Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Patients With an Eating Disorder
Overview
Eating disorders are serious psychiatric illnesses, with high rates of morbidity and mortality. They are most common in young women, but affect people of all ages, genders, races, ethnicities, and socioeconomic statuses. Individuals with eating disorders can develop serious medical complications, requiring multidisciplinary treatment. Given these risks, early detection and treatment of an eating disorder can increase the likelihood of recovery. Medical stabilization is key before beginning mental health treatment, both to ensure the patient's safety and to enhance treatment efficacy.
Epidemiology
Overall, eating disorders are most common in adolescent and young adult females. In a national, cross-sectional survey of 9282 adults, the life-time prevalence was 0.9% for anorexia nervosa (AN), 1.5% for bulimia nervosa (BN), and 3.5% for binge eating disorder (BED) among females; and 0.3% for AN, 0.5% for BN, and 2.0% for BED among males. Further, in an 8-year old prospective study in an all-female ethnically heterogeneous sample, the life-time prevalence by age 20 was 0.8% for AN, 2.6% for BN, and 3.0% for BED. Sub-threshold or atypical presentations, categorized in the Diagnostic and Statistical Manual of Mental Disorders, 5th edition as other specified feeding or eating disorder (OSFED) are also common. One community study of adolescent female twins showed that the prevalence rates for OSFED (5%) were comparable with full-threshold eating disorders (5.4%), with no significant difference in impairment between the two groups.
Although eating disorders are most prevalent in westernized societies, they affect a diverse range of individuals. More broadly, eating disorders affect people in many different cultures and countries across the world. For example, individuals from Hong Kong with eating disorders had diagnostic classifications that corresponded well with life-time eating disorder phenotypes observed in the United States and Europe. Lifetime prevalence rates of eating disorders range from 0.9% to 3.5% for White women; 0.14% to 2.36% for African-American women; 0.12% to 2.67% for Asian-American women; and 0.12% to 2.31% for Latina women.
Onset and Course
The etiology of eating disorders is largely unknown, with likely contributions from genetic, biological, psychological, and sociocultural factors. General risk factors for developing an eating disorder of any kind include low self-esteem, body dissatisfaction, and dieting.
AN typically begins during adolescence and/or young adulthood, with recent data showing a trend toward a decreasing age of onset. Children as young as 7 years old may express body dissatisfaction and a desire to lose weight. The onset of BN is usually later, in late adolescence and young adulthood, with a peak age of onset between 16 and 20 years. The onset of BED is often later still, with an onset at 17 years or later.
The heritability of eating disorders has been well-established. Family studies have shown a higher prevalence of AN, BN, and BED among relatives of individuals with the same disorder. Longitudinal twin data suggest that genetic influence on the development of an eating disorder may be particularly activated during puberty.
A recent longitudinal study of AN and BN showed recovery rates at 31.4% and 68.2%, respectively, after 9 years, and 62.8% and 68.2% after 22 years. The remainder followed a chronic course. Similarly, a 6-year longitudinal study of BED found that 43% of individuals with BED continued to be symptomatic. Longitudinal studies have also illustrated patterns of diagnostic crossover, in which individuals move away from restrictive eating and towards bingeing and/or purging.
Eating disorders are also associated with serious medical complications and a high mortality rate. In a longitudinal study of 246 women with AN, 16 eventually died due either to medical complications of AN (such as cardiac arrest, gastrointestinal hemorrhages, esophageal ulceration, or substance overdose) or by suicide. A recent meta-analysis found that individuals with AN were 5.2 times more likely to die prematurely from any cause, and 18.1 times more likely to die by suicide, than 15–34-year-old females in the general population.
Because nearly half of eating disorder cases are not detected in primary care settings, more specialized outpatient medical services can provide other opportunities for diagnosis. Furthermore, early detection and intervention are crucial to improving prognosis.
Case 1
Ms. D, a 21-year-old Caucasian college student without a psychiatric history, was brought to the clinic by her mother who was concerned that her daughter was exhibiting abnormal eating habits since her return home for summer break. Her mother described Ms. D as visiting the bathroom with increasing frequency after family meals and she had noticed food disappearing from the fridge and cupboards on several occasions. Her mother was concerned about Ms. D's “secretive” behavior and had witnessed her eating large quantities of chocolates and cookies in her room alone.
On interview, Ms. D was oriented, lucid, and avoided eye contact. On mental state exam, she was moderately depressed and quiet. She perceived her weight to be higher than she desired, and described low self-esteem; basing her self-worth almost entirely on her shape and weight. She denied any suicidal ideation but reported she had considered self-harm several times.
Ms. D first binged and purged following a set of disappointing exam results. Over the following 5 months, these episodes increased to 3 to 4 times per week. Following a binge, she described immediate feelings of guilt and shame. She denied using diuretics or diet pills, but reported taking laxatives 5 to 6 times each week. She became increasingly socially anxious and was afraid to eat around others for fear of losing control, and friends observing her behavior. At home she had begun to store food in her room or binge in the kitchen alone at nighttime.
On examination her BMI was 20.2, she had mild parotid swelling and signs of dental erosion. Russell's sign was absent and there was no abdominal bloating or tenderness. Ms. D reported occasional palpitations, generalized fatigue, and regular menstruation. Laboratory tests revealed a mild hypokalemia (3.1 mmol/L); other electrolytes were within normal limits.
Given that her weight was in the normal range and she was medically stable; Ms. D was diagnosed with BN and offered outpatient care and potassium supplementation. The team liaised with her PCP who prescribed 60 mg of fluoxetine daily. She was started on a 4-stage 20-session CBT program to interrupt the binge–purge cycle and reduce over-valuation of shape and weight.
Within the first 4 weeks of treatment, Ms. D was able to reduce her episodes of binging and purging to once every other week, and by completion of CBT, she had not binged or purged in several weeks. Her mood also improved.
Differential Diagnosis and Initial Assessment of Eating Disorders
Given the considerable overlap in symptoms across eating-disorder diagnoses (see DSM-5), differential diagnosis can be challenging. Nonetheless, an accurate and specific diagnosis is crucial in order to select the optimal treatment strategy. Two specific challenges clinicians may face are: (1) not having the time or expertise to screen for an eating disorder, and (2) misattributing physical complaints to a specific medical abnormality, rather than an eating disorder. This is particularly true for symptoms associated with gastrointestinal (GI) distress, weight loss, or dental abnormalities. Patients may initially present to a medical doctor or to a dietitian, and subsequently be referred for psychiatric consultation after an extensive evaluation.
Clinical Detection of an Occult Eating Disorder
Detection of an eating disorder can be challenging, in part due to the reluctance of many patients to acknowledge and receive care for their illness. Thus, some individuals may come to medical attention for an eating disorder only after they arouse clinical suspicion on a medical or pediatric service due to unexplained weight changes or symptoms, or when family members report concerns.
Given the prevalence and associated mortality and morbidity of eating disorders, clinicians should ask sensitive yet direct questions about weight and dieting history in order to detect body image disturbances and weight fluctuations. A national eating disorders screening program study indicated that 91% of individuals who had not previously volunteered eating and weight concerns to a health professional ultimately disclosed such concerns when asked directly.
Because clinical history does not always elicit accurate or complete data, collateral history, physical signs, observable behaviors, and longitudinal clinical course should be used when possible as part of the evaluation. Physical signs may include low weight, parotid hypertrophy, excoriations on the dorsum of the hand (Russell's sign), or characteristic dental erosion. Laboratory studies may yield data suggestive of purging or nutritional compromise but do not have sufficient specificity or sensitivity to be useful in screening. Collateral data can sometimes be helpful, although often, family members are not aware of symptomatic behaviors.
Differential Diagnosis
The DSM-5 characterizes AN by restriction of food intake leading to a significantly low weight (i.e., a weight less than minimally normal for adults, and less than minimally expected for children); fear of gaining weight; and body image disturbance. Clinicians can further specify whether the patient has mild (BMI ≥17 kg/m 2 ), moderate (BMI 16–16.99 kg/m 2 ), severe (BMI 15–15.99 kg/m 2 ), or extreme (BMI <15 kg/m 2 ) AN. An important revision from DSM-IV to DSM-5 was the exclusion of the amenorrhea criterion, given that individuals without amenorrhea may have just as severe an eating disorder as those who are amenorrheic. Individuals who do not endorse a fear of gaining weight, and thus are considered to be “non-fat phobic,” may still be diagnosed with AN in the setting of low weight and persistent behavior that interferes with weight gain. There are two subtypes of AN: restricting (AN-R) and binge eating/purging (AN-BP).
BN is defined in DSM-5 as the presence of recurrent episodes of binge eating and inappropriate compensatory behaviors at least once per week for 3 months or more, in order to control weight. During a binge episode, an individual consumes an objectively large amount of food, coupled with a feeling of lack of control. Individuals with BN also overvalue the importance of shape and weight in ascertaining self-worth. Of note, a key difference between AN-BP and BN is weight status; individuals with BN are not underweight, and are normal weight or even overweight. Patients may utilize a range of inappropriate compensatory behaviors in order to purge calories consumed. Self-induced vomiting is a common purging method, although it is an inefficient means of ridding the body of calories. One study found that individuals retain over half the calories they consume after vomiting. Individuals who purge may also use laxatives, diuretics, or enemas. While these are not as common as vomiting, many of these products are readily available over-the-counter, and can cause serious medical consequences. Patients with co-morbid insulin-dependent diabetes should also be assessed for inappropriately withholding their insulin. This behavior places them at risk for diabetic ketoacidosis, as well as long-term complications of poorly controlled blood sugars, and may necessitate inpatient care for stabilization and safety. The severity rating for BN is based on compensatory episodes per week (mild: 1–3; moderate: 4–7; severe: 8–13; extreme: ≥14) but the clinician is given freedom to increase the severity rating given the levels of other symptoms.
BED is characterized primarily by episodes of recurrent binge eating (at least once per week for 3 months) without compensatory purging. The binge episodes must be associated with three or more of the following: eating rapidly; eating until uncomfortably full; eating large amounts despite lack of hunger; eating in solitude to avoid embarrassment; and/or feelings of disgust or guilt following the episode. If the individual who is bingeing is also underweight or is using compensatory purging behaviors following a binge, a diagnosis of AN or BN (respectively) is more appropriate. Although BED is sometimes associated with obesity, not all individuals who are obese have BED, and not all individuals who have BED are obese.
The DSM-IV diagnosis of eating disorders not otherwise specified (EDNOS) was replaced in DSM-5 with OSFED. As with EDNOS, OSFED serves as a diagnosis for individuals who clearly have a clinically impairing eating disturbance but do not meet criteria for another eating disorder. OSFED comprises five example presentations, including atypical AN (all criteria for AN met, except the patient is not low weight); sub-threshold BN (bingeing and/or purging occurs less than once per week or for less than 3 months); sub-threshold BED (bingeing occurs less than once per week or for less than 3 months); purging disorder (purging without binge episodes); and night-eating syndrome (an individual awakens from sleep and consumes food). DSM-5 also includes a diagnosis called unspecified feeding or eating disorder (UFED) for instances in which there is insufficient information to confer a specific eating-disorder diagnosis, but symptoms are clearly causing distress or impairment.
Medical and other psychiatric causes of poor appetite and weight or vomiting should be excluded with appropriate history, examination, and diagnostic tests. Medical illnesses that the practitioner needs to differentiate from eating disorders include endocrinologic disorders (such as diabetes mellitus and hyperthyroidism), brain tumors, cancer, occult infections, and multiple GI disorders, such as celiac disease, peptic ulcer disease, and gastritis. Moreover, because eating disorders are frequently co-morbid with mood and anxiety disorders, personality disorders, and substance use disorders, a complete assessment of mental status and psychiatric history should be conducted. Finally, a careful assessment of suicide risk is essential.
Weight Assessment
Weight assessment is not only integral to the diagnosis of AN but it is also necessary for therapeutic management across the spectrum of eating disorders. Weight and height should be measured during the initial assessment. Because some patients may attempt to disguise a low weight by adding weight through jewelry, clothing, and hidden items, weights should be measured with the patient in a hospital gown only. If water loading is suspected, patients can be asked to void prior to weighing; urine specific gravity may be informative if it is relatively low.
BMI is the clinical standard for weight assessment for adults. The formula is as follows:
For children, the BMI generally increases with age, and therefore BMI centiles are a more accurate measure. They can be assessed by measuring height and weight and plotting on the U.S. Centers for Disease Control and Prevention BMI-for-age charts.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here