Prevalence of Female Athlete Triad Characteristics in a Club ...

BRIEF REPORT
Prevalence of Female Athlete Triad Characteristics in a Club
Triathlon Team
Anne Z. Hoch, DO, John E. Stavrakos, MD, Jane E. Schimke, AAS
681
ABSTRACT. Hoch AZ, Stavrakos JE, Schimke JE. Prevalence
of female athlete triad characteristics in a club triathlon
team. Arch Phys Med Rehabil 2007;88:681-2.
Objective: To determine the prevalence of the female athlete
triad in club triathletes.
Design: Cross-sectional.
Setting: Academic medical center in the midwestern United
States.
Participants: Fifteen women (mean age, 356y).
Interventions: Not applicable.
Main Outcome Measures: Disordered eating and menstrual
status were determined by questionnaires. Energy status was determined
by a 3-day food record, resting energy expenditure, and exercise
energy output. Bone mineral density (BMD) was measured in the
total left hip and lumbar spine (L2-4) by dual-energy x-ray
absorptiometry.
Results: Sixty percent of the triathletes were found to be in
calorie deficit, 53% had a carbohydrate deficit, 47% had a fat
deficit, 40% had a protein deficit, and 33% had a calcium deficit.
Forty percent of triathletes reported a history of amenorrhea.
BMD was normal in the lumbar spine (L2-4) (1.30.1g/cm 2 ) and
total left hip (1.10.1g/cm 2 ).
Conclusions: Triathletes are at risk for components of the
female athlete triad. Continued efforts need to be directed at
prevention through education of athletes, coaches, parents, and
health care professionals.
Key Words: Amenorrhea; Athletics; Eating disorders; Female;
Rehabilitation.
© 2007 by the American Congress of Rehabilitation Medicine
and the American Academy of Physical Medicine and
Rehabilitation
IN A WORLD INCREASINGLY devoted to participation in
professional and recreational athletic events, the triathlon has
gained significant attention within the past 2 decades. A triathlon
is defined as consecutive swimming, biking, and running events of
varying distances. The diversity of training and the development
of a balanced degree of fitness have contributed to the success of
this athletic endeavor.
The female athlete triad (triad) is an interrelated condition of
disordered eating, amenorrhea, and osteoporosis. It appears to be
more common in endurance and aesthetic sports.
The purpose of this study was to determine the prevalence of
triad characteristics in a recreational women’s triathlon team.
From the Women’s Sports Medicine Program/Sports Medicine Center, Departments
of Orthopaedic Surgery (Hoch, Stavrakos, Schimke) and Physical Medicine
and Rehabilitation (Hoch, Stavrakos), Medical College of Wisconsin, Milwaukee, WI.
No commercial party having a direct financial interest in the results of the research
supporting this article has or will confer a benefit upon the authors or upon any
organization with which the authors are associated.
Reprint requests to Anne Z. Hoch, DO, Women’s Sports Medicine Program/Sports
Medicine Center, Depts of Orthopaedic Surgery/Physical Medicine and Rehabilitation,
9200 W Wisconsin Ave, Medical College of Wisconsin, Milwaukee, WI 53226,
e-mail: azeni@mcw.edu.
0003-9993/07/8805-11346$32.00/0
doi:10.1016/j.apmr.2007.02.035
METHODS
Fifteen women from the same triathlon team volunteered to
participate in this study and signed an informed consent in
accordance with our institutional review board. All of the
triathletes competed at a recreational level except for 2 who
were professionals. Subjects trained on an average of 5 days
per week for at least 90 minutes. Subjects reported starting
athletics at a mean age of 12.6 years. None had a history of
anorexia or bulimia. Subjects completed menstrual and health
questionnaires, which were reviewed by the primary investigator
(AZH). Eating habits and attitudes were assessed by the
Eating Attitude Test (EAT-26), which has been validated for
anorexia. 1 Each subject completed a body image questionnaire.
We calculated energy status on each subject from a prospective
3-day food diary 2 using a Nutrient Analysis program. a
Resting energy expenditure (REE) was calculated by the Harris-
Benedict equation and multiplied by an activity factor of 1.1,
which was used to estimate energy expended from daily activities
of living. 3 Energy burned through exercise was estimated
using known calculations for time of exercise for running,
swimming, and biking. These calculations were based on factors
associated with the participants’ weight, age, sex, and
intensity of exercise. 4 The total calories expended from exercise
was then added to the adjusted energy expenditure (REE
1.1). This value of total energy expended was compared with
the subjects’ caloric intake to determine if an energy deficit was
present. The sports dietitian was able to calculate the participants’
daily carbohydrate, protein, and fat requirements and
compare these values with the amount of macronutrients consumed.
Diet records were recorded during the training season
and consisted of 2 weekdays and 1 weekend day. Finally,
calcium intake was recorded based on the 3-day food diary and
compared with daily requirements for age. 5
We measured bone mineral density (BMD) (in g/cm 2 ) of the lumbar
spine (L2-4) and total left hip by dual energy x-ray absorptiometry
(DXA) b to assess for evidence of reduced BMD for age.
RESULTS
Mean age standard deviation (SD) of the subjects was
356 years. Nutritional analysis revealed that 60% of the
triathletes were in a calorie deficit. Fifty-three percent had a
carbohydrate deficit, 47% had a fat deficit, 40% had a protein
deficit, and 33% had a calcium deficit. The average calcium
deficit was 307283mg, average calorie deficit of the affected
group was 245187kcal, and the average fat deficit
was 1511g. Average carbohydrate and protein deficits were
7030g and 157g, respectively. Calorie deficit is defined
as calorie intake not commensurate with the athlete’s caloric
expenditure. Calcium deficit was defined as an average daily
intake of less than 1300mg/d. 5 The average score on the
EAT-26 was 4.34.4 (15 is abnormal). 1 This self-reported
questionnaire consists of 26 questions and is scored on a 0-to-6
Likert scale. A score of greater than 30 on the EAT-26 represents
a high likelihood of anorexia and a score between 15 and
30 represents a subclinical group with disordered eating habits
and anorectic attitudes. The body image silhouette was
3.20.2. Subjects were asked to circle the body type they
Arch Phys Med Rehabil Vol 88, May 2007

682 PREVALENCE OF TRIAD IN TRIATHLETES, Hoch
considered to be the “ideal” triathlete from a validated 9-body
silhouette scale, 6 which ranges from an extremely thin or
nearly anorexic figure at position 1 to a very overweight figure
at position 9. A calorie deficit (200kcal) was considered
consistent with disordered eating for the purposes of this study.
Forty percent of the participants admitted to a history of
primary or secondary amenorrhea. Average length of amenorrhea
was 142 months. None were currently amenorrheic.
DXA revealed a mean BMD SD of 1.30.1g/cm 2 in the
lumbar spine (L2-4), which corresponds to a T score of 0.7 and
z score of 0.8. The average BMD of the total hip was 1.10.1g/
cm 2 , which corresponds to a T score of 0.7 and z score of 0.8.
Both T score and z score values correspond to values that are
within normative limits based on the World Health Organization
(WHO) 7 classification system and criteria published by the
International Society for Clinical Densitometry (ISCD). 8 In
1994, WHO established criteria for making the diagnosis of
osteoporosis, as well as determining levels that predict higher
chances of fractures for postmenopausal white women. These
criteria are based on comparing BMD in a particular patient to
the average 20-year-old woman. BMD values, referred to as T
score, that fall well below the average for the 20-year-old
(stated statistically as 2.5 SDs below the average) are diagnosed
as “osteoporotic.” If a patient has a BMD T-score value
between 1.0 and 2.5 SDs below average, they are considered
to be “osteopenic.” The z score is the number of SDs
below average for a person of the same age, sex, and race.
ISCD recommends using z scores for premenopausal adolescents
and children. If a z score is less than 2.0 using a
pediatric database of age-matched controls, then ISCD recommends
using the term “low bone mineral density for age.” 8
Subjects with a history of amenorrhea had the lowest BMD.
This was not statically significant given the small sample size,
however.
Overall, 60% of the subjects had at least 1 component of the
triad, with 27% of these having 2 components of the triad (disordered
eating and history of amenorrhea). Finally, 53% of the
subjects were not able to identify all 3 components of the triad and
60% were not aware of daily calcium requirements for age.
DISCUSSION
This was a well-educated group of athletes that showed
relatively normal body image. They showed several triad characteristics,
however. Of the 3 components of the triad, disordered
eating, specifically caloric restriction, was the most prevalent
followed by a history of amenorrhea. None of the subjects
had an abnormal BMD value. Over half of the team showed 1
or more triad characteristics.
The term disordered eating encompasses a spectrum of abnormal
behaviors that may range from a mild preoccupation with
food and exercise to a diagnosis of anorexia or bulimia according
to the Diagnostic and Statistical Manual of Mental Disorders,
Fourth Edition. 9 These disorders often stem from internal or
external pressures to be thin, or they may be the result of the
misconception that thinness improves athletic performance. It is
estimated that anywhere from 15% to 65% of female athletes
suffer from some form of disordered eating pattern. 10 In this study,
60% of the participants were found to be in a calorie deficit.
Athletic associated amenorrhea is a diagnosis of exclusion. It
was formerly believed that the stress of exercise and low body
fat were the causes of exercise associated amenorrhea. Recent
studies by Loucks 11 have found that diminished energy intake,
rather than exercise, stress, or body fat depletion, was the
regulating factor in the cessation of menses in active women.
BMD values in premenopausal women can be calculated by T
score according to WHO. 7 A T score of 1.0 or less is classified
as osteopenia and 2.5 or less is considered osteoporosis. It
should be noted that these labels may not be entirely applicable to
this subject population, because the WHO osteoporosis criteria
were designed for postmenopausal women in whom the mechanism
of low BMD is the result of premature bone loss, and the T
score was designed to predict fracture risk. ISCD 8 recommends
using the term “low BMD” in premenopausal women who do not
have risk factors and have a z score less than 2.0. ISCD reserves
the term osteoporosis for premenopausal women with secondary
risk factors. Fortunately, none of the athletes in this
study showed evidence of abnormal BMD in the hip or lumbar
spine. In addition, BMD was obtained at 1 point in time in this
study. Therefore, we were unable to determine who had a loss,
gain, or no change over time.
CONCLUSIONS
Women triathletes are at risk for components of the female
athlete triad, especially disordered eating. The participants in
this study were college-educated women living in a major
metropolitan area with normal body image, yet over half of
these athletes were unaware of the triad and its potential
dangers. This underscores the need for continued effort directed
at prevention of the triad through education of athletes,
coaches, parents, and other health professionals and for comprehensive
clinical programs to treat the triad.
References
1. Garner DM, Olmsted MP, Bohr Y, Garfinkel PE. The eating
attitudes test: psychometric features and clinical correlates. Psychol
Med 1982;12:871-8.
2. McKeown NM, Day NE, Welch AA, et al. Use of biological
markers to validate self-reported dietary intake in a random sample
of the European Prospective Investigation into Cancer United
Kingdom Norfolk cohort. Am J Clin Nutr 2001;74:188-96.
3. Harris JA, Benedict FG. A biometric study of basal metabolism in
man. Washington (DC): Carnegie Institute of Washington; 1919.
Publication No. 279.
4. Johnson RK. Energy. In: Mahan KL, Escott-Stump S, editors.
Krause’s food, nutrition and diet therapy. 10th ed. Philadelphia:
WB Saunders; 2000. p 19-30.
5. Osteoporosis prevention, diagnosis, and therapy. NIH Consens
Statement 2000;17:1-45.
6. Sorensen TI, Stunkard AJ. Does obesity run in families because of
genes? An adoption study using silhouettes as a measure of
obesity. Acta Psychiatr Scand Suppl 1993;370:67-72.
7. World Health Organization. Assessment of fracture risk and its
application to screening for postmenopausal osteoporosis. Geneva:
WHO; 1994. Technical Report Series 843.
8. Writing Group for the ISCD Position Development Conference.
Diagnosis of osteoporosis in men, premenopausal women, and
children. J Clin Densitom 2004;7:17-26.
9. American Psychiatric Association. Eating disorders. In: First MB,
editor. Diagnostic & statistical manual of mental disorders: DSM-
IV. Washington (DC): APA; 1994. p 539-50.
10. Otis CL, Drinkwater B, Johnson M, Loucks A, Wilmore J. American
College of Sports Medicine position stand. The Female
Athlete Triad. Med Sci Sports Exerc 1997;29:1669-71.
11. Loucks AB. Energy availability, not body fatness, regulates reproductive
function in women. Exerc Sport Sci Rev 2003;31:144-8.
Suppliers
a. Nutritionist Pro, 2nd ed; Axxya Systems, 4035 Willowbend Blvd,
Ste 410, Houston, TX 77025.
b. Lunar Prodigy densitometer, version 2.15; GE Healthcare Lunar,
726 Heartland Trl, Madison, WI 53713-1915.
Arch Phys Med Rehabil Vol 88, May 2007