2004;291:2847–50.
66 MALINA

[35] Malina RM, Bouchard C, Bar-Or O. Growth, maturation, and physical activity. 2nd edition.
Champaign (IL): Human Kinetics; 2004.
[36] Withers RT, LaForgia J, Heymsfield SB. Critical appraisal of the estimation of body com-
position via two-, three-, and four-compartment models. Am J Hum Biol 1999;
11:175–85.
[37] Moulton CR. Age and chemical development in mammals. J Biol Chem 1923;57:
79–97.
[38] Malina RM. Growth and maturation: normal variation and the effects of training.
In: Gisolfi CV, Lamb DR, editors. Perspectives in exercise science and sports med-
icine, vol. II. Youth, exercise, and sport. Indianapolis: Benchmark Press; 1989.
p. 223–65.
[39] Malina RM, Bouchard C, Beunen G. Human growth: selected aspects of current research on
well nourished children. Ann Rev Anthropol 1988;17:187–219.
[40] Guo SS, Chumlea WC, Roche AF, et al. Age- and maturity-related changes in body compo-
sition during adolescence into adulthood: the Fels Longitudinal Study. Int J Obes 1997;21:
1167–75.
[41] Parizkova J. Growth and growth velocity of lean body mass and fat in adolescent boys.
Pediatr Res 1976;10:647–50.
[42] Iuliano-Burns S, Mirwald RL, Bailey DA. The timing and magnitude of peak height velocity
and peak tissue velocities for early, average and late maturing boys and girls. Am J Hum Biol
2001;13:1–8.
[43] Krawczyk B, Sklad M, Majle B. Body components of male and female athletes representing
various sports. Biol Sport 1995;12:243–50.
[44] Schutte JE, Townsend EJ, Hugg J, et al. Density of lean body mass is greater in blacks than in
whites. J Appl Physiol 1984;56:1647–9.
[45] Ortiz O, Russell M, Daley TL, et al. Differences in skeletal muscle and bone mineral mass
between black and white females and their relevance to estimates of body composition.
Am J Clin Nutr 1992;55:8–13.
[46] Gasperino JA, Wang J, Pierson RN, et al. Age-related changes in musculoskeletal mass
between black and white women. Metabolism 1995;44:30–4.
[47] Gallagher D, Visser M, de Meersman RE, et al. Appendicular skeletal muscle mass: effects of
age, gender, and ethnicity. J Appl Physiol 1997;83:229–39.
[48] Hortobagyi T, Israel RG, Houmard JA, et al. Comparison of four methods to assess body
composition in black and white athletes. Int J Sport Nutr 1992;2:60–74.
[49] Malina RM, Battista RA, Siegel SR. Anthropometry of adult athletes: concepts, methods and
applications. In: Driskell JA, Wolinsky I, editors. Nutritional assessment of athletes. Boca
Raton (FL): CRC Press; 2002. p. 135–75.
[50] Malina RM. Crescita e maturazione di atleti bambini e adolescenti praticanti atletica leg-
gera [Growth and maturation of child and adolescent track and field athletes]. Atletica Studi
(Rome) 2006;(Suppl 1, 2):1–464.
[51] National Federation of State High School Assocations. 2006–07 wrestling rule changes.
Available at: www.nfhs.org. Accessed July 10, 2006.
[52] American College of Sports Medicine. Current comment: weight loss in wrestlers. Indianap-
olis: American College of Sports Medicine; 1998.
[53] National Collegiate Athletic Association. 2006 NCAA Wrestling Rules and Interpretations,
Appendix G. Assessment of body composition. Indianapolis: National Collegiate Athletic
Association; 2005. Available at: http://www.ncaa.org/library/rules/2006/2006_
wrestling_rules.pdf. Accessed July 10, 2006.
[54] Roemmich JN, Sinning WE. Weight loss and wrestling training: effects on nutrition,
growth, maturation, body composition and strength. J Appl Physiol 1997;82:
1751–9.
[55] Roemmich JN, Sinning WE. Weight loss and wrestling training: effects on growth-related
hormones. J Appl Physiol 1997;82:1760–4.
BODY COMPOSITION IN ATHLETES 67

[56] Oppliger RA, Utter AC, Scott JR, et al. NCAA rule change improves weight loss among
national championship wrestlers. Med Sci Sports Exerc 2006;38:963–70.
[57] Parizkova J. Longitudinal study of the relationship between body composition and anthropo-
metric characteristics in boys during growth and development. Glasnik Antropoloskog
Drustva Jugoslavije 1970;7:33–8.
[58] Parizkova J. Body fat and physical fitness. The Hague: Martinus Nijhoff; 1977.
[59] Von Dobeln W, Eriksson BO. Physical training, maximal oxygen uptake and dimensions
of the oxygen transporting and metabolizing organs in boys 11–13 years of age. Acta
Paediatr Scand 1972;61:653–60.
[60] Morris FL, Naughton GA, Gibbs JL, et al. Prospective ten-month exercise intervention in pre-
menarcheal girls: positive effects on bone and lean mass. J Bone Min Res 1997;12:
1453–62.
[61] Wilmore JH. Appetite and body composition consequent to physical activity. Res Q
1983;54:415–25.
[62] Siders WA, Bolunchuk WW, Lukaski HC. Effects of participation in a collegiate sport season
on body composition. J Sports Med Phys Fit 1991;31:571–6.
[63] Bolonchuk WW, Lukaski HC, Siders WA. The structural, functional, and nutritional adap-
tation of college basketball players over a season. J Sports Med Phys Fit 1991;31:
165–72.
[64] Meleski BW, Malina RM. Changes in body composition and physique of elite university-
level female swimmers during a competitive season. J Sports Sci 1985;3:33–40.
[65] Fleck SJ. Body composition of elite American athletes. Am J Sports Med 1983;11:
398–403.
[66] Novak LP, Woodward WA, Bestit C, et al. Maximal aerobic power, body composition and
anthropometry of Olympic runners and road cyclists. In: Jungmann H, editor. Sportwissen-
schaftliche Untersuchungen während der XX. Olympischen Spiele, München 1972.
Hamburg: Karl Demeter; 1976. p. 79–90.
[67] Perez HR. The effects of competitive road-racing on the body composition, pulmonary func-
tion, and cardiovascular system of sport cyclists. J Sports Med Phys Fit 1981;21:165–72.
[68] Forbes GB. Toward a new dimension in human growth. Pediatrics 1965;36:825–35.
[69] Novak LP, Hyatt RE, Alexander JF. Body composition and physiologic function of athletes.
JAMA 1968;205:764–70.
[70] Wickkiser JD, Kelly JM. The body composition of a college football team. Med Sci Sports
1975;7:199–202.
[71] Wilmore JH, Parr RB, Haskell WL, et al. Football pros’ strengths—and CV weaknesses—
charted. Phys Sportsmed 1976;4(Oct.):45–54.
[72] Novak LP, Bestit C, Mellerowicz H, et al. Maximal oxygen consumption, body composition
and anthropometry of selected Olympic male athletes. In: Jungmann H, editor. Sportwis-
senschaftliche Untersuchungen während der XX. Olympischen Spiele, München 1972.
Hamburg: Karl Demeter; 1976. p. 57–68.
[73] Raven PB, Gettman LR, Pollock ML, et al. A physiological evaluation of professional soccer
players. Br J Sports Med 1976;10:209–16.
[74] Farmosi I, Apor P, Mecseki S, et al. Body composition of notable soccer players. Hung Rev
Sports Med 1984;25:91–6.
[75] Agre JC, Casal DC, Leon AS, et al. Professional ice hockey players: physiologic, anthropo-
metric and musculoskeletal characteristics. Arch Phys Med Rehab 1988;69:188–92.
[76] Sprynarova S, Parizkova J. Functional capacity and body composition in top weight lifters,
swimmers, runners and skiers. Int Z Angew Physiol 1971;29:184–94.
[77] Sinning WE, Cunningham LN, Racaniello AP, et al. Body composition and somatotype of
male and female Nordic skiers. Res Q 1977;48:741–9.
[78] Pollock ML, Pels AE, Foster C, et al. Comparison of male and female speed skating can-
didates. In: Landers DM, editor. Sports and elite performance. Champaign (IL): Human
Kinetics; 1986. p. 143–52.
68 MALINA

[79] Sinning WE, Wilson JR. Validity of generalized equations for body composition analysis in
women athletes. Res Q Exerc Sport 1984;55:155–60.
[80] Fornetti WC, Pivarnik JM, Foley JM, et al. Reliability and validity of body composition mea-
sures in female athletes. J Appl Physiol 1999;87:1114–22.
[81] Novak LP, Woodward WA, Bestit C, et al. Working capacity (WC170), body composition,
and anthropometry of Olympic female athletes. In: Jungmann H, editor. Sportwissenschaft-
liche Untersuchungen während der XX. Olympischen Spiele, München 1972. Hamburg:
Karl Demeter; 1976. p. 69–78.
[82] Sinning WE. Anthropometric estimation of body density, fat, and lean body weight in
women gymnasts. Med Sci Sports 1978;10:243–9.
[83] Colquhoun D, Chad KE. Physiological characteristics of Australian female soccer players
after a competitive season. Aust J Sci Med Sport 1986;18:9–12.
[84] Fleck SJ, Case S, Puhl J, et al. Physical and physiological characteristics of elite women vol-
leyball players. Can J Appl Sport Sci 1985;10:122–6.
[85] Piechaczek H. Oznaczanie t1uszczu cia1a metodami densytometryczna˛ i antropometrycz-
na (estimation of body fat with densitometric and anthropometric methods). Mater Pr Antro-
pol 1975;89:3–48.