Bulimia nervosa in adults: Clinical features and diagnosis

INTRODUCTION — 

Bulimia nervosa is characterized by recurrent episodes of binge eating and inappropriate compensatory behaviors, as well as frequent comorbid psychopathology [1]. The illness was first distinguished from anorexia nervosa in 1979 [2].

This topic reviews the clinical features, assessment, diagnosis, and differential diagnosis of bulimia nervosa. The medical complications of bulimia nervosa and their management are discussed separately, as is treatment of bulimia nervosa.

●(See "Bulimia nervosa and binge eating disorder in adults: Medical complications and their management".)

●(See "Bulimia nervosa in adults: Cognitive-behavioral therapy (CBT)".)

●(See "Bulimia nervosa in adults: Pharmacotherapy".)

EPIDEMIOLOGY — 

Prevalence estimates of bulimia nervosa are likely low due to the tendency of some individuals to conceal their illness because of shame or perceived stigma [1].

Based upon pooled results from international surveys of adults, the estimated 12-month prevalence of bulimia nervosa is 0.4 percent, and the lifetime prevalence is 1 percent [3]. Prevalence rates in the United States are nearly identical; a nationally representative study found a 12-month prevalence of 0.3 percent and a lifetime prevalence rate of 1 percent [4].

Bulimia nervosa is more common in females than males [1,3]. A meta-analysis of 33 international studies estimated that the lifetime prevalence of bulimia nervosa in females and males was 1.9 and 0.6 percent [5]. In the United States, nationally representative surveys also suggest that the lifetime prevalence of bulimia nervosa is at least three times higher in females [6], and one survey found it was five times higher [7]. In clinical settings, the ratio of females to males with a first-time diagnosis is much higher (13 to 1), as females may be more likely to present for treatment [8].

The median age of onset of bulimia nervosa is 16 years [7]. Premonitory symptoms for the initial onset of bulimia nervosa include overeating, body dissatisfaction, and fasting [9].

PATHOGENESIS — 

The pathogenesis of bulimia nervosa is not known, and studies of the pathogenesis often use a cross-sectional design, making it unclear whether abnormalities (eg, neuroimaging findings) represent etiologic causes, sequelae, neither, or both causes and sequelae. Nevertheless, research suggests that biologic and social/interpersonal factors may be involved.

Factors that may contribute to the onset of bulimia nervosa include the following:

●Genetic – Multiple genes with small effects seem to be involved [10]. One review of twin studies estimated that genetic factors explained 28 to 83 percent of the phenotypic variation for bulimia nervosa (heritability) [11]. A national registry study of sisters (n>780,000) found that the disorder was a heritable trait, such that 41 percent of the variance in bulimia nervosa was due to genetic factors and the remaining 59 percent to unique (nonshared) environmental factors [12].

●Childhood adversity – An established risk factor for onset of bulimia nervosa is childhood adversity and maltreatment. This includes parental psychopathology during childhood and family disruption (child not living with both parents/caregivers) [13], as well as emotional neglect and abuse, exposure to intimate partner violence, harsh physical punishment, and sexual abuse [14,15]. A meta-analysis of 30 studies found that bulimia nervosa was two to three times more likely to occur among individuals with a history of childhood maltreatment [16].

●Neurobiologic factors – Magnetic resonance imaging (MRI) studies demonstrate altered brain structure and function in bulimia nervosa [17-19]. As an example, structural differences between patients with bulimia nervosa and healthy controls include reduction of temporal and parietal cortical areas in patients [20,21]. One study found that reductions were greater in patients with more episodes of binge eating and purging, more preoccupation with shape and weight, and a longer duration of illness; in addition, reduction of cerebral surfaces was associated with poorer neuropsychologic functioning [21]. Subcortical structures in patients also exhibit abnormalities [22].

CLINICAL FEATURES — 

The core features of bulimia nervosa are binge eating and inappropriate compensatory behaviors to prevent weight gain [1].

Binge eating — Binge eating is defined by the following characteristics:

●Excessive amount of food – Binge eating is characterized by consuming an excessive amount of food in a discrete period of time that is definitely larger than most people would eat under similar circumstances; objectively, an eating binge is equivalent to at least two meals or at least 2000 kcal [23,24].

The clinician should consider the context of a meal in deciding whether the patient is binge eating. As an example, an amount of food that is excessive for a typical meal may be regarded as normal during a celebration or holiday [1].

●Loss of control – During binge eating episodes, patients feel they have no control over their eating and cannot stop eating once they have started; the episode often continues until the patient is uncomfortable or painfully full [1,23]. In addition, the binge may occur in a dissociative state.

However, loss of control is not absolute. Patients are usually ashamed of their behavior, try to conceal it, and will usually stop binge eating if someone unexpectedly enters the room. Resignation and abandoned efforts to control uncontrolled eating also qualify as loss of control.

●Discrete episodes – An episode of binge eating occurs during a limited time, such as two hours [1]. However, the episode is not necessarily restricted to one setting; as an example, the patient may begin eating at a restaurant and then continue at home. In addition, binge eating may be planned. Snacking on small amounts of food throughout the day is not considered binge eating.

●Generally avoided foods – During an episode of binge eating, patients frequently consume foods that are otherwise avoided, such as high-calorie sweets. However, the type of food eaten during an eating binge varies across episodes for a given patient and across patients. In addition, patients do not appear to crave any specific nutrient [1].

Inappropriate compensatory reactions — Patients with bulimia nervosa attempt to prevent weight gain through inappropriate compensatory behaviors [1,24].

Many patients employ multiple methods; the most common is self-induced vomiting, which may become an end in itself and occur after eating only small amounts of food. Vomiting is generally induced by using the fingers or instruments (eg, toothbrush) to stimulate the gag reflex; some patients become so adept that they can vomit at will. Ipecac syrup (emetine) is rarely used to induce vomiting; chronic use of ipecac may cause cardiac and skeletal myopathy.

Other inappropriate compensatory behaviors to prevent weight gain include fasting for a day or longer, as well as misuse of laxatives, diuretics, other medications (eg, thyroid hormone), or enemas [1]. Patients with diabetes may omit insulin doses [1,25]. In addition, patients may exercise to the point that it interferes with important activities or occurs at inappropriate times, in inappropriate settings, or despite injury [1].

Medical complications of self-induced vomiting or misuse of laxatives, diuretics, or enemas (purging) are discussed separately. (See "Bulimia nervosa and binge eating disorder in adults: Medical complications and their management", section on 'Medical complications of bulimia nervosa'.)

Prodromal symptoms — Prodromal cognitive symptoms of bulimia nervosa usually precede prodromal behavioral symptoms [26]. The cognitive symptom that most commonly emerges first is overvaluation of body weight and shape. Subsequently, patients can develop fear of gaining weight or feeling fat. The first prodromal behavioral symptom of bulimia nervosa that usually emerges is inappropriate compensatory behavior (eg, self-induced vomiting). Subsequently, patients develop binge eating.

Sequence of maladaptive behaviors — The prototypic sequence of maladaptive behavior in bulimia nervosa consists of [27]:

●Caloric restriction

●Binge eating

●Self-induced vomiting

The binge eating episode triggers further caloric restriction to influence body weight or shape [1], which leads to intense hunger and increases the probability of an additional binge eating episode. This pattern of behavior is known as the Restriction Model [27]. As part of the dietary restraint that occurs between eating binges, patients eat low-calorie food and avoid food that is perceived to be fattening or likely to trigger a binge [1].

Other antecedents to binge eating and purging include dysphoria and interpersonal stressors (Dual Pathway Model) [28-30]. Thus, treatment targets negative affect and stressors as a precipitant for binge-purge behavior [31]. Dysphoria also occurs after a binge eating episode, along with remorse and self-loathing.

Associated features — Associated features of bulimia nervosa include the following:

●Normal body weight – Body weight in bulimia nervosa is usually within or above the normal range [1]. Although bulimia nervosa can include fears of weight gain and becoming fat, the weight of most patients over time is normal. A 22-year, prospective observational study of patients with bulimia nervosa found that in those who either recovered (n = 52) or remained ill (n = 13), weight was normal in 67 and 74 percent [32]. The remaining patients were overweight or obese.

●Impaired cognitive functioning – Impairment of neurocognitive functioning is common in bulimia nervosa. As an example, patients can manifest deficits in executive functions, which include higher level abilities or processes such as inhibitory control, problem-solving, planning, and decision-making. Meta-analyses studies that compared patients with controls found that executive functioning was moderately worse in those with bulimia nervosa [33,34].

●Emotional dysregulation – In addition, emotional dysregulation (eg, loss of behavioral control secondary to negative emotions) often occurs in those with bulimia nervosa [35,36], and bulimia nervosa is usually accompanied by additional psychopathology. (See 'Comorbidity' below.)

●Psychosocial functioning – Psychosocial functioning is typically impaired in bulimia nervosa. A nationally representative study found that among individuals with the disorder, impaired functioning (serious problems with normal daily activities, interpersonal relationships, or fulfilling responsibilities) was present in 61 percent [4].

Nonsuicidal self-injury — Impulsive or compulsive nonsuicidal self-injury (eg, skin cutting, picking, or burning with a lit cigarette) is often seen in bulimia nervosa. Meta-analyses (13 and 20 observational studies) indicate that the prevalence of nonsuicidal self-injury is approximately 33 to 44 percent [37,38]. However, heterogeneity across studies is large.

Information about nonsuicidal self-injury, including its clinical features, assessment, and treatment are discussed separately in the context of youth. (See "Nonsuicidal self-injury in children and adolescents: Clinical features and proposed diagnostic criteria" and "Nonsuicidal self-injury in children and adolescents: Assessment" and "Nonsuicidal self-injury in children and adolescents: Prevention and choosing treatment".)

Suicide — Bulimia nervosa is often accompanied by suicidality, including suicidal ideation, action to prepare for an attempt, nonfatal attempt or self-harm, or death; in addition, the rate of attempts and deaths is elevated, compared with the general population [1,39].

●Suicidal ideation – The prevalence of suicidal ideation in bulimia nervosa is 60 percent, based upon the pooled results from six studies [38].

●Suicide attempts – Meta-analyses (17 and 36 observational studies) in patients with bulimia nervosa indicate that suicide attempts occur in approximately 20 to 25 percent [38,40]. Compared with the general population, individuals with bulimia nervosa are two to six times more likely to attempt suicide [4,41,42].

Factors that are associated with suicide attempts in bulimia nervosa include psychiatric comorbidity (eg, depression, personality disorders, or substance use disorders), a history of hospitalization, a history of self-mutilating behavior (nonsuicidal self-injury), longer duration of bulimia nervosa, poor impulse regulation, and having a sibling with an eating disorder [41-51].

●Suicide deaths – One retrospective study of 906 patients with bulimia nervosa, followed for a mean of 18 years, found that 1 percent died by suicide [52]. Across multiple studies, suicide was approximately four to seven times greater in those with bulimia nervosa than the general population [41,52-54].

However, suicide deaths in bulimia nervosa may be attributable in part to comorbid psychopathology, such as depression [41]. In a registry study, the rate of suicide in bulimia nervosa was approximately one-third of that observed in depressive disorders [53].

Other phenomena — Other clinical phenomena may occur in patients with bulimia nervosa, including dissociative symptoms. In a meta-analysis of eight studies that included 353 patients with bulimia nervosa, the results indicated that symptoms such as absorption, amnesia, depersonalization, and/or derealization were common [55].

In addition, bulimia nervosa may be associated with an increased risk of legal problems. A national registry study followed patients with bulimia nervosa for up to 20 years, as well as the general population [56]. After adjustment for potential confounders such as comorbid psychopathology, the risk of convictions for theft was nearly three times greater in patients than the general population, and the risk of being convicted for other crimes was one and a half to two times greater in patients. Legal issues may create additional stress that interferes with recovery from bulimia nervosa and increases relapse rates.

COMORBIDITY

Psychiatric disorders — Comorbid psychopathology is common in bulimia nervosa [1], such that the prevalence of psychiatric comorbidities in patients with bulimia nervosa exceeds the rate in the general population [57]. A nationally representative survey in the United States estimated that 94 percent of patients with bulimia nervosa had a lifetime history of at least one additional psychiatric disorder (aside from personality disorders) and that the mean number of comorbid disorders was two [57]. Among individuals with no history of an eating disorder, only 58 percent met criteria for one or more lifetime psychiatric disorders.

Every comorbid psychiatric disorder that is present in patients with bulimia nervosa should be diagnosed because of implications for treatment. Comorbidities are associated with a poorer prognosis for bulimia nervosa; however, successful treatment of bulimia nervosa often resolves comorbidity such as anxiety and depressive disorders [58,59].

A history of childhood maltreatment is associated with both onset of bulimia nervosa and with comorbid depressive disorders, anxiety disorders, and borderline personality disorder [60]. (See 'Pathogenesis' above.)

Based upon nationally representative surveys in the United States, the lifetime prevalence of specific comorbid disorders in bulimia nervosa include [57,61]:

●Depressive disorders

•Unipolar major depression – 76 percent of patients with bulimia nervosa

•Persistent depressive disorder (dysthymia) – 35 percent

●Anxiety disorders

•Generalized anxiety disorder – 26 percent

•Panic disorder – 18 percent

•Specific phobia – 14 percent

•Social anxiety disorder (social phobia) – 14 percent

●Substance-related disorders

•Alcohol use disorder – 61 percent

•Nicotine use disorder – 43 percent

•Other drug use disorder – 30 percent

●Posttraumatic stress disorder – 32 percent

●Premenstrual dysphoric disorder – 17 percent

●Personality disorders and traits

•Borderline (48 percent of patients with bulimia nervosa)

•Schizotypal (25 percent)

•Antisocial (11 percent)

Some patients meet criteria for more than one personality disorder.

In addition, patients with bulimia nervosa may have comorbid personality pathology that is not severe enough to meet criteria for a personality disorder but nevertheless causes distress and impairs functioning. This psychopathology includes such traits as [62]:

•Impulsivity – Acting without thought of consequences, failing to resist a drive or temptation that is harmful

•Perfectionism – Pursuing unrealistically high standards despite the occurrence of adverse consequences

•Compulsivity – Need for order, symmetry, exactness, and control

•Narcissism – Excessive concern with physical appearance, need for admiration and external validation from others

Multiple traits are often found in the same patient with bulimia nervosa [63]. One model hypothesizes that personality traits form three distinct patterns of comorbid personality pathology in patients with bulimia nervosa, consisting of emotional dysregulation and impulsivity, emotional inhibition and interpersonal avoidance, and perfectionism and achievement [64].

The prognostic significance of comorbid personality pathology is not clear [62,65]. Some studies have found that bulimia nervosa patients with comorbid personality disorders or traits have a worse outcome compared with patients without personality pathology [58,64,66,67], whereas other studies have found outcomes were comparable [68,69].

General medical disorders — Patients with bulimia nervosa are at risk for general medical complications of their disorder. (See "Bulimia nervosa and binge eating disorder in adults: Medical complications and their management", section on 'Medical complications of bulimia nervosa'.)

COURSE OF ILLNESS — 

The course of illness in bulimia nervosa may be intermittent or chronic [1].

●Duration of illness and recovery – It appears that in most patients, bulimia nervosa persists for at least 12 months. One study of a nationally representative sample from the United States found that among individuals with a lifetime history of bulimia nervosa (n = 92), the disorder persisted for at least 12 months in 55 percent [4].

Prospective and retrospective studies of patients with bulimia nervosa (n = 76 to 1352), with follow-up lasting 6 to 22 years, indicate that approximately 50 to 70 percent eventually improve or recover [70-73].

The likelihood of recovery from bulimia nervosa declines the longer one remains ill [72,74]. As an example, one study found that if patients did not recover within the first decade of follow-up, they were less likely to recover in the second decade of follow-up [73]. In addition, comorbid substance use disorder may be associated with decreased rates of recovery [72], whereas less severe symptoms of bulimia nervosa may be associated with increased rates of recovery [70].

Recovery in females (who compose the large majority of patients with bulimia nervosa in follow-up studies) and males appears to be comparable [75].

●Relapse – Based upon prospective observational studies lasting between 6 months and 22 years, remission of bulimia nervosa is followed by relapse in approximately 15 to 30 percent of patients [70,73,76]. One predictor of relapse is greater symptom severity at baseline (prior to treatment and recovery), including greater frequency of binge eating, self-induced vomiting, and body avoidance behaviors (eg, avoiding either tight-fitting clothing or looking in mirrors) [70,76].

ASSESSMENT

Components of the evaluation — The initial clinical evaluation of patients with a possible diagnosis of bulimia nervosa includes a psychiatric and general medical history, mental status and physical examination, and focused laboratory tests [31,77-79]. At a minimum, we recommended serum electrolytes and renal function tests, and an electrocardiogram [31].

The initial assessment should address the clinical features, diagnostic criteria, differential diagnosis, and comorbid psychopathology of bulimia nervosa, and may require multiple interviews to establish the diagnosis [31]. Subsequently, the clinician should monitor the patient’s response to treatment and assess the patient for new illnesses.

Clinical interview — In patients with a suspected diagnosis of bulimia nervosa, such as those who present with signs or symptoms of the disorder, administering a self-report screening instrument prior to the clinical interview can facilitate the assessment [31,80]. Alternatively, it is reasonable to dispense with screening and begin the assessment by inquiring about the diagnostic criteria for bulimia nervosa [31].

Among the available self-report measures, we suggest the SCOFF, which is a five-item self-report measure that screens for eating disorders [81,82]. Patients who screen positive for an eating disorder with the SCOFF should be asked about the diagnostic criteria for bulimia nervosa. Additional information about the SCOFF is discussed separately. (See "Eating disorders: Overview of epidemiology, clinical features, and diagnosis", section on 'Screening'.)

An alternative, longer self-report measure is the Primary Care Evaluation of Mental Disorders Patient Health Questionnaire (table 1) [83]. Whereas most screening instruments require a follow-up interview to make the diagnosis, the Patient Health Questionnaire both screens for and provides a diagnosis for bulimia nervosa, as well as depressive, anxiety, and alcohol disorders. It was specifically designed for use in primary care, has good diagnostic validity overall (sensitivity 75 percent, specificity 90 percent), excellent diagnostic validity for bulimia nervosa (sensitivity 89 percent, specificity 96 percent), and the median clinician time to review the results is one to two minutes.

Establishing rapport with the patient helps elicit information. Patients are often ashamed of the disorder and may attempt to hide it from clinicians [1]. At the beginning of the interview, simple questions about the patient’s age, occupation, marital status, and referral source can put the patient at ease. In addition, directed and focused questions about eating disorder psychopathology may be helpful for patients who do not respond to open-ended questions [39]. However, it is important to avoid undermining the interview by asking direct questions prematurely and to allow the patient to describe their illness from their own point of view. Conveying a nonjudgmental and knowledgeable attitude is also helpful. Collateral information should be sought from family members, who may report the presence of vomitus in the house, finding hidden laxatives or diuretics, or the patient regularly departing to the bathroom immediately after meals.

The assessment of the psychiatric history and mental status should include questions about [31]:

●Height and weight.

•Current – Clinicians should measure height and weight, as well as ask about them, because the patient’s self-report is often unreliable. Body mass index (calculator 1) and percent ideal body weight (table 2) should then be determined for comparison to population norms.

•Lifetime highest and lowest weights at adult height, and patterns of weight fluctuation.

•Desired weight.

●Frequency of self-weighing.

●Food intake patterns when patients are not binge eating.

●Present and past eating disorder symptoms:

•Restrictive eating.

•Food avoidance.

•Binge eating episodes, including frequency, types and amounts of food eaten, and whether patients feel they cannot control their eating.

•Inappropriate compensatory behaviors such as purging (eg, self-induced vomiting or misuse of laxatives, enemas, diuretics), misuse of other medications such as diet pills or thyroid hormone, prolonged fasting, excessive exercise, and chewing and spitting out food.

•Attitudes about and amount of time preoccupied with body weight and shape, food, and eating.

•Ritualistic eating behaviors (eg, cutting food into extremely small pieces or refusing to mix different types or colors of food on the plate).

●Self-esteem and the degree to which it depends upon body weight and shape.

●Current and past suicidal ideation and behavior – Bulimia nervosa is often accompanied by suicidality, including suicidal ideation, action to prepare for an attempt, and nonfatal attempt or self-harm. (See 'Suicide' above.)

Patients with suicidality should be seen more frequently and perhaps hospitalized, depending upon the estimated level of risk (table 3) [31]. A specific suicide plan of high lethality or intent indicates the need for hospitalization. Risk factors, evaluation, and management of suicidality are discussed separately. (See "Suicidal ideation and behavior in adults".)

●Comorbid disorders, such as anxiety, depressive, impulse control, substance-related, and personality disorders. (See 'Psychiatric disorders' above.)

●Psychosocial functioning.

●Prior treatment for an eating disorder and response:

•Psychotherapy

•Pharmacotherapy

•Self-help groups

●Family history of eating disorders, other psychopathology, and general medical disorders (eg, diabetes mellitus and obesity).

In discussing onset of bulimia nervosa, the clinician should ask about interpersonal problems at that time [31]. However, family members should not be blamed for causing the illness; there is no evidence that families cause eating disorders.

Structured instruments — Structured, interviewer-administered instruments are available for diagnosing bulimia nervosa and other eating disorders, but are seldom used in routine clinical practice. A structured instrument enables the interviewer to clarify ambiguous or contradictory responses and may help differentiate the diagnosis of bulimia nervosa from anorexia nervosa. However, these instruments are labor intensive and generally reserved for specialized evaluation and treatment settings conducting research.

Nevertheless, for clinicians who wish to conduct a structured interview to diagnose patients, multiple options are available [84]. Previously, the gold-standard has been the Eating Disorder Examination [85-88]. Later measures have gained popularity, such as the Eating Disorder Assessment for the American Psychiatric Association's Diagnostic and Statistical Manual, Fifth Edition (DSM-5) [89]. Both instruments systematically assess each criterion for every eating disorder. Each instrument has good to excellent reliability and validity, which are mostly a function of interviewer training and experience.

An online, self-report instrument called the Eating Disorder Questionnaire-Online was developed for diagnosing eating disorders, but we recommend avoiding it because the psychometric properties are poor [90]. As an example, sensitivity for anorexia nervosa is low (0.4), as is positive predictive value for bulimia nervosa (0.5).

Measurement based care — We suggest that clinicians ask patients at each visit how many episodes of binge eating occurred each week since the last visit, and how often they resorted to inappropriate compensatory behavior (eg, purging) to prevent weight gain. Measurement based care can help identify nonresponders to treatment and detect residual symptoms.

Although self-report measures may also be used to assess change in symptoms over time [80], these instruments are generally not used in routine clinical care and are generally reserved for research settings, due to their length. Available instruments include the Bulimia Test-Revised (36 items) [91], Eating Disorder Inventory-3 (91 items) [92], Body Shape Questionnaire (34 items) [93], Three-Factor Eating Questionnaire (51 items) [94], Eating Disorders Questionnaire (108 items) [95], and the Eating Disorder Examination-Questionnaire (38 items) [96]. In addition, newer technologies such as computers, tablets, and smartphones can be used to assess patients via e-mail and text messaging [97].

DIAGNOSIS — 

The Diagnostic and Statistical Manual, Fifth Edition, Text Revision (DSM-5-TR) and the World Health Organization's International Classification of Diseases-11^th Revision (ICD-11) can each be used to diagnose bulimia nervosa. The criteria used by each nosology closely resemble each other.

●Diagnostic and Statistical Manual – A DSM-5-TR diagnosis of bulimia nervosa requires each of the following five criteria (table 4) [1]:

•Episodes of binge eating, which are defined as eating an unusually large amount of food in a discrete period of time (eg, ≤2 hours). In addition, patients experience loss of control over their eating during the episode.

•Inappropriate compensatory behavior to prevent weight gain, such as self-induced vomiting.

•Binge eating and inappropriate compensatory behaviors occur, on average, at least once a week for three months.

•The patient’s self-evaluation is unduly influenced by body shape and weight.

•The disturbance does not occur exclusively during episodes of anorexia nervosa (table 5). (See "Anorexia nervosa in adults: Clinical features, course of illness, assessment, and diagnosis", section on 'Diagnostic and Statistical Manual'.)

The key elements of bulimia nervosa are recurrent episodes of binge eating and inappropriate compensatory behavior to prevent weight gain, which are described in detail elsewhere in this topic. (See 'Binge eating' above and 'Inappropriate compensatory reactions' above.)

●International Classification of Diseases – The diagnostic criteria for bulimia nervosa according to the ICD-11 closely resemble the DSM-5-TR criteria [1,98]. An ICD-11 diagnosis of bulimia nervosa requires each of the following:

•Episodes of binge eating, which are characterized by eating more food than usual during a discrete period of time. Patients feel that they cannot control their eating during the episodes, which are frequent and recurrent (eg, at least once a week for one month or longer).

•Repeated, inappropriate compensatory behavior to prevent weight gain (eg, at least once a week for one month or longer). Examples include:

-Self-induced vomiting, typically within an hour of binge eating

-Fasting and strenuous exercise

-Misuse of diuretics and purgatives (laxatives or enemas)

•Excessive preoccupation with body weight and shape, which can manifest by repeatedly checking one’s weight with different scales or shape with tape measures.

•Marked distress regarding episodes of binge eating and inappropriate compensatory behavior, or clinically significant impairment of interpersonal or occupational functioning.

•Patient does not meet criteria for anorexia nervosa.

●Diagnostic stability – Although patients diagnosed with bulimia nervosa may subsequently develop symptoms that warrant changing the diagnosis to a different eating disorder (eg, anorexia nervosa or binge eating disorder), diagnostic stability is the rule [1,99]. As an example, a prospective observational study followed 128 patients with bulimia nervosa at intake for up to seven years and found that diagnostic crossover to anorexia nervosa occurred in 14 percent [100]. Patients who do crossover to anorexia nervosa commonly revert back to bulimia nervosa, and crossover between the two diagnoses may occur multiple times [1]. According to the criteria for the Diagnostic and Statistical Manual, the current diagnosis is based upon the patient’s clinical presentation for the preceding three months [1].

DIFFERENTIAL DIAGNOSIS — 

Some symptoms of bulimia nervosa (eg, hyperphagia) overlap with symptoms of other psychiatric and medical illnesses [1,101]. However, these other illnesses do not include inappropriate compensatory behaviors to prevent weight gain (eg, self-induced vomiting, misuse of medications, prolonged fasting, or excessive exercise) or excessive preoccupation with weight and body shape, which are characteristic for bulimia nervosa. Some of the illnesses that resemble bulimia nervosa can also occur in conjunction with bulimia nervosa (eg, unipolar major depression and borderline personality disorder). (See 'Comorbidity' above.)

Psychiatric disorders — Anorexia nervosa, binge eating disorder, major depressive disorder, and borderline personality disorder can resemble bulimia nervosa [1,102]. The greatest amount of overlap exists between anorexia nervosa and bulimia nervosa, and distinguishing the two is critical for treatment. Body weight in bulimia nervosa is either normal or high, whereas weight is low in anorexia nervosa.

●Anorexia nervosa – Anorexia nervosa and bulimia nervosa are both marked by overvaluation of body shape and weight [1]. In addition, binge eating and purging (ie, self-induced vomiting or the misuse of laxatives, diuretics, or enemas) occur as part of anorexia nervosa, binge eating/purging subtype. The key distinguishing feature of anorexia nervosa is an abnormally low body weight, as indicated by a body mass index <18.5 kg/m² (calculator 1). By contrast, patients with bulimia nervosa usually maintain a body weight at or above a minimally normal level. Secondarily, the severe weight loss that occurs in anorexia nervosa is accompanied by anatomic and physiologic sequelae that are not found in bulimia nervosa. (See "Anorexia nervosa in adults: Clinical features, course of illness, assessment, and diagnosis", section on 'Diagnosis' and "Anorexia nervosa in adults and adolescents: Medical complications and their management" and "Bulimia nervosa and binge eating disorder in adults: Medical complications and their management".)

Despite these distinctions, anorexia nervosa and bulimia nervosa are related disorders and many experts acknowledge that the line between the two diagnoses is sometimes not clear [103-106]. In addition, the diagnosis for patients can change from one disorder to the other. A seven-year observational study followed 88 patients with anorexia nervosa at study intake and found that at some point, 34 percent met criteria for bulimia nervosa [100]. Of the 128 patients with bulimia nervosa at study intake, 14 percent crossed over to a diagnosis of anorexia nervosa.

●Binge eating disorder – Episodes of binge eating occur in both bulimia nervosa and binge eating disorder [1]. The two disorders are distinguished by the recurrent inappropriate compensatory behaviors that occur as part of bulimia nervosa but are absent in binge eating disorder. In addition, patients with bulimia nervosa typically restrict their diet between binge eating episodes to influence body weight and shape, whereas patients with binge eating disorder do not. (See "Eating disorders: Overview of epidemiology, clinical features, and diagnosis", section on 'Binge eating disorder'.)

●Unipolar major depression – Overeating occurs in bulimia nervosa and often does in major depressive disorder with atypical features [1]. In addition, dysphoria and suicidal ideation and behavior are also frequent in both disorders. However, neither inappropriate compensatory behavior to prevent weight gain nor excessive preoccupation with weight and body shape is present in major depression. Conversely, patients with major depression frequently present with insomnia, anergia, psychomotor agitation or retardation, and anhedonia, which are not seen in bulimia nervosa. (See "Approach to the adult patient with suspected depression".)

●Borderline personality disorder – The impulsivity that characterizes borderline personality disorder can include binge eating; suicidality and disturbed identity are other features that overlap with bulimia nervosa [1]. However, the presence of inappropriate compensatory behavior indicates the diagnosis of bulimia nervosa. In addition, the identity disturbance of patients with bulimia nervosa involves weight and body shape; in borderline personality disorder, the disturbance is more likely to involve goals, values, and vocational aspirations. (See "Borderline personality disorder: Epidemiology, pathogenesis, clinical features, course, assessment, and diagnosis", section on 'Diagnosis'.)

General medical disorders — Hyperphagia can occur in medical disorders such as Prader-Willi syndrome and Klein-Levin syndrome [107,108]. Prader-Willi syndrome is a genetic disorder marked by hyperphagia, obsession with food, and obesity, as well as cognitive impairment (usually intellectual disability) and hypogonadism. Behavioral problems such as tantrums or oppositional behavior can also occur. Klein-Levin syndrome is a self-limited illness that usually affects adolescent males and is marked by increased appetite as well as hypersomnia and behavioral disturbances. Symptoms remit between episodes. Neither syndrome is accompanied by the inappropriate compensatory behaviors and over-concern with shape and weight seen in bulimia nervosa [1]. (See "Prader-Willi syndrome: Clinical features and diagnosis", section on 'Diagnosis'.)

PREGNANCY — 

A separate topic discusses pregnancy, delivery, and postpartum outcomes in the context of bulimia nervosa; the course of bulimia nervosa during and after pregnancy; and management of bulimia nervosa that is specific to pregnant women. (See "Eating disorders in pregnancy".)

SOCIETY GUIDELINE LINKS — 

Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Eating disorders".)

INFORMATION FOR PATIENTS — 

UpToDate offers two types of patient education materials, “The Basics” and “Beyond the Basics.” The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on “patient info” and the keyword(s) of interest.)

●Basics topic (see "Patient education: Bulimia nervosa (The Basics)")

Other useful resources for patients include material dispensed by the following organizations:

●American Psychiatric Association

●National Alliance for Eating Disorders

●National Eating Disorders Association

●National Institute of Mental Health

SUMMARY AND RECOMMENDATIONS

●Epidemiology – Based upon pooled results from international surveys of adults, the estimated 12-month prevalence of bulimia nervosa is 0.4 percent, and the lifetime prevalence is 1 percent. Prevalence rates in the United States are nearly identical. Bulimia nervosa is more common in females than males.

●Clinical features – The core clinical features of bulimia nervosa are:

•Binge eating – Eating an amount of food that is definitely larger than most people would eat under similar circumstances. Binge eating occurs in discrete episodes and is accompanied by a sense of loss of control.

•Inappropriate compensatory behavior to prevent weight gain (eg, self-induced vomiting).

The prototypic sequence of behavior in bulimia nervosa consists of caloric restriction, binge eating, and self-induced vomiting. Body weight is usually within the normal range. (See 'Clinical features' above.)

●Comorbidity – Patients with bulimia nervosa frequently have a lifetime history of at least one comorbid psychiatric disorder, including depressive disorders, anxiety disorders, substance-related disorders, and personality disorders. (See 'Comorbidity' above.)

In addition, patients are at increased risk for general medical complications. (See "Bulimia nervosa and binge eating disorder in adults: Medical complications and their management".)

●Course of illness – In most patients, bulimia nervosa persists for at least 12 months. Approximately 50 to 70 percent of patients eventually improve or recover; however, the likelihood of recovery declines the longer one remains ill. Among patients who remit, relapse occurs in approximately 20 to 30 percent. (See 'Course of illness' above.)

●Assessment – The assessment of the patient’s psychiatric history should include questions about height, weight, frequency of self-weighing, meal patterns, and present and past eating disorder symptoms (restrictive eating; binge eating; inappropriate compensatory behaviors such as purging [eg, self-induced vomiting], prolonged fasting, and excessive exercise; and attitudes about body weight and shape, food, and eating).

The evaluation should also address comorbid psychiatric disorders and suicidality, including suicidal ideation and nonfatal attempt or self-harm. Patients with suicidality should be seen more frequently and perhaps hospitalized, depending upon the estimated level of risk. A specific suicide plan of high lethality or intent indicates the need for hospitalization (table 3). (See 'Clinical interview' above.)

●Diagnosis – The diagnosis of bulimia nervosa according to the American Psychiatric Association’s Diagnostic and Statistical Manual, Fifth Edition, Text Revision, (DSM-5-TR) is summarized in a table (table 4). The diagnostic criteria of the International Classification of Diseases-11^th Revision closely resemble the DSM-5-TR criteria. (See 'Diagnosis' above.)

●Differential diagnosis – There is substantial overlap between anorexia nervosa and bulimia nervosa; however, body weight is usually low in anorexia nervosa and normal or high in bulimia nervosa. Other disorders in the differential diagnosis of bulimia nervosa, including binge eating disorder, unipolar major depression, and borderline personality disorder, do not present with inappropriate compensatory behavior to prevent weight gain or excessive preoccupation with weight and body shape. (See 'Differential diagnosis' above.)