Behavioral Disorders Affecting Food Intake: Eating Disorders and Other Psychiatric Conditions1

Behavioral Disorders Affecting Food Intake: Eating Disorders and Other Psychiatric Conditions1

Janelle W. Coughlin

Margaret Seide

Angela S. Guarda

Eating disorders are driven behavioral disorders resulting in significant functional impairment and, in extreme cases, death. They occur along a spectrum, so that diagnostic boundaries are often blurred. The fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (1) distinguishes three major categories of eating disorders: anorexia nervosa (AN), bulimia nervosa (BN), and eating disorder not otherwise specified (EDNOS).

Binge-eating disorder (BED) is usually accompanied by obesity and is subsumed under the EDNOS category. BED is currently under consideration as a distinct eating disorder in DSM-V and has become the focus of significant clinical and scientific attention.

Unlike BED, AN and BN are perhaps better thought of as “dieting disorders” (2). Both are characterized by an overvalued fear of fatness that drives a set of disturbed behaviors, including restricting food intake, binge eating, excessive exercise, self-induced vomiting, and abuse of laxatives, diuretics, and diet pills. Engagement in these behaviors, coupled with the physiologic consequences of starvation and/or the binge-purge-restrict cycle, sustains and heightens food preoccupation and body image disturbance. This chapter reviews the diagnosis, epidemiology, etiology, complications, and treatment of eating disorders and concludes with a summary of other psychiatric conditions and frequently prescribed psychotropic medications that may affect food intake. Although obesity is often the consequence of repetitive overeating or binge eating, it is primarily a medical condition and is addressed in chapters reserved solely for this topic.


Anorexia Nervosa

AN is a syndrome of self-starvation characterized by weight loss to a level below 85% of expected body weight. Weight loss is accompanied by fear of fatness and, in girls and women, amenorrhea or the absence of three or more consecutive menstrual cycles. AN is further subdivided into a restricting (AN-R) or a binge-eating/purging (AN-P) subtype. Individuals with AN-R restrict food intake and often excessively exercise and fidget in the service of weight loss but do not binge or engage in purging behaviors. By contrast, AN-P includes regular binge-eating and/or purging behaviors (e.g., self-induced vomiting and abuse of laxatives, diuretics, and enemas).

Bulimia Nervosa

BN is a dieting disorder characterized by episodes of binge eating followed by compensatory behaviors aimed at preventing weight gain. Binge eating is defined as consumption of an amount of food definitely larger than
most people would eat in a similar period, under similar circumstances, and is associated with a sense of loss of control over eating. Typical binge foods are high-fat, high-calorie, “forbidden” foods, and amounts consumed are 1000 to 2000 calories or more per binge (3). Between binges, bulimic individuals typically restrict intake and only consume “safe,” low-calorie, low-fat foods. Other compensatory behaviors following a binge can include purging, vomiting, abuse of laxatives or diuretics, or excessive exercise. As in AN, dieting and preoccupation with thinness develop into a consuming passion that is difficult to interrupt and impairs psychologic and social function. The distinction between AN-P and BN is primarily one of weight. Individuals who binge and purge but are less than 85% of ideal body weight or a body mass index (BMI) of about 17.5 and are amenorrheic are given the diagnosis of AN-P, whereas those who are less underweight, or are normal weight or overweight, are given the diagnosis of BN. The two subtypes of BN are purging (BN-P) and nonpurging (BN-NP). Individuals with BN-NP do not self-induce vomiting or abuse laxatives or diuretics; rather, they alternate episodes of binge eating with fasting or excessive exercise to avoid weight gain.

Eating Disorder Not Otherwise Specified

EDNOS is a heterogeneous diagnostic category. It includes partial-syndrome cases of AN and BN, BED, and atypical eating disorders. For partial-syndrome AN or BN, the diagnosis of EDNOS does not imply minor clinical significance. Indeed, these cases may be associated with morbidity equal to or greater than full-syndrome cases of AN or BN (4). An example would be an individual whose baseline weight was obese and who developed intense fear of fatness and extreme dieting behaviors, rapidly losing more than 40% of his or her body weight yet failing to meet the underweight criterion for AN or the binge frequency criterion for BN.

BED is defined as regular binge eating, twice a week or more, associated with a subjective sense of loss of control over eating but lacking the compensatory behaviors typical of BN. BED differs from BN in several additional ways. Individuals with BN restrict their food choices and calorie intake when not bingeing yet often are more impulsive and consume more calories during binges than do individuals with BED. Patients with BED overeat more consistently throughout the day than do patients with BN (5) and are more likely to be overweight or obese.

Examples of atypical eating disorders include globus hystericus, or fear of swallowing, resulting in severe weight loss and functional impairment, and psychogenic vomiting syndromes. In some cases, these may be factitious disorders, conditions in which the behavior persists in part because the sick role has become rewarding to the affected individual.


Epidemiologic data on eating disorders is limited for several reasons. Both AN and BN have relatively low prevalence in the general population. Furthermore, most patients are ambivalent about seeking treatment and minimize their symptoms. A minority of cases reaches clinical attention, so research on clinical samples inevitably underestimates the true incidence of these psychiatric conditions (6).

The prevalence of AN among young women is approximately 0.3%, with girls and women 10 times more likely to develop AN than boys and men (6). The age-and sex-adjusted incidence in the general population is approximately 8 cases per 100,000 population per year. Across the life span, AN is most likely to have its onset among girls and women ages 15 to 19 years, who comprise an estimated 40% of documented new cases. In one epidemiologic sample, incidence rates in this age group increased steadily from 1935 to 1989 (7). It is unclear how much this increase reflects better detection and increased care seeking as awareness of the diagnosis among both clinicians and the general public has increased.

Incidence rates of BN are consistently higher than those reported for AN across studies, approximating 12 cases per 100,000 population per year (6). These are likely to be underestimates because of the more secretive nature of this disorder and because affected individuals lack the starved habitus that makes AN easier to detect. The higher incidence and prevalence of BN are also partly explained by research suggesting that as many as 40% of patients with AN progress to BN over time as a result of the challenges of maintaining a low body weight through primarily restrictive behaviors (8). In comparison with AN, the age of onset in BN is later, with 20- to 24-year-old women being at greatest risk. Prevalence estimates for BN are 1% for girls and women and 0.1% for boys and men, the same gender distribution found in AN (6).

EDNOS is a phenomenologically heterogeneous group, and epidemiologic information on it is scant at best, although the prevalence of partial-syndrome eating disorders is at least twice that of full-syndrome eating disorders (9). Only three population studies of the prevalence of BED have been completed, and these reveal a prevalence of 2% to 3%, a more equal female-to-male distribution (approximately 2:1), and a later age of onset than that of AN or BN of 30 to 50 years (10). Rates of BED are much higher, on the order of approximately 25%, in clinical samples of obese individuals seeking weight-loss treatment (11).


Although knowledge of the pathogenesis of eating disorders remains limited, it is clear that the etiology of these conditions is multifactorial and in most cases includes the interaction of both genetic and environmental predisposing
factors. These interactions and how they contribute to risk remain largely unexplored and are believed to vary significantly among individuals.


Family, twin, and molecular studies suggest that eating disorders are genetically influenced. Cross-transmission of AN, BN, and EDNOS within families suggests a shared familial liability (12). Prevalence of eating disorders in relatives of eating-disordered probands is 7 to 12 times that of controls, and monozygotic twins have significantly higher concordance rates of AN and BN than their dizygotic counterparts (13); however, eating disorders do not necessarily breed true (i.e., relatives of probands with AN have increased rates of AN or BN, not increased rates of AN specifically). Twin studies have found heritability estimates of 58% to 76% for AN, 54% to 76% for BN, 38% to 61% for BED, and 32% to 72% for attitudes commonly associated with eating disorders (e.g., body dissatisfaction and weight preoccupation) (13, 14). Attempts to identify biologic markers for AN and BN have led some researchers to investigate polymorphisms in serotonin- and dopamine-related genes. Other studies have targeted leptin and estrogen receptors, genes involved in weight regulation, feeding, and energy expenditure (15). Although promising, research on these biologic markers has produced inconsistent findings and therefore awaits further investigation.


Research has identified several personality traits associated with eating disorders including elevated harm avoidance (15), neurotic personality features, and low self-esteem (16). Perfectionism, conscientiousness, persistence, and obsessive qualities are often discriminating features of AN, whereas elevated impulsivity, novelty seeking, negative emotionality, stress reactivity, and personality traits associated with antisocial, borderline, histrionic, and narcissistic personality disorders are more commonly associated with BN (15, 17). Family studies have found increased levels of some of these traits in first-degree relatives of individuals with eating disorders, a finding suggesting that the heritability of AN and BN may be related in part to the heritability of these personality characteristics (17).

Developmental Factors

Eating disorders are significantly more prevalent in menstruating girls and women than in prepubertal girls, a finding implicating a role for ovarian hormones and sexual development in the activation of disordered eating (18). Perception of being overweight prepubertally (19) and early-onset menarche (20) have emerged as specific aspects of puberty that may increase eating disorder vulnerability. Early-maturing girls have higher adiposity before menarche, are more dissatisfied with their bodies, and are more likely to engage in weight-loss efforts than girls who go through puberty on time or later in life (20). Environmental changes associated with the transition to college, including high levels of stress, performance and achievement demands, and role and identity changes, are factors significantly related to disordered eating (21) and may make this developmental milestone one that places late adolescents at risk of developing eating disorders. Past trauma, namely, childhood sexual abuse, may heighten the risk of developing eating disorders; however, early sexual abuse has been associated with other psychiatric conditions, thus making it difficult to determine whether a direct link between eating disorders and childhood sexual abuse exists or whether early sexual abuse and mental health are more broadly linked (22).

Sociocultural Factors

The sociocultural model of eating disorders posits that eating disorders and body image disturbances are the result of pervasive societal pressures on girls and women to be thin. According to this model, messages idealizing thinness are transmitted to members of society through the mass media, peers, and families. Although this model explains why eating disorders are more common in Western cultures that value thinness, only a small minority of the population develops eating disorders; therefore, sociocultural factors alone are not a sufficient explanation for the development of an eating disorder. However, pressures to be thin starting around the age of puberty may trigger the onset of dieting behavior in otherwise vulnerable individuals.

Mass Media

Slender female models and images (e.g., cartoons, computer graphics) saturate the Western mass media. Internalization, or acceptance, of these societal standards of thinness may lead to low self-esteem, negative affect, dieting, and/or eating disorders in girls and women (23). Experimental studies have consistently shown that girls and women exposed to media images of thinness experience greater body dissatisfaction in comparison with those exposed to heavier or neutral images (24). The negative effect of these images is heightened when girls and women viewing slender images have already internalized thin beauty ideals or have high baseline levels of body image disturbance.


Pressure to be thin from peers (25) and past history of weight-related teasing (26) may impact body dissatisfaction among girls and women and may increase the risk of disordered eating behaviors; however, at least one study reported that weight-related teasing does not predict body dissatisfaction in adolescent girls (27). Because most studies of teasing are retrospective, they are also affected by recall bias, which may be stronger in individuals with body dissatisfaction who are at risk of eating disorders.

Similarly, although members of adolescent female cliques often have comparable levels of body image disturbance, it is unclear whether body image is directly influenced by peers or whether adolescents simply seek homogeneous peer groups. Among college sorority sisters, bingeing and purging behaviors are passed from one individual to another, much like the spreading of a disease (a contagion effect), a finding suggesting a direct social influence of peers on disordered eating (28).


Parents are the most dominant sociocultural factor affecting young children, and parents’ direct comments about their child’s weight, particularly comments of mothers, have been identified as the most consistent factor associated with children’s concerns and behaviors related to weight and shape (29, 30). Familial dynamics are also important predisposing factors in eating disorders. Girls who eat alone, who have parents who are not married (31), or who perceive their family communication, parental caring, and parental expectations as low (32) are at increased risk of disordered eating. Indeed, investigators have suggested that low perceived social support from family, coupled with low self-esteem, high body concern, and use of escape-avoidance coping, places women at high risk of developing eating disorders (33).

Jul 27, 2016 | Posted by in PUBLIC HEALTH AND EPIDEMIOLOGY | Comments Off on Behavioral Disorders Affecting Food Intake: Eating Disorders and Other Psychiatric Conditions1

Full access? Get Clinical Tree

Get Clinical Tree app for offline access