Abstract
DSM-V describes three eating disorders (anorexia nervosa, bulimia nervosa, and binge eating disorder), three feeding disorders (avoidant/restrictive food intake disorder, pica, and rumination disorder), and two residual feeding and eating disorder categories (APA, 2013). Although these disorders contain some overlapping features, an individual can receive just one feeding or eating disorder diagnosis at a time. The only exception is pica, which can be diagnosed concurrently with another feeding or eating disorder if the pica behavior is severe enough to warrant additional clinical attention.
The Spectrum of Eating Disorders
Classification
DSM-5 describes three eating disorders (anorexia nervosa, bulimia nervosa, and binge-eating disorder), three feeding disorders (avoidant/restrictive food intake disorder, pica, and rumination disorder), and two residual feeding and eating disorder categories (APA, 2013). Although these disorders contain some overlapping features, an individual can receive just one feeding or eating disorder diagnosis at a time. The only exception is pica, which can be diagnosed concurrently with another feeding or eating disorder if the pica behavior is severe enough to warrant additional clinical attention.
Eating Disorders
The hallmark feature of anorexia nervosa (AN) is a significantly low body weight, accompanied by an intense fear of weight gain, and body image disturbance (for example, feeling “fat” while very thin, or an inability to recognize the self-relevance of health consequences associated with low weight). Although clinical judgment is necessary to determine whether low body weight reaches clinical significance, it would conventionally fall below the 18.5 kg/m2 lower limit of normal weight range for adults and below the fifth percentile for sex, height, and age for children. Individuals with AN may maintain their low weight by eating very little (AN restricting type) or by attempting to compensate for any episodes of actual or perceived overeating by engaging in purging behaviors such as self-induced vomiting or misuse of laxatives, diuretics, or other medications (AN binge-eating/purging type; APA, 2013).
In contrast, individuals with bulimia nervosa (BN) are typically normal-weight or overweight. BN is characterized by recurrent binge eating (i.e., consuming large amounts of food while feeling out of control) and inappropriate behaviors intended to compensate for the potential impact of these binge episodes on one’s body weight. Such compensatory behaviors can take the form of self-induced vomiting; misuse of laxatives, diuretics, or other medications; fasting; or excessive exercise. Binge eating and compensatory behaviors must occur, on average, at least once per week for at least three months before a diagnosis of BN can be conferred. Individuals with BN also base their self-worth primarily or even exclusively on their current body shape or weight, resulting in fluctuating or poor self-esteem (APA, 2013).
Binge-eating disorder (BED) is also characterized by recurrent binge eating, but in the absence of the compensatory behaviors that are the sine qua non of BN. The majority of individuals with BED are overweight or obese. Although many individuals with BED also overvalue their shape and weight, this is not a diagnostic criterion. Instead, the ascertainment of BED focuses on distinguishing binge eating from episodes of ordinary overeating. Specifically, to meet diagnostic criteria for BED, binge episodes must be characterized by three or more of the following key features: rapid eating, eating beyond satiety to the point of discomfort, eating alone due to guilt or shame, eating in the absence of hunger, and/or feeling dysphoria or self-reproach because of the overeating. Binge eating must be associated with significant distress, and occur, on average, at least once per week for at least three months, before a BED diagnosis can be made (APA, 2013).
Feeding Disorders
DSM-5 avoidant/restrictive food intake disorder (ARFID) is a revision of the DSM-IV category of feeding or eating disorder of infancy or early childhood, which has been expanded to account for similar presentations in adolescents and adults. Individuals with ARFID are unable to meet their nutritional needs as evidenced by low weight, nutritional deficiency, reliance on enteral feeding/nutritional supplements, and/or psychosocial impairment. In contrast to the body weight concerns observed in AN and BN, body shape or weight concerns are not present in ARFID. Instead, persons with ARFID typically exhibit a lack of interest in feeding, avoidance of food due to sensory features (e.g., texture, temperature), or a fear of a traumatic eating-related experience (e.g., choking, involuntary vomiting). To meet diagnostic criteria, food avoidance cannot be associated with a normative cultural practice (APA, 2013).
Individuals with pica engage in the persistent consumption of nonnutritive, non-food substances, such as paper, cloth, or dirt. A minimum duration of one month is required before the diagnosis can be made. Commonly ingested non-nutritive substances (e.g., low- or no-calorie foods) do not meet the non-food requirement. To qualify for a diagnosis of pica, the behavior cannot be part of normative development or a locally accepted cultural practice. DSM-5 therefore suggests a minimum age of two, as non-aberrant mouthing behaviors are common during infancy and toddlerhood (APA, 2013).
Rumination disorder is characterized by the persistent and effortless regurgitation of food over a period of at least one month. After regurgitating, individuals with rumination disorder typically re-chew, re-swallow, or spit out the previously ingested food. Although the regurgitation may be at least partially voluntary, it is distinct from the self-induced vomiting characteristic of BN, because the latter is intended to compensate for calories ingested during a binge-eating episode. It can also be distinguished (by physical examination or testing) from esophageal reflux and other gastrointestinal conditions (APA, 2013).
Other Feeding and Eating Problems
Some feeding or eating disturbances are associated with clinically significant distress and/or impairment but do not meet criteria for one of the specific diagnoses described above. This residual category was previously termed eating disorder, not otherwise specified (EDNOS) until the nomenclature changed in DSM-5. A new term, other specified feeding or eating disorder (OSFED), now captures these residual cases, and includes five example presentations. Atypical anorexia nervosa describes anorexic features at normal or above-normal weight. Bulimia nervosa (of low frequency and/or limited duration) and binge-eating disorder (of low frequency and/or limited duration) are appropriate diagnoses when individuals have engaged in relevant behaviors less frequently than once per week or for fewer than three months. Purging disorder describes a symptom profile that includes self-induced vomiting and/or misuse of laxatives, diuretics, or medications at least once per week for three months, in the absence of frank binge episodes. Night eating syndrome comprises either recurrent evening or nocturnal eating (e.g., a large amount of food consumed after dinner or eating after awaking from sleep) that results in distress or impairment. Lastly, if insufficient information is available to confer a definitive feeding or eating disorder diagnosis (e.g., in an emergency room setting) and/or the reasons that a clinical presentation does not meet diagnostic criteria for a feeding or eating disorder remain unspecified, a diagnosis of unspecified feeding or eating disorder (UFED) may be applied.
Epidemiology and Impact
Although eating disorders do affect boys and men, available data indicate that these disorders are significantly more common in girls and women. In a recent epidemiological study of women in the United Kingdom, 15.3 percent had experienced a DSM-5 eating disorder by the fourth or fifth decade of life, with the most common presentation being other specified feeding or eating disorder (Micali et al., 2017). Aggregated data from the United States and Western Europe indicate that, among young females, AN has a 12-month prevalence of 0.4 percent, while BN is two to three times more common, with a 12-month prevalence of 1.0–1.5 percent (APA, 2013). The sex ratio of BED is more balanced; with BED affecting 1.6 percent of females and 0.8 percent of males in a 12-month timeframe (APA, 2013). ARFID may be equally common in both females and males, but less is known about its prevalence in the general population. The prevalence of pica and rumination disorder is similarly unknown, but both disorders are thought to be more common among individuals with intellectual disabilities. Eating disorders also have global distribution, and although prevalence is highest across North America, Europe, Australia, and New Zealand, evidence supports that prevalence has been increasing more rapidly in regions in the Global South over the past decade (IHME, 2018).
Course, Mortality, Recovery
Course and Outcome
The long-term course and outcome of eating disorders is variable. Many individuals with an eating disorder will achieve full remission or substantial improvement, while others will have an illness course marked by chronicity, fluctuation in symptoms with periods of improvement punctuated by relapse, or even fatality. Although the course and outcome of eating disorders have been the focus of many longitudinal and treatment studies, interpretation of the literature is challenged by methodological variability across investigations and inconsistent operationalization of outcome. Taken together, extant work suggests the longitudinal course may be influenced by factors including age at onset, diagnosis, symptom presentation, treatment participation, and comorbidity, among others.
Recovery and Relapse
Anorexia nervosa. Anorexia nervosa (AN) typically onsets during adolescence or early adulthood. Considerable longitudinal research has described the long-term course and outcome of anorexia nervosa. Comprehensive reviews of outcome studies have suggested that half of surviving patients achieve full recovery, one third improve but continue to experience symptoms, and approximately 20 percent persist in chronic illness (Steinhausen, 2009). However, a more recent study showed more promising outcomes, with 62.8 percent of women with anorexia nervosa classified as recovered after an average of 22 years of study follow-up (Eddy et al., 2017). Younger age at illness onset was associated with improved outcomes, and indeed, research suggests that rates of recovery are generally higher among adolescents than adults. Taken together, data suggest that the majority of adolescents will achieve full recovery, depending on the definition of outcome used, with weight remission preceding the resolution of cognitive symptoms.
In addition to younger age, a shorter duration of illness and a milder illness (e.g., higher body weight, less entrenched cognitive symptoms) have been associated with improved outcomes. Lower body weight and the presence of bulimic symptoms may mark a more severe course. Whereas the subject of relapse has received less research attention, studies of clinical samples suggest that approximately one-third of patients treated for anorexia nervosa who recover will relapse.
Notably, understanding the course and outcome of AN comes primarily from studies of treatment-seeking individuals, which may reflect a selection bias. The findings from the National Comorbidity Survey Replication (NCS-R) suggested that spontaneous recovery from AN in non-treatment seeking individuals may occur often, although an understanding of factors contributing to this outcome is incomplete (Hudson et al., 2007).
Bulimia nervosa. Bulimia nervosa onsets in late adolescence or early adulthood. The longitudinal course of bulimia nervosa is also variable, most often marked by recovery or substantial improvement, but for some patients, it is characterized by chronicity or relapse. Reviews indicate 50–75 percent will achieve full recovery and that 20–25 percent will have a chronic protracted illness course (Steinhausen, 2009). For example, in a recent study, 68.2 percent of individuals with bulimia nervosa were classified as recovered after an average of 22 years of study follow-up (Eddy et al., 2017). As in the studies of anorexia nervosa, reconciliation of the findings in bulimia nervosa was challenged by variable definitions of recovery. Longer duration of illness, history of treatment refractoriness, comorbid impulsive-spectrum problems (e.g., substance abuse, borderline personality disorder) may be associated with poorer outcome. Approximately one-third of those who recover will relapse. Persistent cognitive eating disorder symptoms, including body image disturbance or overvaluation of weight and shape, may increase the likelihood of behavioral relapse.
Binge-eating disorder. Less research exists describing the longitudinal course and outcome of binge-eating disorder (BED). Community-based research suggests that the diagnosis is unstable over time, with some data indicating that over six months nearly half of those with BED no longer met full criteria for the disorder. Prognosis is generally positive, with other findings suggesting that fewer than one-fifth of those with BED still having clinically significant eating disorder symptoms by five years of follow-up. However, rates of concomitant obesity increase in these samples over time, which may be an important health outcome to assess in addition to the eating disorder. Relapse appears to be less likely for those with BED compared to those with anorexia nervosa or bulimia nervosa.
Mortality
Eating disorders are associated with an increased risk of premature death. In spite of diagnostic and treatment advances, the prognosis for adult anorexia nervosa does not seem to have improved during the twentieth century. Recent meta-analytic findings indicate standardized mortality ratios (SMRs) of 5.9 overall among patients with anorexia nervosa (Arcelus et al., 2011), and of 31.0 for death by suicide specifically (Preti et al., 2011). Factors noted to predict death in anorexia nervosa include lower body weight, longer duration of illness, comorbid substance use disorders, poorer social functioning, and history of suicidality. Death is often ascribed to medical complications of the illness, including cardiac failure, as well as suicide. SMRs among those with bulimia nervosa are estimated to be lower than those in anorexia nervosa at 1.9 overall (Arcelus et al., 2011) and 7.5 for suicide (Preti et al., 2011). Whereas mortality among other types of eating disorders has received less research attention, some data suggest SMR for those with low weight EDNOS may be comparable to that observed in anorexia nervosa. However, those with low weight EDNOS may constitute a heterogeneous group as other research (e.g., focused on non-fat phobic AN) has suggested prognosis may be more favorable (Becker, Thomas, & Pike, 2009). Across diagnoses, the prognosis is better for child and adolescent patients (Rome et al., 2003).
Diagnostic Crossover
The course of eating disorders is characterized by symptom fluctuation and, frequently, diagnostic crossover. Crossover between the anorexia nervosa subtypes, and from anorexia nervosa to bulimia nervosa, appears to be the most common. Approximately 50 percent of those with anorexia nervosa will develop regular binge eating and/or purging behaviors, crossing between the subtypes (e.g., from restricting type to binge/purge type) or to bulimia nervosa if the development of these symptoms is concomitant with weight gain. Research suggests that crossover from primary restriction to binge/purge symptoms most commonly occurs within the first three to five years of illness, although longitudinal data do suggest that crossover is frequent, bi-directional, and can occur even many years into illness (Eddy et al., 2008).
Crossover from bulimia nervosa to anorexia nervosa is less common and generally limited to individuals who had a previous history of anorexia nervosa. Further, diagnostic migration from binge-eating disorder to bulimia nervosa or anorexia nervosa appears to be rare, but prospective research in this area is limited.
Given that symptom fluctuation is common, achievement of symptomatic improvement – even in the absence of full remission – is common across eating disorder diagnoses. This can give the impression of crossover from anorexia nervosa, bulimia nervosa, and binge-eating disorder, to the range of OSFED presentations (e.g., those that narrowly miss full criteria for AN or BN). However, these transitions most often resemble the initial eating disorder diagnosis and are associated with improvements in psychosocial functioning, suggesting that they may be better conceptualized as a transition to a partially recovered state, rather than to a new eating disorder diagnosis (Eddy et al., 2010).
Feeding Disorders
Less is known about the course and outcome of the feeding disorders including pica, rumination disorder, and ARFID. Pica can onset in early childhood and also can be observed in school-age children as well as adults. Rumination disorder often begins during the first year of life and can occur in the context of developmental delays; however, it can be diagnosed across the lifespan and does also occur in typically developing individuals. Spontaneous remission occurs for many, but for those with co-occurring intellectual disabilities, a more protracted course may be expected (APA, 2013). ARFID is a newly revised, heterogeneous diagnosis and thus the long-term course and outcome have not yet been reported on. While some of those youth with selective eating or failure to thrive may receive early intervention and achieve healthy growth and development over time, others will persist in feeding or eating problems. There is some suggestion that those with early feeding problems may be at increased risk for the development of eating disorders over time, but further research about the longitudinal relationship between early feeding and prospective eating disorders is needed.
Differential Diagnosis
Differential diagnosis of a clinical presentation falling within the broad category of feeding and eating disorders is rendered especially challenging for several reasons. First, there is potentially moderate phenomenologic overlap in presentation across the diagnostic categories, ARFID, anorexia nervosa, bulimia nervosa, and binge-eating disorders, as well as other specified feeding and eating disorders (OSFED) and unspecified feeding and eating disorders (UFED). As noted previously, some patients migrate from one diagnostic category to another during their longitudinal course, and definitions of remission, partial remission, and recovery are not well-established empirically. Second, medical illness needs to be considered as potentially driving some or all of the signs and symptoms such as appetite, weight, and gastrointestinal discomfort or dysfunction, and then also rigorously evaluated, excluded, and/or managed when appropriate. Next, observed, collateral, and longitudinal data are often indispensable to establishing an eating disorder diagnosis, depending on the patient’s capacity for insight and willingness to share important diagnostic information during an initial clinical encounter. Limited insight especially characterizes child and adolescent patients who may not formulate their behaviors as motivated (e.g., by weight or shape concerns) or even as voluntary at all. Patients with an eating disorder are commonly reluctant to disclose symptoms to a clinician because of associated shame or stigma, or perceived secondary gains associated with the disorder. Finally, eating disorders are highly comorbid with other mental disorders – and particularly with mood and anxiety disorders – and symptoms and signs such as appetite and weight changes might be partially attributable to the comorbid disorder in some cases.
Although AN is characterized by low weight, intense fear of weight gain (and/or behaviors undermining appropriate weight gain), and body image disturbance, it overlaps phenomenologically with BN with respect to clinically significant concerns about weight and behaviors that prevent weight gain – or at least are intended to prevent it by compensating for calories consumed during a binge eating episode. Both patients with AN and BN can meet criteria for these respective disorders through dietary restriction, purging (vomiting), laxative use, and excessive physical activity, among other behaviors or a combination of these. For example, an estimated one half of individuals with AN engage in regular binge-eating and/or purging, and conversely, some individuals with BN engage in fasting and/or exercise to neutralize calories taken in during a binge, but do not induce vomiting or misuse laxatives. In this respect, a useful clinical sign distinguishing AN from BN is a significantly low weight. Likewise, because both AN and ARFID are characterized by low weight, the two can be difficult to distinguish clinically. Whereas the onset of ARFID, which can begin in infancy, tends to be younger than AN, which has its peak onset in adolescence, underweight adolescent patients who disavow an intense concern with fatness or weight gain may eventually meet diagnostic criteria for either disorder. A key distinguishing characteristic would be that a patient with AN may present as unable to recognize or accept the medical risks of the behavior, whereas a patient with ARFID will often be distressed by the behavior and may have more overt fears that underlie food avoidance. As noted previously, patients may not receive a concurrent diagnosis of more than one feeding or eating disorder, with the exception of pica. As a result, if criteria are met for both AN and BN, a diagnosis of AN is made. BN and binge-eating disorder are mutually exclusive; the latter is distinguished by the absence of routine compensatory behaviors to prevent weight gain after a binge.
Atypical depression with weight loss is generally distinguishable from AN by the actual loss of appetite; by contrast, individuals with AN typically override hunger in order to restrict their diet. In addition, whereas individuals with atypical depression commonly lose interest in food and eating, individuals with AN may show interest in preparing food for others and may experience eating or mealtimes as unpleasant or affectively charged. Body image disturbance and distress associated with AN and BN can be similar to the intense scrutiny and distress associated with an imagined defect in physical appearance as seen in body dysmorphic disorder. When the attention is focused on weight or shape and other associated criteria are met, a diagnosis of either AN or BN should be made. Although the body image disturbance associated with AN can result in distortions that can present in similar ways as do somatic delusions, they do not typically warrant a separate diagnosis of delusional disorder.
Food-related anxiety and/or aversions to particular tastes/textures/temperatures common to ARFID may also be present in certain specific phobias or in those with autism spectrum disorders. If the food avoidance behaviors are sufficiently significant to warrant independent treatment, an ARFID diagnosis is appropriate.
Differentiating an Eating Disorder from a Medical Disorder
Patients presenting with failure to gain weight or new-onset weight loss should be evaluated for medical causes or contributions to weight or appetite loss, hypermetabolic states, or failure to thrive, as well as for an eating disorder. Occult malignancies, thyroid and other autoimmune disease, diabetes, infectious and parasitic diseases, and gastrointestinal disease should be considered in the differential diagnosis of weight loss. Anorexia nervosa not infrequently follows a medically precipitated weight loss, so the course and its relation to known medical illness may yield important clinical data that would lead a clinician to consider anorexia nervosa in the differential diagnosis. Weight and shape concerns are prevalent among adolescent and adult women, so clinicians should take care not to prematurely attribute profound weight changes solely to dietary restriction or exercise. Likewise, clinicians should not prematurely exclude eating pathology based on patient self-report alone, but rather should seek corroborating diagnostic data through judicious deployment of appropriate medical diagnostic testing, clinical records detailing past medical and psychiatric history as well as treatment, and should systematically aim to reconcile any incongruities among clinical history, patient report, physical exam, laboratory tests, and observed behavior. If nutritional rehabilitation is indicated, clinicians should assess whether the patient is able to progress with weight gain as expected (for example, outpatient treatments typically aim for a weight gain of 1–2 lbs/week), and if not, whether and how these longitudinal data are informative to the diagnosis. If a patient chooses either to conceal or not disclose weight concerns, dietary restriction, binge eating, purging, and other behaviors preventing weight gain, an eating disorder may be a diagnosis of exclusion after medical causes are evaluated and ruled out and the longitudinal course supports that the patient is undermining weight gain. Sometimes laboratory tests can be informative when they are suggestive of chronic vomiting or laxative abuse (e.g., if there is an unexplained hypokalemia or hyperamylasemia) but they are neither sensitive nor specific for an eating disorder. Bulimia nervosa commonly presents without any physical abnormalities on examination, so is an even more difficult diagnosis to establish if a patient does not disclose symptoms. That said, a dental exam revealing a wear pattern consistent with chronic exposure to vomitus and Russell’s sign (excoriation on the back of the hand from trauma after chronic scraping against the incisors) are suggestive of BN, but do not offer a highly sensitive test for BN. Their prevalence is unknown; the former is not readily visible in the context of a mental health examination and the latter is relatively uncommon. Finally, patients with BED are frequently overweight or obese, but rarely have non-weight related medical complications related to binge eating per se.
Differentiating an Eating Disorder from Normative Behaviors
For some underweight and overweight patients, the diagnostic criteria related to cognitive or affective symptoms (e.g., intense fear of weight gain, marked distress following a binge eating episode) must be assessed against the local social context and associated norms governing diet, exercise, and weight management behavior. Some patients will present their behaviors as motivated by a desire to achieve a “healthy lifestyle” or become more competitive athletically (e.g., through implementing rigid dietary rules or a rigorous exercise regimen). Under such scenarios, clinicians need to discern whether the behaviors have resulted in clinically significant distress or impairment or carry medical risks or result in complications that the patient is unable to recognize or accept.
Evaluation
Evaluation of a known eating disorder will almost always comprise psychological/psychiatric, medical, and nutritional dimensions. Indeed, since signs can potentially first manifest as psychological, medical, or nutritional or, in some cases, can be absent or subtle, primary care clinicians, medical sub-specialists (including gastroenterologists and cardiologists), or general psychiatrists may be the first to detect an eating disorder. This chapter focuses on the psychiatric or psychological component of the evaluation and we emphasize that the psychiatrist or psychologist is often the clinician who coordinates team communications, referrals, and the management plan and thus should have an excellent grasp of the spectrum of clinical evaluation and care required.
The evaluation of a patient with an eating disorder will be very different for a patient who has – versus one who has not – acknowledged symptoms and accepts the importance of providing a candid history in order to evaluate their clinical significance. If an eating disorder is unconfirmed, but suspected because of the history, the initial evaluation will focus on ascertaining and piecing together relevant clinical and collateral data. These sources include history of present illness and presenting symptoms, past medical and psychiatric history, physical and mental status examination, and laboratory and/or findings from other diagnostic testing, as well as collateral sources of information from family members, teachers, and chart review with particular attention to unexplained changes in weight, diet, and possible physiologic sequelae of chronic purging (e.g., perimolysis, hypokalemia, hyperamylasemia or poor nutrition). For children and adolescents in particular, who may have limited insight into their motivations or behaviors due to cognitive maturity and therefore find it challenging to self-report on symptoms, collateral assessment of parents or other caretakers is critical.
An eating disorder will often first come to light in a primary or specialty medical care setting. That being said, the diagnosis is easily missed as well. Studies show that up to half of eating disorder cases are not recognized in primary care settings. Moreover, an estimated one half of individuals with an eating disorder have not received specialty care for this problem. In primary care settings, patients may not manifest any clinical signs of an eating disorder, except low weight, in the case of anorexia nervosa. Further, unless a patient is motivated to seek treatment for symptoms – or an accompanying family member requests an evaluation on his or her behalf – the eating disorder may not be easily apparent. For these reasons, primary care clinicians and mental health generalists should maintain vigilance for undisclosed eating disorder symptoms and understand that patients are frequently reluctant to seek care for this problem. There is typically a several-year delay between onset and treatment seeking; postponement of treatment allows the physiologic and psychological impacts of the disorder to accrue. Although patients may be reluctant to volunteer information about these symptoms, evidence also supports that doctors frequently fail to inquire. When doctors do ask, patients often will admit to symptoms and a more thorough and prompt evaluation can ensue. It is understandable that busy primary care clinicians must be selective in screening questions they pose to their patients and therefore may choose to focus screening for eating pathology in the demographic that they believe will yield the greatest number of cases. Whereas it is undeniably important to exercise watchfulness for eating pathology in the evaluation of adolescent and young adult women, the demographic in which eating disorders are most likely to onset, clinicians should recognize that eating disorders can present in childhood through late adulthood, in both men and women and across all major US ethnic groups.
The eating disorder screening questionnaire, the SCOFF, has clinical utility as a self-report first stage screener (Hill et al., 2009), though importantly it may not identify feeding disorders such as ARFID, pica, or rumination disorder. Alternatively, posing a general screening question, such as “Do you have any concerns (or has anyone else expressed concerns) that you might have an eating disorder or problems with your eating?,” may be useful but has uncertain yield. In the absence of a patient report, clinicians may look for clinical signs heralding an incipient or occult eating disorder, such as low weight, poor weight gain, weight losses and gains, signs of poor nutrition or evidence of potassium wasting (possibly indicative of chronic vomiting, laxative, or diuretic use), or physical signs consistent with chronic vomiting, as noted previously.
Evaluating a Suspected Eating Disorder
If a new-onset eating disorder is suspected, a directed history should ascertain the timeline and severity of weight loss, weight gain, or weight cycling. Clinicians should take special note of rapid and substantial weight loss while ascertaining any of its precipitants. A previous history of the weight nadir at the patient’s present height is an informative benchmark that will guide urgency of intervention, as will history of weight loss or decline in BMI centile with respect to the patient’s growth trajectory. A targeted review of systems should identify signs and sequelae of poor intake, undernutrition, and/or purging and other compensatory behaviors, such as lightheadedness, palpitations, syncopal episodes, seizure, and skipped or irregular menstrual periods. Patients often feel reluctant or ashamed to describe purging symptoms, so it is advisable for clinicians to inventory the type, duration, and frequency of inappropriate compensatory behaviors in an empathic, non-judgmental way that optimally elicits pertinent history. In some cases, patients may not recognize the extreme medical risk posed by purging behavior (e.g., inducing vomiting through syrup of ipecac use or by underdosing or withholding insulin) and prompt education about serious health impact and appropriate intervention is imperative. On the other hand, clinicians should be mindful that their history taking could potentially suggest modes of purging to some patients (e.g., withholding of insulin dose by a patient with insulin-dependent diabetes) and should exercise appropriate caution in evaluating these possibilities.
Medical history and physical and laboratory examination should consider and exclude medical conditions that might have precipitated or exacerbated weight loss. These include gastrointestinal, infectious, rheumatologic, and endocrine disorders. Physical and laboratory exam should also evaluate the potential manifestations and medical consequences of low weight and chronic undernutrition, as well as chronic purging and other compensatory behaviors, including any history of ipecac use. Physical findings rarely confirm a diagnosis of an eating disorder, but they can augment the clinical data informing the differential diagnosis of weight loss. They are also essential in determining the urgency of and optimal setting for therapeutic intervention. Anorexia nervosa and bulimia nervosa are associated with potentially serious medical comorbidity as well as elevated mortality, as noted previously. Because some patients first present with severe illness, the initial evaluation should ascertain whether the patient requires immediate inpatient medical or psychiatric admission for safety and optimal management. For example, rapid and substantial weight loss or a body weight below 75 percent expected for height and age; marked hypotension, bradycardia, or hypothermia; cardiac dysrhythmia or markedly abnormal serious electrolyte derangement such as marked hypokalemia or hypomagnesemia; or other serious complications such as a seizure or syncope are each an indication for inpatient medical care to stabilize and manage medical complications that are either manifest, or that could imminently emerge. Clinicians should be aware that nutritional rehabilitation can precipitate the refeeding syndrome. The refeeding syndrome is a potentially life-threatening condition involving cardiac, respiratory, hematologic, and neuromuscular complications arising during the course of nutritional rehabilitation in severely emaciated individuals. The syndrome is commonly associated with hypophosphatemia, hypomagnesemia, and hypokalemia, and at-risk patients may therefore require intensive monitoring with clinical evaluation, serial laboratory examination, and telemetry for their medical safety.
Because eating disorders commonly have physiological impacts, a comprehensive clinical evaluation of an eating disorder comprises psychological, nutritional, and medical dimensions. Nutritional assessment and management are beyond the scope of this text, but all health professionals treating patients with an eating disorder should be familiar with weight parameters that signal medical risk.
Patients with AN require laboratory examination to evaluate additional medical complications. A comprehensive examination includes screening for (1) electrolyte abnormalities, hypoglycemia, elevated transaminases, elevated amylase and lipase, anemia, leukopenia, thrombocytopenia; (2) electrocardiographic abnormalities, including low voltage, bradycardia, QT interval prolongation, and irregular rhythm; and (3) osteopenia with bone densitometry. Targeted evaluation of endocrine function to evaluate weight changes and menstrual abnormalities may be indicated as well. Patients with BN as well as patients with OSFED/UFED who purge by induced vomiting, laxative, and/or diuretic misuse require evaluation of electrolytes, including potassium and magnesium.
Patients with BED can present with or develop overweight and obesity, which are associated with substantial medical comorbidity and require medical evaluation and management. In addition, because BED can develop subsequent to overweight or weight gain, clinicians should consider medical causes of increased appetite, weight gain, and obesity in their differential diagnosis of patients presenting with BED when clinically appropriate.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
