classified as having a subclinical eating disorder
termed anorexia athletica (Sundgot-Borgen
1994a).
It is assumed that many cases of anorexia
nervosa and bulimia nervosa begin as subclinical
variants of these disorders. Early identification
and treatment may prevent development of the
full disorder (Bassoe 1990). Subclinical cases are
more prevalent than those meeting the formal
diagnostic criteria for anorexia nervosa and
bulimia nervosa (Sundgot-Borgen 1994a).
Prevalence
Data on the prevalence of eating disorders in ath-
letic populations are limited and equivocal.
Most studies have looked at symptoms of eating
disorders such as preoccupation with food and
weight, disturbed body image, or the use of
pathogenic weight control methods.
Female athletes
Estimates of the prevalence of the symptoms
of eating disorders and true eating disorders
among female athletes range from less than 1% to
as high as 75% (Gadpalle et al. 1987; Burckes-
Miller & Black 1988; Warrenet al. 1990; Sundgot-
Borgen 1994a).
Methodological weaknesses such as small
sample size, lack of definition of the competitive
level or type of sport(s) and lack of definition of
the data collection method used characterize
most of the studies attempting to study the
prevalence of eating disorders (Sundgot-Borgen
1994b).
Only one study has used clinical evaluation
and the DSM criteria applied across athletes and
controls (Sundgot-Borgen 1994a). The preva-
lence of anorexia nervosa (1.3%) seems to
be within the same range as that reported in
non-athletes (Andersen 1990), whereas bulimia
nervosa (8.2%) and subclinical eating disoders
(8%) seem to be more prevalent among female
athletes than non-athletes (Sundgot-Borgen
1994a). The prevalence of eating disorders was
significantly higher among athletes competing
in aesthetic and weight-dependent sports than
among other sport groups where leanness is con-
sidered less important (Fig. 39.1).Male athletes
Results from existing studies on male athletes
indicate that the frequency of eating distur-
bances and pathological dieting practices varies
from none to 57%, depending on the definition
used and the population studied (Dummer et al.eating disorders in athletes 511
50403020100
G1 G2 G3 G4 G5 G6
Sports groups Non-athletesSubjects with eating disorder (%)
*Fig. 39.1Prevalence of eating
disorders in female elite athletes
representing: G1, technical sports
(n=98); G2, endurance sports
(n=119); G3, aesthetic sports
(n=64); G4, weight-dependent
sports (n=41); G5, ball games
(n=183); G6, power sports
(n=17); and non-athletes
(n=522). The data are shown as
mean and 95% confidence
intervals. The shaded area is the
95% confidence interval for the
control group of non-athletes.