Pediatrics International (2000) 42, 1–7
Eating disorders in adolescent girls*
S JEAN EMANS
Division of Adolescent/Young Adult Medicine, Children’s Hospital and Harvard Medical School,
Boston, Massachusetts, USA
Abstract Eating disorders are common in contemporary society. New information is emerging on the pathogenesis of
anorexia nervosa and bulimia nervosa and includes psychologic, biologic, family, environmental, genetic and
social factors. The physician providing care to adolescents is challenged to carry out a careful evaluation and
monitor the patient for complications, especially loss of bone mass. Treatment requires a multidisciplinary
Key words adolescent health, anorexia nervosa, bulimia nervosa, eating disorder, osteoporosis.
Background and epidemiology than the 85th percentile using National Health and Nutrition
Examination Survey (NHANES) II data.
Eating disorders are common among adolescent girls in the In both American and Japanese societies, an obsession
United States and in industrialized nations. A number of with thinness permeates the media with images of thin,
studies have suggested an increase in the incidence of successful models, diets that purport to bring rapid results
anorexia nervosa and bulimia over the past 50 years in the and numerous advertisements for low-caloric, low-fat foods.
USA, although there may have been a recent decrease in Diet pills and weight-loss programs are a billion dollar
binge/purging behaviors among college students.1–5 The industry. Dieting behaviors are manifest even at ages as
actual prevalence of anorexia nervosa (AN) and bulimia early as 8–10 years and up to 20% of adolescent girls score
nervosa (BN) has been difﬁcult to measure in the USA in the abnormal range on standardized tests of eating
because of variable case deﬁnition and reporting and, attitudes.1,9 The Commonwealth Fund Survey of the Health
importantly, case ascertainment in an illness characterized of Adolescent Girls,10 published in 1997, documented that as
by denial and secrecy. The onset of AN typically occurs girls progress through adolescence they diet increasingly and
during adolescence or young adulthood, although exercise less. In grades 5–8, 39% of girls gave a history of
manifestations may be apparent later in life or in childhood. dieting and by 9th to 12th grades, 58% had dieted. Among
It is estimated that 0.48% of 15–19-year-olds have AN, the same girls, 81% of girls in grades 5–8 had exercised
1–5% of adolescents have BN and 3–5% of 15–30-year-old three or more times per week, compared with only 67% of
women have eating disorders not other speciﬁed (ED 9th to 12th grade girls. In the 1997 Youth Risk Behavior
NOS).1,3,5–7 In comparison, in Japan, AN has been reported Survey,11 which is administered to 9th to 12th grade USA
in 0.08% of 12–14-year-olds and 0.1% of 15–18-year-olds students, 33.5% of girls thought they were overweight,
and bulimia in 2.9% of college students.8 Anorexia nervosa nearly 60% were attempting to lose weight, 45.7% were
represents the third most common chronic disease in the dieting and 65.4% were exercising to lose weight. All of
USA and 95% of cases of AN occur in females. Although these behaviors were much more common in girls than in
the focus of the present paper is on eating disorders, obesity boys. Similar results have been noted in a review of the
is much more prevalent in the USA than both BN and AN. Japanese literature.8 A survey of 1799 Japanese female high
Eleven percent of adolescents have a body mass index school students found that 85% who were a normal weight
(BMI) in excess of the 95th percentile and 22% are greater wanted to be thinner and 45% who were 10–20%
underweight wanted to be thinner.8 A survey of 712 junior
high school students found that 48% had dieted to lose
Correspondence: S Jean Emans, Chief, Division of Adolescent weight and 75% were concerned about their weight.12 Mukai
and Young Adult Medicine, The Children’s Hospital Medical and colleagues8 studied 197 11th grade female Japanese
Center, 300 Longwood Avenue, Boston, MA 02115, USA. students, administering a 26-item eating attitudes test (EAT).
*Presented in part at the 102nd Meeting of the Japan Pediatric The mean score was 16.99 and 35% scored over 20.
Society, 24 April 1999. Interestingly, actual weight status was not associated with
Received 10 May 1999. higher scores on the EAT but the feeling of being overweight
2 SJ Emans
was associated. Teens’ reports of their mothers’ views of foods in patients with AN and controls, noting different
weight issues were particularly related to high EAT scores. psychologic and physiologic responses to high caloric foods
Having her mother view her as fat, being encouraged to lose along with decreased activity in a brain region mediating
weight by her mother and the frequency of talking with her oral sensation. Other studies have found magnetic resonance
mother about dieting accounted for 14% of the variance in imaging (MRI) changes in girls with AN, including
EAT scores for girls. decreased white and gray matter and increased cerebrospinal
Many studies to date have looked primarily at clinical ﬂuid (CSF) volume.18,19
samples of girls with AN and BN and have not examined The family history is often striking in girls with
population-based samples. Assumptions about the socio- disordered eating and points to both environmental and
demographics of girls with anorexia have tended to focus on genetic factors. Not infrequently, the clinician may ﬁnd a
white, afﬂuent teens. More recent data suggest that a more family history of eating disorders, depression, alcoholism,
diverse population is affected by eating disorders.1,13,14 In the substance abuse and other mental illness during the
cross-sectional study of the National Longitudinal Study of evaluation. There is no one family type that is associated with
Adolescent Health (1994–95), Forman and Goodman13 the development of eating disorders, although in clinical
deﬁned a group of girls (‘AN-like syndrome’) who had samples, families often have high expectations, poor
BMI<85% for age, were trying to lose weight and reported communication styles and marital tension.1,3,5 Genetic studies
that their weight was about right to very overweight. The have found variable rates of eating disorders within families
overall prevalence of this syndrome was 0.2% and the and twins.20 It appears that young women with affected ﬁrst-
sample was more diverse in racial, ethnic and income degree relatives have a 5–10-fold increased risk of
characteristics than previously noted for clinical samples. developing an eating disorder.21 In a study of 24 probands
with AN and 44 controls, Gershon et al. found a rate of AN
and BN among ﬁrst-degree relatives of 2% and 4.4%,
Pathogenesis respectively, compared with 0% and 1.3% among control
probands.22 In a small study of twins, Holland and
The pathogenesis of eating disorders remains elusive, but is colleagues23 reported that the concordance rate for AN was
clearly multifactorial and includes psychologic, biologic, 56% for monozygotic twins and 7% for dizygotic twins. In
family, environmental, genetic and social factors. For example, the Virginia Twin Registry study, the concordance for BN
an adolescent with low esteem, biologic predisposition or was 22.9% in monozygotic twins and 8.7% in dizygotic
family trauma may develop an eating disorder to establish a twins.24 The same Registry study did not ﬁnd as high a
sense of control and stability.5 Similarly, an adolescent concordance for AN as in the Holland study, but estimated
participating in an athletic or artistic endeavor, such as ballet, that the risk was increased 5–50-fold for a twin with AN.20,25
track and gymnastics, may experience praise and success
accompanying weight loss.
Young women with eating disorders have high rates of Deﬁnitions of eating disorders
affective disorders, depression, anxiety disorders, obsessive
compulsive disorders, personality disorders and substance The DSM-IV deﬁnitions of AN, BN and ED NOS are noted
use (the latter particularly associated with BN). Studies have in Table 1.26 For girls with AN, there are two subtypes: (i)
yielded conﬂicting evidence on the association of sexual those who use restricting alone to reduce weight; and (ii)
abuse and the development of eating disorders,3,5,15 but a those who use bingeing/purging to control their weight. Of
history of non-consensual sexual contact should always be importance in growing adolescents, girls going through
obtained as part of the evaluation because it may be a puberty may fail to make normal weight gains and gradually
predisposing or trigger event for some adolescents with fall below the 85th percentile of the expected weight for
eating disorders. height or lose the equivalent of 15% of expected bodyweight
Biologic issues involved in the pathogenesis of eating for height. Linear growth failure may also result from
disorders have been highlighted recently with research on inadequate caloric intake at a critical stage of puberty and, at
neurotransmitters (norepinephrine and serotonin), changes in the same time, the small amounts of gonadal hormones
leptin levels and alterations in cerebral blood ﬂow. Studies secreted may still advance bone age, resulting in loss of ﬁnal
have suggested that leptin levels are low in girls with AN adult stature. Standards of BMI used in adults are not
and amenorrhea and that levels return towards normal in applicable to early adolescents and growth charts need to be
girls who have both weight recovery and restoration of followed longitudinally, in addition to BMI. For example,
menses.16 Using positron emission tomography (PET) scans, the 85% BMI for 12–13 years old is 16, for 14–16 years old
Gordon and colleagues17 have examined the effect of food is 17 and for 17–18 years old is 18.13 Weight targets and
challenges with neutral foods and provocative high caloric ranges are established using ideal bodyweight (IBW) tables
Eating disorders in adolescent girls 3
Table 1 Criteria for diagnosing eating disorders in adolescents (adapted from DSM IV (1994))26
A. Refusal to maintain bodyweight over a minimal normal weight for age and height (bodyweight less than 85% of that expected) or
failure to make expected weight gain during period of growth.
B. Intense fear of gaining weight or becoming fat, even though underweight.
C. Disturbance in the way in which one’s bodyweight or shape is experienced or denial of the seriousness of the current low bodyweight.
D. In postmenarcheal females, amenorrhea of at least three consecutive menstrual cycles.
A. Recurrent episodes of binge eating. An episode of binge eating is characterized by eating, in a discrete period of time, an amount of
food that is deﬁnitely larger than most people would eat during a similar period of time AND a sense of lack of control over eating
during the episode.
B. Recurrent inappropriate compensatory behavior in order to prevent weight gain, such as self-induced vomiting; misuse of laxatives,
diuretics or enemas; fasting; or excessive exercise.
C. Binge eating and compensatory behaviors occur, on average, at least twice a week for 3 months.
D. Self-evaluation is unduly inﬂuenced by body shape and weight.
E. The disturbance does not occur exclusively during episodes of anorexia nervosa.
Eating Disorder Not Otherwise Speciﬁed
Disorders of eating that do not meet the criteria for a speciﬁc Eating Disorder, for example:
The criteria for anorexia nervosa are met but the individual has regular menses.
The criteria for anorexia nervosa are met except that, despite signiﬁcant weight loss, the individual’s current weight is in the normal range.
The criteria for bulimia nervosa are met but the binge eating and compensatory mechanisms occur at a frequency less than twice a week
or for a duration of less than 3 months.
The regular use of inappropriate compensatory behavior by an individual of normal bodyweight after eating small amounts of food.
Repeatedly chewing and spitting out, but not swallowing, large amounts of food.
for age (National Health Statistics standards), estimated fat that initiated weight loss, death or illness in a family member,
percentages (using calipers or Frisch tables27) and BMI. A a change in family constellation and substance abuse.
number of investigators have pointed out mean thresholds A careful history of pubertal progression or delay and menses
associated with the re-establishment of menses in girls with is essential, because girls with eating disorders may have
AN, based on estimates of percentages of body fat using delayed puberty, impaired growth, delayed menarche,
height and weight measurements,27 percentage of ideal amenorrhea or oligomenorrhea. The social history should
bodyweight28 and BMI.29 However, girls may remain include a complete review of risk issues focusing on the
amenorrheic despite attaining the necessary weight for home, education, activities, depression and drugs, sexuality
height, because of stress and ongoing preoccupation with and suicide (HEADSS) survey, a notation of the family
food, dieting and weight. Further studies are needed to better structure and an observation and discussion of parental
understand the interrelationship between height, weight, dynamics. A family history should include information about
pubertal development, expected growth trajectory and eating disorders within the family, obesity, depression,
menstrual function in girls with eating disorders. alcoholism, substance use or other mental illness.
A review of systems should include questions about
abdominal pain, bloating or constipation, esophagitis assoc-
iated with bulimia, hair loss or texture change associated
Patient evaluation with anorexia nervosa, cold intolerance, fatigue, weakness,
fainting, substance use and depression. The level of partici-
The evaluation of the adolescent girl in whom an eating pation and hours per day of physical exercise and a history
disorder is suspected includes a careful patient and family of stress fractures should be obtained. There is a ﬁne line
history, physical examination, laboratory tests and mental between classic AN and the so-called ‘female athlete triad’:
health and nutritional assessments. The history from the osteoporosis, amenorrhea and eating disorders.30,31 Girls with
patient and family should include a history of weight this triad have an increased risk of stress fracture, not only
changes, self perceptions of weight and desired weight, a because of osteopenia, but also because of training errors
history of bingeing and out of control eating and purging and and an inability to stop exercising and rest with the onset of
use of laxatives, ipecac and diet pills. Triggers for weight loss pain. Girls with AN may also have poor concentration,
include teasing at school or comments about weight that irritability and moodiness. Girls with BN may have cavities
occurred at home or in the physician’s ofﬁce, a mild illness and acid erosion of the anterior teeth.
4 SJ Emans
The dietary history should include a 24 h recall of intake, tests, a urine pregnancy test if the patient is sexually active
with the clinician realizing that the amounts may be and amenorrheic and an electrocardiogram (EKG) if the
inaccurate because teens with AN may over-report intake. patient is bradycardic or will be using medications with
Triggers to bingeing, such as stress and conﬂict, are cardiac effects. The QT interval may be prolonged in these
important to ascertain and address. The calcium intake patients even without electrolyte abnormalities, although the
should be estimated by determining the number of servings exact prevalence of this abnormality reported in studies has
of dairy per day or the use of calcium supplements, in order been variable.1,3,5 Pancytopenia or isolated mild neutropenia,
to plan interventions to assure adequate calcium and vitamin anemia or thrombocytopenia may be noted. Hypokalemia is
D intake because of the increased risk of osteoporosis in common in girls with vomiting and can be life-threatening.
AN. It is also important to ask about caffeine-containing Hypophosphatemia may become apparent, particularly with
beverages, because these may dull the appetite and also refeeding during hospitalization. Thyroid tests are often
increase the heart rate at the time of evaluation. typical of ‘sick euthyroid’ syndrome. Cortisol levels are
Many programs also administer standardized assessment frequently elevated in girls with AN, although not associated
instruments, such as the Beck Depression Index, a 21 item with any clinical stigmata of hypercortisolism (with the
scale (scored as 0–9, normal; 10–15, mild; 16–19, mild to exception of osteopenia). Follicle stimulating hormone
moderate; 20–29, moderate to severe; and >30, severe), and (FSH) and prolactin levels are obtained in girls with
the 26 or 40 item EAT. A score above 30 on the 40 item test persistent or unexplained amenorrhea before instituting
indicates a signiﬁcant chance of being an eating disordered hormonal replacement therapy. Central nervous system
patient. imaging should be considered in a patient with an early and
The physical examination should include vital signs to unusual presentation of an eating disorder, growth failure or
assess bradycardia, hypotension, orthostasis and hypo- neurologic signs or symptoms. Other tests are ordered
thermia. The weight and height should be recorded in a depending on the clinical assessment.
gown, after urination. The urine speciﬁc gravity should be
measured, because some patients will water load or hide
weights in their clothing to falsely elevate their weight. Management
During the skin examination, the clinician should look for
lanugo, dry skin, a yellow tint to skin associated with Treatment of a girl with AN or bulimia requires a
hypercarotenemia, hair changes and calluses on the dorsum multidisciplinary team approach. The medical provider
of the hand (indicative of bulimic behaviors). In a girl with typically assumes the role as manager of the care team,
AN, the abdomen is often scaffoid with palpable stool, the performing weekly vital signs and weight checks and
breasts atrophied, the vaginal mucosa hypoestrogenic and coordinating the overall communication with the family. The
the extremities cool and wasted. The cardiac examination nutritionist works with the adolescent and family around
should assess for bradycardia, arrhythmias and mitral valve meal planning and assessing and making recommendations
prolapse. A girl with AN is often in denial or irritable or has for caloric requirements and calcium intake. The
a ﬂat affect. In a girl with BN and normal weight, the genital psychotherapist provides individual and/or family therapy.
exam may be normal. There may be dental caries or acid The psychopharmacologist provides medication, which can
erosion of the anterior teeth and parotid hypertrophy from be helpful if there is concomitant depression or obsessive
vomiting. compulsive features or bulimia present. A number of trials
As the clinician assesses the history and physical have found reduction in binge vomiting in double-blind,
examination of the adolescent, the possibility of other placebo-controlled trials of tricyclic antidepressants and of
diagnoses must be entertained but are usually excluded by selective serotonin reuptake inhibitors (SSRI).1,3,32–34 If the
the typical history and a few simple laboratory tests. Other family will allow it, the school nurse or counsellor can
possibilities for weight loss include malignancy, central provide extra support and structure for the patient who is
nervous system (CNS) tumor, inﬂammatory bowel disease, eating lunch at school.
celiac disease, malabsorption, diabetes mellitus, hypo- The indications for hospitalization include unstable vital
thyroidism, hypopituitarism, Addison’s disease, primary signs, hypotension, orthostasis, bradycardia, hypothermia,
depression (with secondary anorexia), human immuno- severe malnutrition (less than 75 to 80% ideal body weight),
deﬁciency virus (HIV) infection and chronic illness, among dehydration, abnormal electrolytes, arrhythmias, acute food
others. Thus, the typical laboratory examination obtained at refusal, uncontrollable bingeing and purging, suicidality and
the ﬁrst visit of an evaluation for AN are complete blood failure of outpatient therapy. Treatment options include
count (CBC), differential, sedimentation rate, urinalysis, medical hospitalization, psychiatric hospitalization if the
electrolytes, glucose, calcium, magnesium, phosphorous, patient is medically stable, day treatment program, evening
blood urea nitrogen (BUN), creatinine, thyroid function groups (which may be treatment groups or support groups)
Eating disorders in adolescent girls 5
and long-term residential treatment in particularly severe colleagues40 have reported a positive effect of estrogen
cases. progestin therapy on bone density for patients who were at
Over the past six years at Boston Children’s Hospital, we <70% of ideal bodyweight. Gordon and colleagues41 have
have observed an increase in the severity of patients begun examining the effect of oral dehydroepiandrosterone
presenting with AN to our Eating Disorders Program. (DHEA), which has resulted in improvement in some bone
Bravender et al.35 have reported that although the age and markers. Grinspoon and colleagues have studied IgF-1 for its
BMI stayed the same between 1991 and 1996, the heart rate potential effect on bone density.42 Calcium and vitamin D
was lower in 1996 (52 beats/min vs. 66 beats/min in 1991), supplements are important to improve bone density;
the duration of amenorrhea was longer in 1996 (7.5 vs. 3.7 adolescent girls require 1300–1500 mg of calcium and 400
months) and more patients were admitted on the day of ﬁrst units of vitamin D daily. Physical activity in the presence of
presentation (45% vs. 0%). The changes in characteristics of normal menses is associated with increased bone mass.
patients presenting and the impact on various insurance In addition to deaths from cardiac complications and
policies and guidelines for the use of preventive services, electrolyte abnormalities, suicide attempts and completion in
mental health and nutritional referrals deserve further the course of eating disorders remain a signiﬁcant source of
investigation. morbidity and mortality. A 10-year follow-up study of AN
found an overall mortality of 6.6%.43
The complications of AN and BN include problems such as
osteoporosis, cardiac impairment including mitral valve Several research studies have examined long-term outcomes
prolapse, long QTC interval, arrhythmias and sudden death, in adolescents with AN and BN.43–48 Strober and
cognitive changes, difﬁculties in psychosocial functioning, colleagues,46 in a follow-up study of patients who had been
gastrointestinal dysfunction, endocrinologic changes in- hospitalized in the UCLA Eating Disorders Program (mean
cluding impaired growth, electrolyte abnormalities, dental BMI 14.1; 69.9% of IBW), have reported that full recovery
erosion and enlarged salivary glands in bulimic patients and was evident in 76% of patients with a median time of 79
suicide.36–38 In the early 1980s, Pugliese and colleagues months, a partial recovery in 10% of patients with a median
published data showing that fear of obesity and failure to time of 57 months and a chronic course was observed in
gain weight during adolescence were associated with 14%. A more protracted recovery was predicted by hostile
impairment of linear growth, which reversed with adequate attitudes towards a patient’s family and extreme compul-
weight gain.39 sivity in daily routines. The probability of chronic outcome
Osteoporosis is a critical issue for adolescents with AN was associated with extreme compulsive drive to exercise
and amenorrhea. The lowered bone density is often not reported at the time of discharge and a history of poor social
totally reversible and may be a source of short-term and relating prior to the onset of the illness. A patient’s poor
long-term morbidity. The degree of osteoporosis has been relationship with her family was a predictor of poor
associated with low weight for height and the duration of outcome. Ash and colleagues47 undertook a follow up of 583
estrogen deﬁciency. As little as 6 months of estrogen female adolescents with suspected AN seen at Children’s
deﬁciency may impact on bone density. Other factors Hospital Boston. One hundred and sixty-one of 286 actually
include inadequate calcium and vitamin D intake, hyper- completed the questionnaires at a mean age of 21.6~4.2
cortisolism and adrenal androgen deﬁciency. The most years and 5.3~3.4 years after presentation. Sixty-ﬁve
important approach to the prevention and treatment of percent of the subjects satisﬁed all four DSM-IV criteria for
osteoporosis in AN is the restoration of normal bodyweight. AN and 35% met two or three criteria and were labeled the
Hotta et al. have found that a BMI>16.4~0.3 kg/m2 has ‘partial group.’ It was noted that the full group compared
been associated with improved bone density,29 and Shomento with the partial group had no difference at follow up in
and Kreipe have found that a mean>92~7% of ideal eating attitudes, body distortion, depression, object relations,
bodyweight was associated with the return of menses.28 ego identity or global outcome, even though at initial intake
Interestingly, Golden and colleagues have noted that even the full group had lower BMI, a lower percentage ideal
with restoration of normal bodyweight, persistent amenorrhea bodyweight, greater fears of weight gain, greater body
has been associated with low leptin levels.16 Return of image disturbances and greater menstrual irregularity. Of
menses is obviously important to provide normal estrogen particular note, patients who did not satisfy psychosocial
effect and improved bone mass. Hormonal therapies have criteria (fear of becoming fat and/or distorted body image)
been used with mixed results and estrogen progestin therapy had better outcomes in global outcomes, eating attitudes,
has been subjected to only a few trials. Klibanski and attachment and social competence. There was no difference
6 SJ Emans
in outcome if patients failed to satisfy weight or amenorrhea 3 Becker AE, Grinspoon SK, Klibanski A, Herzog DB. Eating
criteria. disorders. N. Engl. J. Med. 1999; 340: 1092–8.
4 Woods ER. Eating Disorders. In: Carr PL, Freund KM,
A study by Keel et al. of 221 participants with BN
Somani S (eds). The Medical Care of Women. WB Saunders
11.6~1.9 years after presentation found that 30% were still Co., Philadelphia, 1995; 702–11.
engaging in bingeing and purging behavior and 70% had full 5 Forman SF. Eating Disorders: Epidemiology, pathogenesis,
or partial remission.48 Of note, 0.6% satisﬁed criteria for and clinical features. In: Rose BD (ed.). UptoDate (CD-
AN, 11% for BN and 18.5% ED NOS. Substance abuse and ROM). Wellesley, MA, 1999.
6 Lucas AR, Beard CM, O’Fallon WM et al. 50-year trends in
longer duration of illness were associated with a poor
the incidence of anorexia nervosa in Rochester Minn: A
prognosis. population-based study. Am. J. Psychiatry 1991; 148: 917–22.
7 Fairburn CG, Beglin SJ. Studies of the epidemiology of
bulimia nervosa. Am. J. Psychiatry 1990; 147: 401–8.
Prevention 8 Mukai T, Crago M, Shisslak CM. Eating attitudes and weight
preoccupation among female high school students in Japan. J.
Child Psychol. Psychiatry 1994; 35: 677–88.
Prevention requires multiple strategies that focus on the 9 Maloney MJ, McGuire J, Daniels SR, Specker B. Dieting
home, school, social and physical environments of young behavior and eating attitudes in children. Pediatrics 1989; 84:
people and assure preventive services in health-care 482–9.
settings.49 A number of school-based programs have early 10 Schoen C, Davis K, Collins KS et al. The Commonwealth
Fund Survey of the Health of Adolescent Girls. New York,
results suggesting that a change in attitudes can be
The Commonwealth Fund, 1997.
accomplished. Whether behaviors such as disordered eating 11 Kann L, Kinchen SA, Williams BI et al. Youth Risk Behavior
and the prevention of clinical AN and BN will change Surveillance – United States, 1997. Mor. Mortal. Wkly Rep.
remains to be determined. To date, it appears that the CDC Surveill. Summ. 1998; 47: (SS-3) 1–89.
important elements of school-based programs include media 12 Takeuchi S, Hayano J, Kamiya T et al. Body image and self
image in 712 junior high school students. Shinshin Igaku
skills, appreciation of body shapes and sizes, healthy eating
1991; 31: 367–73.
at home and school, physical activity and exercise, outreach 13 Forman SF, Goodman E. Characteristics of teens with an
to parents and communities and referrals for services. A anorexia nervosa-like syndrome from a recent population-
National Screening Day for Eating Disorders has been based nationally representative survey. (Abstract) Society for
carried out in colleges over the past few years and will be Adolescent Medicine. Los Angeles, CA, March 1999. J.
Adolesc. Health 1999; 24: 95.
introduced to high schools in 2000. Internet resources, such
14 Rastam M, Gillberg C. The family background in anorexia
as our web site at youngwomenshealth.org, offer another nervosa: A population based study. J. Am. Acad. Child
way to educate adolescents and their parents. Further Adolesc. Psychiatry 1991; 30: 283–9.
research is clearly needed to understand the multifactorial 15 Shrier LA, Pierce JD, Emans SJ, DuRant RH. Gender
etiology of disordered eating in adolescents and to develop differences in risk behavior associated with forced or
pressured sex. Arch. Pediatr. Adolesc. Med. 1998; 152: 57–63.
strategies to promote healthy eating, nutrition and physical
16 Golden NH, Kretzer PM, Yoon DJ et al. Leptin, resumption of
exercise over the life span. menses, and anorexia nervosa. (Abstract) Eighth New York
International Conference on Eating Disorders, New York City,
NY, April 1998.
Acknowledgements 17 Gordon CM, Emans SJ, Grace E et al. Regional cerebral blood
ﬂow difference during high-caloric food challenge in patients
with anorexia nervosa versus normal controls. Abstract,
My thanks for Sara Forman MD, for her leadership of the American Pediatric Society/Society for Pediatric Research,
Eating Disorders Program at Children’s Hospital Boston and New Orleans, 1 May 1998.
her critical help in reviewing the literature, and to Maria 18 Katzman DK, Lambe EK, Mikulis DJ et al. Cerebral gray
Luoni BS, for manuscript preparation. This work was matter and white matter volume deﬁcits in adolescent females
with anorexia nervosa. J. Pediatr. 1996; 129: 794–803.
supported in part by MCJ MA 259195 from the Maternal 19 Katzman DK, Zipursky RB, Lambe EK et al. A longitudinal
and Child Health Bureau, Health Resources and Services magnetic resonance imaging study of brain changes in
Administration. adolescents with anorexia nervosa. Arch. Pediatr. Adolesc.
Med. 1997; 151: 793–7.
20 Strober M, Lilenfeld LR, Kaye W, Bulik C. Genetic factors in
anorexia nervosa and bulimia nervosa. In: Cooper PJ, Stein A
References (eds). Feeding Problems and Eating Disorders, 2nd edn.
Harwood Publishers, Chur, Switzerland, in press.
1 Fisher M, Golden N, Katzman D et al. Eating disorders in 21 Woodside DBA. Review of anorexia nervosa and bulimia
adolescence: A background paper. J. Adolesc. Health 1995; nervosa. Curr. Probl. Pediatr. 1995; 25: 67–89.
16: 420–37. 22 Gershon ES, Schreiber JL, Hamovit JR et al. Anorexia
2 Harper G. Eating disorders in adolescents. Pediatr. Rev. 1994; nervosa and major affective disorders associated in families:
15: 72–7. A preliminary report. In: Guze SB, Earls FG, Barrett JE
Eating disorders in adolescent girls 7
(eds). Childhood Psychopathology and Development. Raven 36 Fisher M. Medical complications of anorexia and bulimia
Press, New York, 1983; 279–84. nervosa. Adolesc. Med. 1992; 3(3): 487–502.
23 Holland AJ, Hall A, Murray R et al. Anorexia nervosa: A 37 Kreipe RE, Harris JP. Myocardial impairment resulting from
study of 34 twin pairs. Br. J. Psychiatry 1984; 145: 414–19. eating disorders. Pediatric Ann. 1993; 21: 760–68.
24 Kendler KS, MacLean C, Neale M et al. The genetic basis of 38 Palla B, Litt IF. Medical complications of eating disorders in
bulimia nervosa. Am. J. Psychiatry 1991; 148: 1627–37. adolescents. Pediatrics 1988; 81: 613–23.
25 Walters EE, Kendler KS. Anorexia nervosa and anorexic-like 39 Pugliese MT, Lipshitz F, Grad G et al. Fear of obesity: A
syndromes in a population-based twin sample. Am. J. cause of short stature and delayed puberty. N. Engl. J. Med.
Psychiatry 1995; 152: 64–71. 1983; 309: 513–18.
26 American Psychiatric Association. Diagnostic and Statistical 40 Klibanski A, Biller BMK, Schoenfeld DA et al. The effects of
Manual of Mental Disorders, 4th edn. American Psychiatric estrogen administration on trabecular bone loss in young
Association, Washington, DC, 1994. women with anorexia nervosa. J. Clin. Endocrinol. Metab.
27 Frisch RE, McArthur JW. Menstrual cycles: Fatness as 1995; 80: 898–904.
determinant of minimum weight for height necessary for their 41 Gordon CM, Grace E, Emans SJ, Goodman E, Crawford MH,
maintenance or onset. Science 1974; 185: 949–50. LeBoff MS. Changes in bone turnover markers and menstrual
28 Shomento SH, Kreipe RE. Menstruation and fertility following function after short-term oral DHEA in young women with
anorexia nervosa. Adolesc. Pediatr. Gynecol. 1994; 7: 142–6. anorexia nervosa. J. Bone Miner. Res. 1999; 14: 136–45.
29 Hotta M, Shibasaski T, Sato K, Demura H. The importance of 42 Grinspoon S, Baum H, Lee K et al. Effects of short-term
body weight history in the occurrence and recovery of recombinant human insulin-like growth factor I administration
osteoporosis in patients with anorexia nervosa: Evaluation by on bone turnover in osteopenic women with anorexia nervosa.
dual X-ray absorptiometry and bone metabolic markers. Eur. J. Clin. Endocrinol. Metab. 1996; 81: 3864–70.
J. Endocrinol. 1998; 139: 276–83. 43 Eckert ED, Halmi KA, Marchi P et al. Ten-year follow-up of
30 Emans SJ. Delayed puberty and menstrual irregularities. In: anorexia nervosa: Clinical course and outcome. Psychol. Med.
Emans SJ, Laufer MR, Goldstein DP (eds). Pediatric and 1995; 25: 143–56.
Adolescent Gynecology. Lippincott Raven, Philadelphia, 1998. 44 Kreipe RE, Uphoff M. Treatment and outcome of adolescents
31 Putukian M. The female triad – eating disorders, amenorrhea, with anorexia nervosa. Adolesc. Med. 1992; 3: 519–40.
and osteoporosis. Med. Clin. N. Am. 1994; 78: 345–56. 45 Kreipe RE, Churchill BH, Strauss J. Long-term outcome of
32 Forman SF. Treatment of anorexia nervosa and bulimia adolescents with anorexia nervosa. Am. J. Dis. Child. 1989;
nervosa. In: Rose BD (ed.). UptoDate (CD-ROM). Wellesley, 143: 1322–7.
MA, 1999. 46 Strober M, Freeman R, Morrell W. The long-term course of
33 Pope Jr HG, Hudson JL, Jonas JM, Yurgelun-Todd D. Bulimia severe anorexia nervosa in adolescents: Survival analysis of
treated with imipramine: A placebo-controlled, double-blind recovery, relapse, and outcome predictors over 10–15 years in
study. Am. J. Psychiatry 1983; 140: 554–8. a prospective study. Int. J. Eat. Disord. 1997; 22: 339–60.
34 Fluoxetine Bulimia Nervosa Collaborative Study Group. 47 Ash JB, Piazza E, Harrison R et al. DSM-IV criteria for
Fluoxetine in the treatment of bulimia nervosa. A multicenter, anorexia nervosa as predictors of outcome among eating
placebo controlled, double-blind trial. Arch. Gen. Psychiatry disordered adolescents. J. Adolesc. Health 1997; 20: 150.
1992; 49: 139–47. 48 Keel PK, Mitchell JE, Miller KB et al. Long-term outcome of
35 Bravender T, Robertson L, Woods ER, Gordon C, Forman S. bulimia nervosa. Arch. Gen. Psychiatry 1999; 56: 63–9.
Is there an increased clinical severity of patients with eating 49 Neumark-Sztainer D. School-based programs for preventing
disorders under managed care? J. Adolesc. Health 1999; in eating disturbances. J. Sch. Health 1996; 66: 64–71.