Food-Related Disorders: Anorexia Nervosa

Heather Munro Prescott. Cambridge World History of Food. Editor: Kenneth F Kiple & Kriemhild Conee Ornelas. Volume 1. Cambridge, UK: Cambridge University Press, 2000.


Anorexia nervosa is a psychophysiological disorder—usually of young females—characterized by a prolonged refusal to eat or maintain normal body weight, an intense fear of becoming obese, a disturbed body image in which the emaciated patient feels over-weight, and the absence of any physical illness that would account for extreme weight loss. The term “anorexia” is actually a misnomer, because genuine loss of appetite is rare and usually does not occur until late in the illness. In reality, most anorectics are obsessed with food and constantly struggle to deny natural hunger.

In anorexia nervosa, normal dieting escalates into a preoccupation with being thin, profound changes in eating patterns, and a loss of at least 25 percent of the original body weight. Weight loss is usually accomplished by a severe restriction of caloric intake, with patients subsisting on fewer than 600 calories per day. Contemporary anorectics may couple fasting with self-induced vomiting, use of laxatives and diuretics, and strenuous exercise.

The most consistent medical consequences of anorexia nervosa are amenorrhea (ceasing or irregularity of menstruation) and estrogen deficiency. In most cases amenorrhea follows weight loss, but it is not unusual for amenorrhea to appear before noticeable weight loss has occurred. The decrease in estrogen causes many anorectics to develop osteoporosis, a loss of bone density that is usually seen only in post-menopausal women (Garfinkel and Garner 1982).

By the time the anorectic is profoundly under-weight, other physical complications resulting from severe malnutrition begin to appear. These include bradycardia (slowing of the heartbeat), hypotension (loss of normal blood pressure), lethargy, hypothermia, constipation, the appearance of “lanugo” or fine silky hair covering the body, and a variety of other metabolic and systemic changes.

In addition to the physical symptoms associated with chronic starvation, anorectics also display a relatively consistent cluster of emotional and behavioral characteristics, the most prominent of which grows out of the anorectic’s deviation from normal eating habits. Severe restriction of food intake is sometimes alternated with bulimic phases, in which the anorectic engages in uncontrolled or excessive eating followed by self-induced vomiting and laxative abuse. Other unusual eating habits may include monotonous or eccentric diets, hoarding or hiding of food, and obsessive preoccupation with food and cooking for others.

Emotionally, anorexic patients are often described as being perfectionist, dependent, introverted, and overly compliant. Although studies have failed to find a consistent psychiatric symptom pattern for the disorder, frequently reported neurotic traits include obsessive-compulsive, hysterical, hypochondriacal, and depressive symptoms. A decrease in or disappearance of sexual interest is also a frequent concomitant of anorexia nervosa.

A distorted body image is an almost universal characteristic of anorectics, with many patients insisting they are overweight even when their bodies are extremely emaciated. As a result, most individuals with anorexia nervosa deny or minimize the severity of their illness and are usually highly resistant to therapy. The anorectic’s refusal to acknowledge her nutritional needs, and her steadfast insistence that nothing is wrong, make anorexia nervosa one of the most recalcitrant disorders in contemporary medicine.

Distribution and Incidence

Once considered to be extremely rare, the reported incidence of anorexia nervosa has more than doubled during the past 20 years (Herzog and Copeland 1985).The disorder is especially prevalent among adolescent and young adult women. Ninety to 95 percent of anorectics are young and female, and as many as 1 in 250 females between 12 and 18 years of age may develop the disorder. The exact incidence of anorexia and other eating disorders is difficult to determine, however, because of problems in conducting reliable epidemiological studies and the small samples on which many such studies are based (Crisp, Palmer, and Kalucy 1976).

The onset of anorexia nervosa occurs most often during adolescence, although some patients have become anorexic as early as age 11 and as late as the sixth decade of life. Patients are typically high achievers, with normal or above average intelligence. They also tend to come from middle- or upper-class families, although evidence of anorexia nervosa among working-class and poverty-class women is growing (Bulik 1987; Gowers and McMahon 1989; Dolan 1991).

Anorexia nervosa is comparatively rare in men: Approximately 5 to 10 percent of anorectics are male. The clinical picture for male anorexic patients is also much different from that for women. In general, male anorectics tend to display a greater degree of psychopathology, are often massively obese before acquiring the disorder, are less likely to be affluent, and are even more resistant to therapy than their female counterparts (Garfinkel and Garner 1982). There is growing evidence, however, that anorexia nervosa and bulimia are more common among men than previously believed. This is particularly true among homosexual men, who tend to experience more dissatisfaction with body image than do heterosexual men (Yager et al. 1988; Silberstein et al. 1989; Striegel-Moore,Tucker, and Hsu 1990).

Anorexia nervosa was also once thought to be comparatively rare among American blacks, Hispanics, Native Americans, lesbians, first- and second-generation ethnic immigrants, and individuals from disadvantaged socioeconomic backgrounds (Herzog and Copeland 1985). Recent research in this area, however, has indicated that the incidence of anorexia and other eating disorders among these groups is much higher than previously thought (see the section “Geographic and Demographic Features”).

Etiology and Epidemiology

Although the etiology of anorexia nervosa is an area of intense investigation, researchers have yet to reach a consensus about the origin of the disorder. The most sophisticated thinking on the subject regards anorexia nervosa as a multidetermined disorder that involves an interplay of biological, psychological, and cultural factors. Advocates of this model view these three etiological factors as reciprocal and interactive and believe it is simplistic to isolate one component as the single underlying cause of the disorder (Garfinkel and Garner 1982; Brumberg 1988).

Joan Brumberg (1988) has developed a multidetermined etiological model based on a two-staged conceptualization of anorexia nervosa that delineates the relative impact of sociocultural influences and individual biological and psychological variables in precipitating the disorder. In the first stage the “recruitment” phase of the illness—sociocultural factors play the dominant role. During this period, cultural assumptions that associate thinness with female beauty lead certain women into a pattern of chronic dieting. Indeed, research on the sociocultural causes of anorexia nervosa has linked the increased incidence of anorexia nervosa and other eating disorders with the tremendous cultural attention given to dieting and food, increasingly thinner standards of beauty, and the fitness movement (Schwartz, Thompson, and Johnson 1982; Chernin 1985; Orbach 1986; Bordo 1993). Yet sociocultural variables alone cannot explain why some women but not others move from chronic dieting to anorexia nervosa. Therefore, other individual factors must be implicated in the final development of the illness.

Brumberg’s model of anorexia nervosa relies on a second stage—career or acclimation—to correct the shortcomings of sociocultural explanations of the disorder. During the career phase, specific biological and psychological features determine which individuals develop the full-blown psychopathology of anorexia nervosa. In order to explain the transition between the recruitment and career phases of anorexia nervosa, Brumberg relies on recent research in the biological and social sciences, which has sought to uncover the unique physiological and psychological characteristics of anorexic patients.

Since the early 1900s, a number of different endocrinological and neurological abnormalities have been postulated as underlying biological causes of anorexia nervosa: hormonal imbalance, dysfunction in the satiety center of the hypothalamus, lesions in the limbic system of the brain, and irregular output of vasopressin and gonadotropin (Herzog and Copeland 1985). The search for a biomedical cause of anorexia nervosa is made difficult, however, by the fact that chronic starvation itself produces extensive changes in hypothalamic and metabolic function. Researchers in this area have yet to find a common biological characteristic of the anorexic population that is unmistakably a cause rather than a consequence of extreme weight loss and malnutrition (Brumberg 1988).

A more satisfactory explanation of the biological factors that contribute to the “career” phase of anorexia nervosa is the “addiction to starvation” model proffered by the British psychiatrists George I. Szmukler and Digby Tantum (1984). According to Szmukler and Tantum, patients with fully developed anorexia nervosa are physically and psychologically dependent on the state of starvation. Much like alcoholics and other substance abusers, anorectics find something gratifying or tension-relieving about the state of starvation and possess a specific physiological substrate that makes them more susceptible to starvation dependence than individuals who merely engage in chronic dieting. Szmukler and Tantum add, however, that starvation dependence is not the total explanation of anorexia nervosa. Rather, they believe that starvation dependence acts in conjunction with a range of sociocultural, psychological, and familial factors that encourage certain individuals to use anorexic behavior as a means of expressing personal anguish.

Current psychological models of anorexia nervosa fall into three basic categories: psychoanalytic, family systems, and social psychology. In both the psychoanalytic and family systems models, anorexia nervosa is seen as a pathological response to the developmental crisis of adolescence. Orthodox psychoanalysts, drawing on the work of Sigmund Freud, view the anorectic as a girl who fears adult womanhood and who associates eating with oral impregnation (Brumberg 1988). Family systems theory, however, offers a more complex explanation of the relationship between adolescence and anorexia nervosa. On the basis of clinical work with anorectics and their families, family systems therapists have found that the majority of anorexic patients are “enmeshed,” meaning that the normal process of individuation is blocked by extreme parental overprotectiveness, control, and rigidity. Anorexia nervosa is therefore seen as a form of adolescent rebellion against parental authority (Minuchin, Rosman, and Baker 1978; Bruch 1988).

Research in social psychology and the field of personality has devised several psychological tests to distinguish the psychological characteristics of anorectics from others in their age-group. One study has shown that although many of the psychological traits of anorectics and other women are indistinguishable, anorectics display a markedly higher degree of ineffectiveness and lower amount of self-esteem. Other studies have proposed that anorectics have actual cognitive problems with body imaging; still others suggest a relationship between anorexia nervosa and sex-role socialization (Garfinkel and Garner 1982).

Some researchers have attempted to fit anorexia nervosa within other established psychiatric categories such as affective disorders and obsessional neurosis. Many anorectics, in fact, display behavior patterns associated with obsessive-compulsive disorders: perfectionism, excessive orderliness and cleanliness, meticulous attention to detail, and self-righteousness. This correlation has led a number of researchers to suggest that anorexia nervosa is itself a form of obsessive-compulsive behavior (Rothenberg 1986). Depressive symptoms are also commonly seen in many patients with anorexia nervosa. Various family, genetic, and endocrine studies have found a correlation between eating disorders and depression. However, the association between anorexia nervosa and other psychiatric disorders remains controversial (Garfinkel and Garner 1982;Herzog and Copeland 1985).


The extent to which people suffered from anorexia nervosa in the past has been a subject of much historical debate. Some clinicians and medical historians have suggested that anorexia nervosa was first identified in 1689 by the British doctor Richard Morton, physician to James II (Bliss and Branch 1960; Silverman 1983). The medieval historian Rudolph Bell (1985) has dated the origins of anorexia nervosa even earlier, claiming that the medieval female saints, who were reputed to live without eating anything except the Eucharist, actually suffered from anorexia nervosa.

Other historians, however, have argued that attempts to label all historical instances of food refusal and appetite loss as anorexia nervosa are simplistic and maintain that the historical record is insufficient to make conclusive diagnoses of individual cases (Bynum 1987; Brumberg 1988). Although these historians agree that the final physiological stage of acute starvation may be the same in contemporary anorectics and medieval ascetics, the cultural and psychological reasons behind the refusal to eat are quite different.Thus, to reduce both to a single biomedical cause is to overlook the variety of social and cultural contexts in which certain individuals have chosen to refuse food.

The modern disease classification of anorexia nervosa emerged during the 1860s and 1870s, when the work of public asylum keepers, elite British physicians, and early French neurologists partially distinguished anorexia nervosa from other diseases involving loss of appetite (Brumberg 1988). In 1859, the American asylum physician William Stout Chipley published the first American description of sitomania, a type of insanity characterized by an intense dread or loathing of food (Chipley 1859). Although Chipley found sitophobia in patients from a broad range of social groups and age-groups, he identified a special form of the disease that afflicted adolescent girls.

Chipley’s work was ignored by his contemporaries, however, and it was not until the 1870s, when two influential case studies by the British physician William Withey Gull and the French alienist Charles Lasègue (Lasègue 1873; Gull 1874) were published, that physicians began to pay significant attention to anorexia in girlhood. Gull’s primary accomplishment was to name and establish anorexia nervosa as a coherent disease entity, distinct from mental illnesses in which appetite loss was a secondary feature and from physical “wasting” diseases such as tuberculosis, diabetes, or cancer. Despite widespread acclaim for Gull’s work with anorexic patients, however, late-nineteenth-century clinicians generally rejected the conception of anorexia nervosa as an independent disease. Instead, they viewed it either as a variant of hyteria that affected the gastrointestinal system or as a juvenile form of neurasthenia (Brumberg 1988).

Nineteenth-century physicians also tended to focus on the physical symptom of not eating and ignored the anorexic patient’s psychological reasons for refusing food. An important exception was Lasègue, who was the first to suggest the significance of family dynamics in the genesis and perpetuation of anorexia nervosa. Because of the somatic emphasis of nineteenth-century medicine, however, most medical practitioners of that time disregarded Lasègue’s therapeutic perspective. Instead, they directed medical intervention toward restoring the anorectic to a reasonable weight and pattern of eating rather than exploring the underlying emotional causes of the patient’s alleged lack of appetite (Brumberg 1988).

In the twentieth century, the treatment of anorexia nervosa changed to incorporate new developments within medical and psychiatric practice. Before the Second World War, two distinct and isolated models dominated medical thinking on anorexia nervosa.The first approach was rooted in late-nineteenth-century research in organotherapy, a form of treatment based on the principle that disease resulted from the removal or dysfunction of secreting organs and glands (Brumberg 1988). Between 1900 and 1940, a variety of different endocrinologic deficiencies were proposed as the cause of anorexia nervosa. In 1914, Morris Simmonds, a pathologist at the University of Hamburg, published a clinical description of an extreme cachexia due to destruction of the anterior lobe of the pituitary. Because patients with anorexia nervosa and those with Simmonds’s disease shared a set of common symptoms, many clinicians assumed that a deficiency in pituitary hormone was the cause of both conditions (Brumberg 1988).

Other researchers implicated thyroid insufficiency as the cause of anorexia nervosa. Research conducted at the Mayo Clinic in Rochester, Minnesota, during the period between the two world wars established the relationship between thyroid function and body weight and led many physicians to regard anorexia nervosa as a metabolic disorder caused by a deficiency in thyroid hormone. Throughout the 1920s and 1930s, insulin, antuitrin, estrogen, and a host of other hormones were also employed in the treatment of anorexia nervosa (Brumberg 1988).

The second major approach to anorexia nervosa in the early twentieth century grew out of the field of dynamic psychiatry, which emerged during the 1890s and early 1900s. Beginning in the last decade of the nineteenth century, practitioners in dynamic psychiatry increasingly focused on the life history of individual patients and the emotional sources of nervous disease. Two of the leading pioneers in this new field—Sigmund Freud and Pierre Janet—were the first to suggestively link the etiology of anorexia nervosa with the issue of psychosexual development. According to Freud, all appetites were expressions of libido or sexual drive. Thus, not eating represented a repression of normal sexual appetite (Freud 1959). Similarly, Janet asserted that anorexic girls refused food in order to retard normal sexual development and forestall adult sexuality (Janet 1903).

Because of the enormous popularity of endocrino-logic explanations, the idea of anorexia nervosa as a psychosexual disturbance was generally overlooked for more than 30 years. By the 1930s, however, the failure of endocrinologic models to establish either a predictable cure or a definitive cause of anorexia nervosa, the growing reputation of the Freudian psychoanalytic movement, and increased attention to the role of emotions in disease led a number of practitioners to assert the value and importance of psychotherapy in the treatment of anorexia nervosa. Although biomedical treatment of the disorder continued, most clinicians argued that successful, permanent recovery depended on uncovering the psychological basis for the anorectic’s behavior. Following up on the work of Freud and Janet, orthodox psychiatrists during this time postulated that refusal to eat was related to suppression of the sexual appetite and claimed that anorexic women regarded eating as oral impregnation and obesity as pregnancy (Brumberg 1988).

After World War II, a new psychiatric view of eating disorders, shaped largely by the work of Hilde Bruch, encouraged a more complex interpretation of the psychological underpinnings of anorexia nervosa. Although Bruch agreed that the anorectic was unprepared to cope with the psychological and social consequences of adulthood and sexuality, she also stressed the importance of individual personality formation and factors within the family that contributed to the psychogenesis of anorexia nervosa. Here, Bruch revived Las-gue’s work on the role of family dynamics in anorexia nervosa. According to Bruch, the families of most anorexic patients were engaged in a dysfunctional style of familial interaction known as “enmeshment”: Such families are characterized by extreme parental overprotectiveness, lack of privacy of individual members, and reluctance or inability to confront intrafamilial conflicts. Although superficially these families appeared to be congenial, Bruch wrote, this harmony was achieved through excessive conformity on the part of the child, which undermined the child’s development of an autonomous self. Anorexia nervosa, according to Bruch, was therefore a young woman’s attempt to exert control and self-direction within a family environment in which she otherwise felt powerless (Bruch 1973, 1988).

Bruch was also primarily responsible for the tremendous growth in the popular awareness of anorexia nervosa and other eating disorders in the 1970s and 1980s. Through her book, The Golden Cage: The Enigma of Anorexia Nervosa (1978), which sold over 150,000 copies, and numerous articles in Family Circle and other popular magazines, Bruch brought anorexia nervosa into common American parlance.

At the same time that the American public was becoming increasingly aware of anorexia nervosa, the number of reported cases of the disorder grew tremendously. This phenomenon has led some clinicians and social commentators to suggest that the popularization process itself may promote a “sympathetic host environment” for the disorder (Striegel-Moore, Silberstein, and Rodin 1986). As Bruch herself observed: “Once the discovery of isolated tormented women, it [anorexia nervosa] has now acquired a fashionable reputation, of being something to be competitive about. … This is a far cry from the twenty-years-ago anorexic whose goal was to be unique and suggests that social factors may impact the prevalence of the disorder” (Bruch 1988: 3-4).

Geographic and Demographic Features

Until recently, anorexia nervosa was believed to be a disorder largely confined to the United States, Canada, and Western Europe. Researchers also thought that the disease was virtually nonexistent in people of color and/or those from disadvantaged socioeconomic backgrounds. As early as 1880, S. Fenwick observed that anorexia nervosa “is much more common in the wealthier class of society than amongst those who have to procure their bread by daily labor” (Fenwick 1880: 11).This image of anorexia nervosa as a disease of abundance has persisted into the present day. Many researchers suggest that individuals from non-Western societies, minority groups, and impoverished backgrounds are “protected” from eating disorders because thinness is not highly valued in these communities, and fatness is often viewed as a sign of health and prosperity (Andersen and Hay 1985; Gray, Ford, and Kelly 1987; Gowers and McMahon 1989). The apparent absence of the disorder in developing nations and its high incidence among affluent social groups in Westernized countries led many clinicians to classify anorexia nervosa as a “culture bound” syndrome, meaning a disorder that is restricted to certain cultures primarily because of their distinctive psychosocial features (Prince 1985).

As a result of these views, none of the early literature on anorexia nervosa mentioned individuals from minority groups, non-Western countries, or lower socioeconomic classes (Bruch 1966; Kendall et al. 1973; Garfinkel and Garner 1980). The first cases of nonwhite anorectics appeared in a paper by M. P. Warren and R. L. Vande Wiele (1973), which noted 1 Chinese and 1 black person out of 42 patients seen at their New York clinic between 1960 and 1971. Other articles from the late 1970s and early 1980s mentioned one or two cases of nonwhite anorectics but did not offer any explanations of this phenomenon (Jones et al. 1980; Hedblom, Hubbard, and Andersen 1981; Garfinkel and Garner 1982; Roy-Byrne, Lee-Benner, and Yager 1984).

More recently, research on nonwhite and non-Western anorectics has grown significantly: Investigators have identified cases of the disorder in Malaysia (Buhrich 1981), Greece (Fichter, Elton, and Sourdi 1988), Nigeria (Nwaefuna 1981), Zimbabwe (Buchan and Gregory 1984), and Ethiopia (Fahy et al. 1988). The non-Western country to receive the most attention from researchers has been Japan, probably because it is one of the most Westernized East Asian countries. In Japan, anorexia nervosa and a binge-eating syndrome called Kirbarashi-gui have been well documented by researchers for a number of years (Nogami and Yabana 1977; Azuma and Henmi 1982; Nogami et al. 1984; Suematsu et al. 1985).

Within the United States and Great Britain, there has been a growing body of research on the incidence of anorexia nervosa in blacks (Pumariega, Edwards, and Mitchell 1984; Andersen and Hay 1985; Nevo 1985; Robinson and Andersen 1985;White, Hudson, and Campbell 1985; Silber 1986; Gray et al. 1987; Hsu 1987; Thomas and James 1988), Hispanics (Silber 1986; Hiebert et al. 1988; Smith and Krejci 1991), Asian-Americans and British-Asians (Nevo 1985; Root 1990), and Native Americans (Rosen et al. 1988; White-house and Mumford 1988; Smith and Krejci 1991), as well as recent immigrants from Eastern Europe (Bulik 1987), the Middle East (Garfinkel and Garner 1982), and the Caribbean (Thomas and Szmukler 1985; Holden and Robinson 1988). Researchers have also challenged the notion that lesbians are “protected” from eating disorders because lesbian ideology challenges culturally prescribed beauty ideals (Striegel-Moore et al. 1990; Thompson 1994).

Although all of this recent research indicates that the number of reported cases of anorexia nervosa and other eating disorders is substantially lower in non-whites, lesbians, and individuals from non-Western countries, there is a difference of opinion on what this implies about the actual incidence of eating disorders in these groups. Some researchers who have investigated anorexia nervosa in racial minorities have suggested that the disorder is linked more to socioeconomic class than to race and argue that the growing incidence of nonwhite anorectics reflects the growing economic prosperity of certain minority group members. These researchers argue that as non-whites become more prosperous, their exposure to white, middle-class beauty standards increases, thereby making nonwhites more vulnerable to anorexia and other eating disorders. Because fewer nonwhites than whites belong to the middle and upper economic classes, fewer nonwhites become anorexic (Pumariega et al. 1984; Andersen and Hay 1985; Robinson and Andersen 1985; White et al. 1985; Gray et al. 1987; Hsu 1987; Thomas and James 1988).

Other investigators, however, have exposed methodological and philosophical flaws behind this kind of argument. Some have suggested that the reason there are so relatively few nonwhite anorectics is because people of color do not have the same access to health-care facilities as whites. Because most of the studies of anorexia nervosa record only those sufferers who come to the attention of medical and psychiatric facilities, nonwhites who lack access to these facilities will not be acknowledged by health-care researchers. Moreover, even those minorities who do have access to medical care may feel threatened by a white-dominated medical profession and/or may be embarrassed to seek help for a mental health problem. Thus, the actual number of nonwhites with anorexia nervosa in the general population may be greater than indicated by case reports (Rosen et al. 1988; Root 1990; Dolan 1991; Smith and Krejci 1991; Thompson 1994).

In addition, some have argued that racial stereotypes about who is most vulnerable to anorexia nervosa can explain the apparent rarity of the disorder in minority groups. Anorexia nervosa is frequently referred to in medical and popular literature as a “Golden Girl’s Disease” that afflicts only young girls from white, Western European, privileged backgrounds (Root 1990). Consequently, this ethnocentric bias may lead medical personnel to misdiagnose or underdiagnose eating disorders in persons of color (Silber 1986; Hiebert et al. 1988; Rosen et al. 1988; Dolan 1991; Thompson 1994).

Even those who agree that minority group status may “protect” nonwhites from eating disorders also argue that this status does not necessarily protect specific individuals within these groups. As Maria Root (1990: 534) notes in a recent article on eating disorders in women of color: “Individuals within each racial/ethnic group are subject to the standards of the dominant culture, particularly when the culture-of-origin is devalued by the dominant culture.” Because thinness in Western and Westernized societies is associated with higher social class, and the attendant social power, resources, and opportunities, some individuals of color may see the pursuit of a slim body-type as a ticket to upward social mobility and acceptance by the dominant culture (Root 1990; see also Silber 1986 and Thompson 1994).

Whatever the explanation, the standard image of anorexia nervosa as a privileged white girl’s disease is increasingly being called into question. The disorder has been detected in a variety of racial, ethnic, and socioeconomic groups and in both Western and non-Western societies, although at the moment the number of cases among these groups appears to be relatively rare. Clearly, more research is needed before any definitive statements on the incidence and form of eating disorders in nonwhite and non-Western groups can be made.