More than eight million Americans suffer from eating disorders. Approximately 90% of them are young women; however, middle-aged women, children, and men are also affected. The prevalence of eating disorders appears to vary by the population being studied. Recently, the recognition of binge eating disorder (BED), a putative diagnosis currently a form of “Eating Disorder Not Otherwise Specified,” has changed the face of eating disorders. While anorexia (AN) and bulimia nervosa (BN) appear to primarily affect women, the ratio of women to men with BED is approximately 3:2. There are also significant cross-cultural differences in the prevalence and presentation of AN, BN, and BED.

Eating disorders are more prevalent in industrialized societies (where food is abundant and thinness is valued as attractive) than in developing countries. Women in western countries traditionally have exhibited greater concern for body habitus than those in developing countries, who appear to be more accepting of and comfortable with a fuller body shape. In many of the latter societies, a fuller figure has been considered the cultural stereotype of attractiveness, although this ideal appears to change when individuals from these societies integrate into western culture.

Westernization has affected many countries, and individuals from other cultures should not be excluded from consideration of an eating disorder diagnosis. Immigration from non-Western to Western cultures appears to place individuals at greater risk for eating disorders. Indeed, degree of acculturation into American society is associated with eating disorder risk, likely due to the adoption of Western body-image ideals. As individuals (particularly girls and women) from cultures in which AN is unknown or extremely rare immigrate to westernized societies with higher rates of AN, they tend to develop disorders as they attempt to acculturate.

Two core features are common across all eating disorders; namely, severe disturbance in eating habits, and excessive concern and/or dissatisfaction with body shape and weight. However, on the surface, individuals with AN and BED present quite differently. Further, AN, BN, and BED differ in terms of prevalence, demographic correlates, and medical ramifications. Classification of both overeating and under-eating disorders into a single category poses difficulties for the conceptualization and treatment of these conditions.

Before detailing the clinical characteristics of various eating disorders, it is necessary to identify what is meant by “normative eating.” In so doing, it becomes apparent that a great deal of dieting occurs in Western culture as part of normal eating. In fact, estimates suggest that anywhere from 15% to 80% of the population may be dieting at a given time. Despite these statistics, over 65% of the adult population, and about 16% of children and adolescents aged 2-19 years, are considered overweight or obese. High prevalence of obesity appears to disproportionately affect those of racial and ethnic minorities. About 70% of African American and Mexican American adults, compared to about 62% of non-Hispanic white adults, are considered overweight or obese. Similarly, 21% of African American and Mexican American children, compared with 14.6% of non-Hispanic white children, are overweight.

The term “dieting” in lay culture has been used to describe a wide variety of behaviors ranging from healthful (eg, eating more vegetables, increasing exercise) to extreme (eg, prolonged fasting, self-induced vomiting). Appropriateness of dieting should be considered in light of the specific behaviors that comprise dieting. Further, consideration of dieting in the context of an individual’s weight status is an important factor in evaluating whether dieting is pathological (in underweight or non-overweight individuals) versus appropriate (in overweight individuals). It has been suggested that dieting typically precedes eating disorder onset in cases of BN, while for BED, binge eating has been reported as preceding the onset of dieting in approximately half of cases.

Women are most likely to restrict their food intake to control their weight or lose weight, but increasingly men are also engaging in dieting behavior. Perhaps most worrisome is the prevalence of dieting among adolescents and even children. Data suggest that that 40% of 9-year-old girls have dieted, and even 5-year-olds voice concern about their diet that appear to be linked to cultural standards for body-image. Although most individuals who diet do not develop an eating disorder, dieting, in combination with other factors, may be an important precipitant to the development of eating disorders. The acceptance of dieting as “normative” may prohibit recognition of problem eating.

The prevalence and incidence rates for eating disorders vary significantly, depending on the disorder and the population. Generally speaking, of patients with classic signs and symptoms of AN or BN, 90% are female, 95% are white, and 75% are adolescent when they develop the disorder. These data are substantiated by several cross-cultural studies that have reported few, if any, cases in rural areas of Africa, the Middle East, or Asia with the exception of Japan, the only non-Western country that has seen a substantial and persistent increase in eating disorders. AN has been implicated as a “culture-bound syndrome” because certain cultural mores are reflected in the signs and symptoms of the disorder. In adults, recent studies of racial and ethnic differences in eating disorder prevalence within a nationally-representative sample suggest that AN is less common in African American and Latino populations compared to non-Hispanic whites. However, several studies have shown that other abnormal eating behaviors may be as common more so among African American women (eg, purging by laxatives vs vomiting, binge eating). African American women are also more likely to develop BN or BED than AN, and a recent study found a strong association between BED and obesity in this population. Given the high rates of obesity in ethnic minority populations, experts have postulated that BED is a significant problem among these groups. The predominance of non-Hispanic whites among cases of AN may contribute to cultural bias in diagnosis, with less recognition of eating disorders among ethnic minorities.

Compared to adults, the prevalence of eating disorders among adolescents is characterized by a differential racial and ethnic pattern. Indeed, among high school students, Hispanic and non-Hispanic white girls tend to report similar levels of eating disordered attitudes and cognitions, such as excessive shape and weight concern and extreme dieting, which may be risk factors for full-syndrome disorders. African American girls, however, report lower body weight concerns and behaviors than girls of other ethnicities. Among adolescent boys, nearly all ethnic minorities report more eating disorder symptoms and weight concerns compared to Caucasian boys. How subthreshold disturbances and risk factors manifest into differential prevalence of full-syndrome disorders across cultures is not well understood.

It is traditionally thought that AN and BN tend to affect adolescent girls of middle to upper socioeconomic status. Age- and sex-specific estimates suggest that about 0.5%-1% of adolescent girls develop AN, whereas 5% of older adolescent and young adult women develop BN. This population also exhibits a high frequency of coexistence between AN and BN. It has been reported that as many as 50% of AN patients may exhibit bulimic behaviors while 30%-80% of patients with BN have a history of AN. Although constituting a small segment of patients with eating disorders, male adolescents must not be forgotten. Most, however, tend to have a diagnosis of BN or BED.

Data suggest that approximately 3%-5% of people surveyed in a general population have BED. Although being overweight is not a criterion for the diagnosis of BED, it has been estimated that slightly over 11% of individuals who join Weight Watchers and 30% of individuals who present to hospital-based weight control programs meet the diagnosis of BED. In contrast to AN and BN, BED appears to afflict adults of all socioeconomic strata and education level equally. Furthermore, BED is often diagnosed in middle-aged adults.

Finally, it should be noted that despite the emphasis on AN, BN, and BED in both the literature and the media, the diagnostic category of “eating disorder not otherwise specified” (EDNOS)—excluding BED—is the most prevalent eating disorder in the United States, affecting 6%-10% of young women. Recent research suggests that individuals diagnosed with EDNOS did not differ significantly from AN and BED in terms of eating or general pathology. However, individuals with BN exhibited greater eating and general psychopathology compared to EDNOS. Girls meeting all criteria for AN except amenorrhea did not differ from full-syndrome AN. Further, nearly 40% of individuals diagnosed with EDNOS went on to develop either AN or BN with 1-2 years. Thus, EDNOS seems to represent a heterogenous category whose symptoms overlap substantially with traditional eating disorders. Clinicians should monitor possible progression of EDNOS to full syndrome AN, BN, or BED, especially given the paucity of treatment recommendations for EDNOS.

The origins of eating disorders are extremely complex and poorly understood. However, biological, psychological, cultural, and societal factors are likely contributors to the predisposition, precipitation, and perpetuation of these disorders. Typically, individuals with eating disorders are thought to have a biological or genetic predisposition that is activated by environmental (ie, sociocultural, psychosocial) factors.

Risk factors for developing an eating disorder include participation in activities that promote thinness (eg, ballet dancing, modeling, and athletics) and certain personality traits, such as low self-esteem, difficulty expressing negative emotions, difficulty resolving conflict, being perfectionistic, and neuroticism/anxiety. Mounting data also support substantial biological predispositions to AN and BN. Mothers and sisters of probands who had AN were found to have eight times the risk of developing an eating disorder compared with the general population. Genetic studies also lend strong support to the underlying biological supposition regarding eating disorders. Twin studies have shown heritability estimates in the 50%-90% range for AN and 35%-50% for BN, with monozygotic twins having higher concordance than dizygotic twins. A strong association between AN and BN in families has also been found in the Virginia Twin Registry.

Eating disorders may also be precipitated by psychosocial factors in vulnerable individuals. These precipitating factors often relate to developmental tasks of adolescence and include maturation fears, particularly those related to sexual development, peer group involvement, independence and autonomy struggles, family conflicts, sexual abuse, and identity conflicts. Two other psychological factors that figure permanently in the pathogenesis of BN or BED are sexual trauma and depression. Patients with either of these disorders are predisposed to have a family and personal history of depression. Therefore, it is important to note the presence of depression or history of sexual trauma during the initial patient assessment.

In the past two decades, the number of men who openly report dissatisfaction with their physical appearance has tripled. Today nearly as many men as women say they are unhappy with how they look. Fifty percent more men reportedly seek evaluation and treatment for eating disorders than they did in the 1990s. This trend may by rooted in an obsession with “six-pack abs” and bulging biceps that seems especially common among athletes and fitness enthusiasts. Pursuit of muscularity to the exclusion of healthful habits may be a precursor to eating disorder development in males. Additionally, exercise status and sexual orientation are two risk factors for eating disorders in men. Often men who develop eating disorders have a history of being overweight when they were younger. Men considered to be at increased risk of developing eating disorders include:

• Athletes, especially those participating in sports that work against gravity, such as gymnastics.
  
• Men with gender issues.
  
• Men with personality traits such as perfectionism and impulsive behaviors, and those who have anxiety.
  
• Obese boys who face teasing and have low self-esteem.

Increasingly, research on the risk-factors for eating disorders has indicated that ED symptoms may emerge as early as middle childhood. Dieting, concerns about shape and weight, and body dissatisfaction are present in children as young as 5 years of age, and common among children aged 8-12 years. Early factors shown to be associated with later development of AN include feeding and gastrointestinal difficulties during infancy and early childhood, maternal body dissatisfaction, and dieting and shape concerns during early and middle childhood. Similarly, early concerns with eating and weight are potential red-flags for future development of BN. Overweight status during childhood has been shown to be related to future development of both BN and BED.

Loss of control eating, defined as the feeling of being unable to control what or how much one is eating regardless of the amount of food consumed, has been reported in young children. Loss of control eating is fairly common among children, especially among those who are overweight (prevalence estimates among overweight youth range from 4% to 45%). LOC eating in youth is analogous to binge eating in adults, and may be a precursor for future development of BED. Children who have loss of control eating are at risk for gaining excess weight as they grow, and they have more disturbed eating patterns, symptoms of depression and anxiety, and behavior problems compared to children without such behaviors. Children who are at risk for becoming overweight adults, including those who are overweight and who have overweight parents, should be queried about loss of control eating to monitor potential progression to BN or BED.

Eating disorders are serious and complex problems, and the earlier an eating disorder is identified, the better the patient’s chance of recovery. This makes a compelling argument for targeted screening of at-risk groups, including gymnasts, runners, body builders, wrestlers, dancers, rowers, and swimmers. These groups warrant close monitoring because their sports or livelihood dictate weight restriction. The populations at highest risk for AN and BN are female adolescents and young adults, and screening should occur throughout adolescence, especially at ages 14 and 18 years. This correlates with the transition to high school and college and the associated stressors.

In contrast to AN and BN, which typically emerge during adolescence, BED is most frequently detected during middle adulthood, even though many individuals report an onset of the disorder in their mid-twenties and initial binge eating behaviors can begin even earlier. Because of the relatively high prevalence of BED in community samples (3%-5%), routine screening for binge eating among overweight adults may be warranted. Almost all individuals seeking treatment for weight control should be screened for BED because of the high incidence of this disorder in this group (about 30%-50%). Although these individuals may fall short of meeting the full criteria for BED, the problematic attitudes associated with the disorder will likely be uncovered. The tools used for screening can be very sophisticated and vary with the population being assessed. However, there are some that are easily incorporated into the routine primary care office visit (Table 11-1). These questions are very helpful for the early detection of BN and BED; some individuals with these disorders can be uncovered using self-report alone. Those with AN, on the other hand, are more often resistant to self-reporting and usually require reporting by others (ie, parents, friends). Therefore, it is imperative that parents, friends, teachers, family, dentists, and physicians become educated about the possible signs and symptoms associated with these difficult-to-manage disorders to facilitate prevention or early management of these individuals. Some cases of BN may be similarly difficult to detect, because many patients with BN are of a healthy weight status, and maintain secrecy surrounding bulimic behaviors.