Guideline Statements and Implementation

Implementation

Estimates of the prevalence and disease burden associated with eating disorders vary by country and depend on the methodology of the epidemiological study (Galmiche et al. 2019; Santomauro et al. 2021; Wu et al. 2020). Among individuals in the United States assessed in the 2012–2013 National Epidemiologic Survey Alcohol and Related Conditions-III (NESARC-III), the 12-month prevalence estimates for AN, BN, and BED were 0.05% (standard error of the mean [SEM] 0.02%), 0.14% (SEM 0.02%), and 0.44% (SEM 0.04%), whereas lifetime estimates were 0.80% (SEM 0.07%), 0.28% (SEM 0.03%), and 0.85% (SEM 0.05%), respectively (Udo and Grilo 2018). Somewhat different estimates were found in prior smaller studies such as the National Comorbidity Replication Survey (Hudson et al. 2007) and pooled data from the National Institute of Mental Health (NIMH) Collaborative Psychiatric Epidemiological Studies (Marques et al. 2011). Other studies suggest that the prevalence of eating disorders may be increasing (Favaro et al. 2009; Galmiche et al. 2019). Furthermore, many of these studies did not fully assess for unspecified or other specified eating disorders. As a result, the actual burden of eating disorders is likely to be underestimated (Feltner et al. 2021; Harrop et al. 2021; U.S. Preventive Services Task Force 2022; Ward et al. 2019).

In the NESARC-III findings, women were more likely to have a 12-month diagnosis or a lifetime diagnosis as compared to men (adjusted odds ratio [OR] for 12-month diagnosis 6.48 for AN, 5.16 for BN, and 2.37 for BED, and for lifetime diagnosis 12.00 for AN, 5.80 for BN, and 3.01 for BED; Udo and Grilo 2018). A lifetime diagnosis of BN was just as likely in Hispanic white and non-Hispanic Black individuals as in non-Hispanic white individuals; however, a lifetime diagnosis of AN was more likely in non-Hispanic white individuals than in Hispanic and non-Hispanic Black individuals, whereas BED was more common in non-Hispanic white than non-Hispanic Black individuals (Udo and Grilo 2018). LGBTQ+ individuals were also more likely to have a lifetime eating disorder diagnosis than cisgender heterosexual individuals, with adjusted ORs of 1.93 for AN, 3.69 for BN, 2.32 for BED, and 1.96 for any eating disorder (Kamody et al. 2020). In addition, more recent data suggest an increasing incidence of eating disorders and inpatient care for eating disorders, particularly AN, during the COVID-19 pandemic, and these increases appear to be unrelated to prior COVID-19 infection (Agostino et al. 2021; Asch et al. 2021; Lin et al. 2021; Otto et al. 2021; Taquet et al. 2021; Toulany et al. 2022).

The U.S. Preventive Services Task Force (2022) notes that there is insufficient evidence for routine screening for eating disorders in adolescents and adults (age 10 years or older) who have no signs or symptoms of an eating disorder. On the other hand, it can be challenging to identify eating disorder signs, symptoms, or risk factors without specific attention to these elements during the evaluation. In addition, it is important to note that the presence of an eating disorder diagnosis cannot be predicted simply by assessing weight or BMI. Data from the Collaborative Psychiatric Epidemiology Surveys of 2001–2003 showed an increase in the adjusted OR for any 12-month or lifetime eating disorder among overweight and obese men and women relative to normal-weight individuals, with the greatest increase among those with Class III obesity (Duncan et al. 2017). Women with a low BMI also had an increased adjusted OR of any 12-month or lifetime eating disorder, but most underweight adults did not meet criteria for an eating disorder (Duncan et al. 2017). Furthermore, many individuals with an eating disorder do not receive help, even when this is broadly defined to include use of self-help or support groups. In the NESARC-III study, the prevalence of seeking any help was 34.5% for AN, 34.5% for BN, and 62.6% for BED, but there was substantial variability based on sex, race, and ethnicity (Coffino et al. 2019). In AN, the likelihood of seeking help was less in Hispanic as compared to non-Hispanic white individuals (adjusted OR 0.30), whereas in BED the likelihood of seeking help was less in men than in women (adjusted OR 0.29) and in non-Hispanic Black individuals (adjusted OR 0.25) and Hispanic individuals (adjusted OR 0.46) as compared to non-Hispanic white individuals (Coffino et al. 2019). Consequently, screening for eating disorder symptoms will be important in order to identify eating disorders and to reduce disparities in receipt of treatment (Marques et al. 2011). Systematically collected prevalence data are less available in gender-diverse individuals, but there appear to be higher rates of eating disorder diagnoses as well as weight and shape concerns among transgender and gender non-binary youth as compared to cisgender youth (Coelho et al. 2019; Grammer et al. 2021). Early recognition of an eating disorder is also essential because of the relatively young age of onset for eating disorders in many individuals. More specifically, the median age of onset in the NESARC-III study was 17.4 years in AN, 16.0 years in BN, and 21.1 years in BED (Udo and Grilo 2018), although some studies suggest that the median age of onset has been decreasing in recent years (Favaro et al. 2009; Galmiche et al. 2019). Mean ages of onset were slightly higher (Udo and Grilo 2018). A long duration of illness was common in the NESARC-III study, with a median duration of the episode of illness of 4.9 years in AN, 8.0 years in BN, and 10.6 years in BED; mean episode durations were 11.4 years (SEM 0.4), 12.2 years (SEM 0.67), and 15.9 years (SEM 0.36), respectively (Udo and Grilo 2018). Psychosocial impairment was also common in individuals with an eating disorder, highlighting the importance of early identification and intervention (Udo and Grilo 2018). For example, in individuals with AN, onset before 15 years of age was associated with greater illness severity, higher rates of lifetime psychiatric comorbidity, and more psychosocial difficulties (Grilo and Udo 2021).

Given the prevalence and typical age of onset of eating disorders in adolescence or young adulthood, the American Academy of Pediatrics recommends that pediatricians ask all preteens and adolescents about eating patterns and body image as well as screen for eating disorders and be alert to potential signs and symptoms of disordered eating (Hornberger et al. 2021).

Prevalence rates of eating disorders among patients receiving psychiatric treatment are likely to be considerably higher than in the general population, given the significant co-occurrence of eating disorders with other psychiatric disorders (see Statement 4). For example, one study of 260 individuals referred to a community-based mental health service for treatment of anxiety or depression noted ratings of eating problems (as measured by a score above 1 on the SCOFF) in 18.5% and a DSM-IV (American Psychiatric Association 1994) eating disorder in 7.3% of the total sample (Fursland and Watson 2014).

Other individuals who appear to have an increased likelihood of an eating disorder include individuals who have experienced teasing or bullying (Day et al. 2022; Lie et al. 2019; Solmi et al. 2021) or childhood sexual abuse (Solmi et al. 2021), athletes (Eichstadt et al. 2020; Sundgot-Borgen and Torstveit 2004), and patients with celiac disease (Lebwohl et al. 2021; Mårild et al. 2017) or type 1 diabetes mellitus (Hall et al. 2021; Toni et al. 2017). Despite their prevalence and importance, eating disorders may remain undetected unless systematic screening occurs. Individuals with an eating disorder may not have insight into the presence or severity of eating disorder signs and symptoms (Arbel et al. 2014; Gorwood et al. 2019; Konstantakopoulos et al. 2011, 2020). Males and individuals from racial, ethnic, and gender minorities may be less likely to be asked about the presence of eating disorder symptoms due to a perception that eating disorders primarily affect certain demographic groups (e.g., young white females). However, eating disorders occur among all populations, age groups, genders, and cultural groups, although clinical presentations may vary (Alegria et al. 2007; Cachelin and Striegel-Moore 2006; Hudson et al. 2007; Makino et al. 2004; Marques et al. 2011; Ricciardelli et al. 2007; Taylor et al. 2007; Udo and Grilo 2018). In fact, eating disorder diagnoses may be more frequent among transgender and gender non-binary individuals than in those who identify as cisgender (Coelho et al. 2019; Grammer et al. 2021). Clinicians may also erroneously overlook an eating disorder, including atypical AN, in individuals whose BMI is in the normal range or higher. To this end, the clinician should be sure to ask all patients about the presence of eating disorder symptoms as part of their standard psychiatric evaluation. For example, as part of the clinical interview, a patient could be asked “Have you or others worried that your preoccupation with weight, body shape, or food is excessive?” and “Have you felt that your weight or body shape excessively affects how you feel about yourself?” Screening questionnaires can also be used (see Table 2), although questions may need to be adapted based on the patient’s developmental and cognitive level. In terms of structured rating scales, the SCOFF questionnaire is most frequently used for screening purposes (Kutz et al. 2020; Morgan et al. 1999). It is a five-item tool that has been translated into multiple languages (e.g., Garcia et al. 2010; Garcia-Campayo et al. 2005; Richter et al. 2017), has been studied in adolescents as well as adults (Kutz et al. 2020), and can be used as a written self-report tool or with questions asked by the interviewer (Perry et al. 2002). The SCOFF has high sensitivity and specificity (Morgan et al. 1999), particularly for identifying the presence of AN or BN in young women with eating disorder symptoms who have two or more positive responses to the SCOFF questions (Kutz et al. 2020). In more diverse populations, the predictive value of the SCOFF is reduced (Kutz et al. 2020; Solmi et al. 2015). It may also have less ability to detect unspecified or other specified feeding or eating disorders, including atypical AN (Maguen et al. 2018). In addition, it was developed before criteria for BED were established, and it performs less well in detecting the presence of BED (Kutz et al. 2020). For this reason, the SCOFF could be supplemented by adding the initial question from Binge Eating Disorder Screener–7 (Herman et al. 2016): “During the last 3 months, did you have any episodes of excessive overeating (i.e., eating significantly more than what most people would eat in a similar period of time)?” A follow-up question could ask whether any such episodes were associated with a loss of control or inability to stop eating.

Other questionnaires, including the Screen for Disordered Eating (Maguen et al. 2018) and the Eating Disorder Screen for Primary Care (Cotton et al. 2003), have also been proposed as screening tools. Both have a greater sensitivity than the SCOFF (Cotton et al. 2003; Maguen et al. 2018), but these screening tools have not been well studied among representative patient populations.

Implementation

A careful assessment of the patient’s history, symptoms, behaviors, and mental status is the first step in making a diagnosis of an eating disorder. This assessment can take several visits to complete and should also address the recommendations of the American Psychiatric Association Practice Guidelines for the Psychiatric Evaluation of Adults (American Psychiatric Association 2016). Information can be obtained through face-to-face interviews, standardized assessment tools, physical examination, laboratory testing, and input from collateral sources, such as family members, other health professionals, and medical records. For a variety of reasons (e.g., ambivalence about changing behavior, stigma, impaired insight), individuals with eating disorders may underreport symptoms (e.g., amount of food consumed, time spent exercising, episodes of binge eating and/or purging). Consequently, family members, partners, or others may observe changes in eating or other behaviors that the patient does not report. In the assessment of children and adolescents, it is essential to involve parents or guardians and, whenever appropriate, school personnel and health professionals who routinely work with the patient. Because many symptoms of eating disorders are cognitive in nature (e.g., fear of weight gain, overvaluation of weight and body shape) and depend on abstract reasoning ability, children may not have the capacity to articulate or endorse such symptoms, and greater reliance on behavioral indicators may be necessary (Lock et al. 2015a). In addition, children and adolescents may exhibit different psychosocial changes than adults as their eating-related symptoms evolve (Hornberger et al. 2021; Lock et al. 2015a). Thus, as a complement to the psychosocial assessment conducted as part of any initial evaluation (American Psychiatric Association 2016), it can be helpful to identify changes in school performance, athletic pursuits, or other differences in psychosocial functioning when assessing children or adolescents with an eating disorder. Clinicians should also keep in mind that the clinical presentation of an eating disorder may be influenced by cultural considerations, although individuals of all age groups, genders, and cultural groups may develop an eating disorder (George and Franko 2010; Hudson et al. 2007; Makino et al. 2004; Perez et al. 2016; Udo and Grilo 2018).

The initial assessment of a patient with a possible eating disorder should include a thorough history of the patient’s height and weight, including lifetime maximum and minimum weights. Clinicians should assess the degree of recent weight loss because medical complications (e.g., refeeding syndrome) are predicted by both the rapidity and total magnitude of weight loss (Garber et al. 2019; Whitelaw et al. 2018). Asking the patient about the weight that would be most comfortable for them can help align treatment-planning goals and provide additional useful information about the patient’s degree of insight (e.g., if they select a desired weight that is significantly below the normal range). The clinician should also document any changes in adult height because this may reflect bone loss resulting from chronic nutritional deficiency (Misra et al. 2016).

For children and adolescents, obtaining historical height and weight percentiles through growth curves or charts (either documented directly or obtained from the patient’s pediatrician or family physician) may help to identify growth retardation associated with AN and is extremely important for characterizing changes in the patient’s weight and height trajectory (Marion et al. 2020; Modan-Moses et al. 2003, 2021; Swenne and Thurfjell 2003). It can also be helpful to take a developmental history of feeding and eating, as described below for ARFID.

A thorough assessment of food and eating patterns, and any changes to patterns of eating behavior, is critical for evaluation of a possible eating disorder. Food repertoire changes (as reflected by typical daily diet habits) may include choosing foods with different (often lower) caloric density, changing to a specific type of diet (e.g., vegan/vegetarian, high protein/paleo, gluten free), reducing the variety of foods eaten, avoiding entire food groups (e.g., dairy) or macronutrients (e.g., fat, carbohydrates), or developing food intolerances, phobias, or aversions. Consequently, it is important to determine whether food ingestion is sufficient to sustain healthy growth and development, whether patterns of food restriction or avoidance may contribute to possible nutritional deficiencies, and whether significant changes in food preferences represent onset of health conditions other than eating disorders. It can also be helpful to ask about ways in which patients find eating behaviors to be helpful to them, including their motivations for food restriction or avoidance. Such inquiries can help patients feel understood and can identify gaps in coping strategies that may warrant attention in treatment.

Individuals with eating disorders, particularly individuals with AN (Gianini et al. 2015), may exhibit abnormal mealtime behaviors, such as excessively tearing or cutting up foods, chewing each bite a certain number of times, delaying onset of eating, or avoiding particular combinations of foods. Other eating disorder-related behaviors may include chewing and spitting, rumination, or concern about gastrointestinal (GI) effects of eating (e.g., fullness, bloating, abdominal pain). Individuals with an eating disorder may also avoid social situations because of heightened sensory experiences of others’ food, feelings of disgust at watching others eat, or self-consciousness about others commenting on their food choices.

Clinicians should inquire about the presence of binge eating, during which the individual experiences a sense of loss of control over eating and consumes an amount of food that is definitely larger than most people would eat under similar circumstances (American Psychiatric Association 2013). It is helpful to learn what occurs during a patient’s typical binge, because their subjective description may provide additional insights into their eating behaviors. Patients should also be asked about purging behavior, which may involve self-induced vomiting or laxative or diuretic use, as well as the frequency of purging behavior, including details on the type and quantity of laxatives or diuretics used and whether there have been recent changes in the pattern and frequency of purging behavior. In addition, clinicians should inquire about compensatory behaviors, such as the use of medication to manipulate weight (e.g., diuretics, caffeine, stimulants, diet pills, nutritional or herbal supplements, muscle building supplements, insulin omission or dose manipulation, thyroid hormones) or excessive exercise. Indicators of compensatory or excessive exercise may include an unwillingness or inability to adapt one’s exercise regimen when injured and rigidity and/or preoccupation with one’s exercise routine to the extent that it contributes to social avoidance (e.g., being so concerned with waking up to exercise that one does not go out with friends in the evening).

Individuals with a possible eating disorder should be asked about the amount of time they spend preoccupied with thoughts about eating, weight, or body shape. Although not all individuals with an eating disorder will report these features, disturbance in the experience of the body and/or overvaluation of weight and body shape may be reflected in negative subjective evaluations of one’s appearance (e.g., feelings of self-disgust toward one’s body), body dissatisfaction (including concern with muscle definition or specific body areas), difficulties trusting perceptions of interoceptive and proprioceptive sensations (e.g., not trusting experiences of hunger as real, concerns about GI symptoms), behavioral rituals (e.g., frequent weighing, checking size of body areas), and cognitive preoccupations (e.g., fear of gaining weight, anxiety about eating, disgust with food, concerns about eating in social situations). In men, there can be a greater focus on muscularity rather than weight per se (Lavender et al. 2017). Bullying and cyberbullying are common and may contribute to body dissatisfaction, among other psychological effects (Day et al. 2022; Lie et al. 2019, 2021a, 2021b; Solmi et al. 2021). In addition, it can often be helpful to ask about social media interactions or peer groups influences that affect patients’ views of eating, weight, or body shape (Padín et al. 2021; Scott et al. 2019).

In transgender and gender non-binary youth, gender-affirming motivations can lead to dietary restriction or other compensatory behaviors to prevent puberty onset or progression (Avila et al. 2019; Coelho et al. 2019). It is also important to learn whether the patient has had gender-affirming medical interventions, including hormone therapy or surgical interventions, and their association with changes in eating disorder symptoms (Jones et al. 2018; Nowaskie et al. 2021; Uniacke et al. 2021).

Gathering information on prior treatment and treatment response can be helpful in conceptualizing the severity and course of illness and in formulating the initial treatment plan. In addition to treatment settings and levels of care, clinicians should inquire about prior experience with, and response to, both psychotherapeutic and pharmacological interventions. Premorbid personality traits (e.g., perfectionism, conscientiousness, and obsessionality in AN; impulsivity with binge eating or purging behavior) may influence symptom severity, treatment planning, and outcomes for individuals with an eating disorder (Dahlenburg et al. 2019; Dufresne et al. 2020; Hower et al. 2021; Legg and Turner 2021; Lilenfeld et al. 2006; Waxman 2009).

The initial assessment should also include a thorough family history. Patients should be asked about a family history of eating disorders, binge eating, dieting or restrictive eating, obesity, or other weight-related issues as well as about family and cultural attitudes toward eating, exercise, and appearance. Clinicians should also inquire about family history of conditions that may be common in individuals with eating disorders, such as diabetes mellitus, inflammatory bowel disease, and other psychiatric disorders, including depression, anxiety, OCD, and substance use disorders (Hudson et al. 2007). Although asking about a family history of suicide is important in every psychiatric evaluation, it is particularly relevant when evaluating a patient with a possible eating disorder given the elevated rates of suicide in this population (Arcelus et al. 2011). When assessing adolescents, clinicians should also consider the role of family interactions and attitudes (Blissett and Haycraft 2011; Lydecker and Grilo 2016) that may require attention as part of the treatment plan.

A patient’s degree of insight and capacity to make a reasoned choice about the need for treatment should be assessed because insight and judgment may be impaired by a constellation of factors, including the effects of restrictive eating on cognition. Eating disorders, especially AN, are also characterized by ambivalence toward treatment because interventions that target disordered eating and weight control behaviors may induce anxiety. Patients who have experienced frequent relapses or an extended history of unsuccessful treatment may feel hopeless about the prospects of improvement. For adolescents, the ability to assess future risk depends on the patient’s level of cognitive development and can complicate assessments of insight and capacity.

Implementation

In the assessment of a patient with a possible eating disorder, obtaining the patient’s weight and quantifying recent or current eating and weight control behaviors can help detect and determine the severity of eating disorder behaviors and associated symptoms. Height should also be obtained, as described in Statement 6.

APA’s Practice Guidelines for the Psychiatric Evaluation of Adults, 3rd Edition (American Psychiatric Association 2016) provides a general description of the use of quantitative measures, which can include use of formal rating scales or quantifying the number or characteristics of relevant behaviors. The intent of using a quantitative measure is not to establish a diagnosis but rather to complement other aspects of the screening and assessment process. Depending on the measure, it can aid in treatment planning by providing a structured, replicable way to document the patient’s baseline symptoms. It can also help to determine which symptoms should be the target of intervention based on factors such as frequency of occurrence, magnitude, associated distress to the patient, and potential for associated harm to the patient or others. On the other hand, it is important to be mindful of the fact that some individuals will underreport symptoms, particularly if they do not feel comfortable with the therapeutic relationship, are not motivated for treatment, lack awareness of having a disorder, wish to avoid disappointing the clinician, feel shame about their behavior, or have experienced prior bullying or criticism for their behavior, weight, or appearance.

As treatment proceeds, use of quantitative measures will often allow more precise tracking of whether pharmacological, psychotherapeutic, and other nonpharmacological treatments are having their intended effects or whether a shift in the treatment plan is needed. Standardized assessments can be useful for demonstrating improvement to patients who may feel unmotivated or disappointed with their response to treatment. They can also provide helpful information about the actual effects of prior treatments. Again, however, underreporting may occur, or rates of symptom reporting may increase as motivation to change or insight improve with treatment. In addition, patients’ ratings can be compared with family members’ impressions of treatment effects to clarify the longitudinal course of the patient’s illness.

One approach to quantitative assessment is to focus on measures of eating disorder severity as described in DSM-5 criteria (i.e., weight loss for AN, episodes of inappropriate compensatory behaviors per week for BN, binge-eating episodes per week for BED). In addition to behavior frequency, measures of behavioral intensity or time spent on dietary restriction, binge eating, purging, exercise, and other compensatory behaviors can also be obtained. Weight is a key measure in individuals with an eating disorder and should be obtained with patients dressed in light clothing and with shoes removed. Whether the measured weight should be shared with the patient or not is unclear and may depend on the treatment approach being used as well as patient-specific considerations.

Although use of an eating disorder rating scale is not necessary to quantify eating and weight control behaviors, a number of patient- and clinician-rated scales and screening tools for eating disorders have been developed and validated (Schaefer et al. 2021). If a scale is used, the choice of a scale should consider the age of the patient, clinical setting, time available for administration, and therapeutic objective (i.e., screening vs. diagnosis vs. ongoing monitoring). As discussed in Statement 1, the SCOFF questionnaire is a five-item tool for eating disorders that has high sensitivity and specificity for identifying AN and BN when screening for an eating disorder (Kutz et al. 2020; Morgan et al. 1999). The Eating Disorder Assessment for DSM-5 (EDA-5) is a freely available clinician-rated, semistructured interview that shows good reliability for diagnosis of AN, BN, and BED in adolescents and adults (www.eda5.org; Sysko et al. 2015). It also has a youth version available for ages 8–14 as well as versions in Spanish and other languages. The Eating Disorders Examination Questionnaire (EDE-Q) is a relatively brief, freely available, and well-validated self-report measure that is adapted from the semistructured Eating Disorders Examination (EDE) interview (Fairburn 2008). It has been validated in Spanish (Grilo et al. 2012a; Peláez-Fernández et al. 2012) and multiple other languages (Lichtenstein et al. 2021). It may have lower validity in men (Smith et al. 2017), but it appears to be a useful self-report measure in transgender and gender-diverse individuals, although further validation is warranted (Avila et al. 2019; Duffy et al. 2021; Nagata et al. 2020a, 2020c; Nowaskie et al. 2021). The EDE-Q is reliable in adults as well as in adolescents (Mond et al. 2004; Schaefer et al. 2018), and a children’s version of the EDE-Q has been validated for use in 7- to 18-year-olds (Kliem et al. 2017). Another scale developed for children and adolescents ages 8–14 years is the Kids’ Eating Disorders Survey (Brewerton 2001; Childress et al. 1993). For assessing longitudinal changes in eating-related cognitions and behaviors, the Eating Disorders 15 (ED-15) has been developed (Tatham et al. 2015) and has a corresponding version for youth (Accurso and Waller 2021a) and for reporting by parents or caregivers (Accurso and Waller 2021b). The Clinical Impairment Assessment (CIA; www.psytoolkit.org/survey-library/eating-cia.html) is a self-report measure available in English and Spanish to measure psychosocial impairment associated with an eating disorder (Bohn et al. 2008; Jenkins 2013; Maraldo et al. 2021; Martín et al. 2015; Raykos et al. 2019), although different thresholds for total scores may be needed in men and women (Richson et al. 2021).

For assessment of other disorders related to eating, the Eating Disorders in Youth Questionnaire has been validated for use in 8- to 13-year-olds (Goldberg et al. 2020) and the Nine Item Avoidant/Restrictive Food Intake Disorder Screen (Zickgraf and Ellis 2018) has been validated for use in adults. The Pica, ARFID, and Rumination Disorder Interview (PARDI; Bryant-Waugh et al. 2019) is a multi-informant, semistructured instrument that is used in research; however, clinicians may benefit from reviewing the items on the PARDI to learn about clinical features of these disorders.

Implementation

Co-occurring health conditions are important to identify as part of the evaluation of a patient with a possible eating disorder. Some co-occurring health conditions may be a sequela of an eating disorder (e.g., gastroesophageal reflux disease, irritable bowel syndrome, gastroparesis, other GI motility disorders), whereas others (e.g., diabetes mellitus, celiac disease, inflammatory bowel disease) can place restrictions on eating behaviors and dietary variety and can exacerbate or increase the likelihood of developing an eating disorder (Nikniaz et al. 2021; J.E. Peters et al. 2022; Young et al. 2013). Even when a health condition is independent of an eating disorder, it can influence the choice of treatment or a need for medical stabilization.

Identification of preexisting or co-occurring psychiatric conditions and obtaining information on their onset and course is also important for treatment planning. When another psychiatric condition is present, outcomes are worse (Franko et al. 2018; Keshishian et al. 2019; Lydecker and Grilo 2021; Riquin et al. 2021) and mortality is greater (Himmerich et al. 2019a, 2019b; Kask et al. 2016, 2017). Thus, it is essential to provide care for both the eating disorder and other psychiatric conditions. The relationship between eating disorders and psychiatric symptoms is complex; careful clinical assessment is needed to discern whether symptoms of depression, anxiety, or obsessionality reflect an independent co-occurring disorder or have developed as a result of the eating disorder. For example, starvation has been shown to lead to depressive symptoms, including low mood, impaired concentration, low energy, and sleep disturbance, as well as increased anxiety and obsessionality (Keys et al. 1950). On the other hand, eating disorders frequently co-occur with other psychiatric disorders, particularly depression, anxiety, OCD, PTSD, autism spectrum disorder, substance use disorders, and personality disorders (Hudson et al. 2007; Steinhausen et al. 2021). Individuals with an eating disorder also have a greater likelihood of ADHD than individuals without an eating disorder (Brewerton and Duncan 2016; Nazar et al. 2016). Physical restlessness is commonly observed in low-weight patients with AN and can manifest as persistent fidgeting or refusal to sit for extended periods of time, independent of the presence of ADHD.

A history of trauma may also increase risk for development of disordered eating behaviors (Emery et al. 2021; Russon et al. 2019) or an eating disorder. Reports of prior sexual trauma are elevated in patients with eating disorders (Lie et al. 2021a; Madowitz et al. 2015; Solmi et al. 2021), but rates of physical or emotional abuse or neglect are also increased (Afifi et al. 2017; Coffino et al. 2020; Hazzard et al. 2019; Kimber et al. 2017; Lie et al. 2021a; Molendijk et al. 2017; Pignatelli et al. 2017). In addition, many individuals who have an eating disorder have experienced bullying or criticism of their weight or appearance (Day et al. 2022; Lie et al. 2019, 2021b; Solmi et al. 2021). Consequently, all patients with a possible eating disorder should be asked about a history of trauma; physical, emotional, or sexual abuse; bullying (including cyberbullying); or neglect (including food insecurity) and assessed for symptoms related to PTSD (Ferrell et al. 2020).

Suicide is the second leading cause of death among individuals with AN, and rates of suicidal behavior are elevated in individuals with BN and BED (Smith et al. 2018). The initial examination should include a thorough assessment of suicide risk, including current suicidal ideas, plans, or intentions, prior suicidal plans or attempts, and the presence of non-suicidal self-injury (see Guideline III, “Assessment of Suicide Risk,” in the APA’s Practice Guidelines for the Psychiatric Evaluation of Adults; American Psychiatric Association 2016). Such assessments can be conducted through clinical interview, mental status examination, or use of quantitative measures.

For any patient who is undergoing an initial psychiatric evaluation, it is important to assess the patient’s use of caffeine, tobacco, alcohol, cannabinoids, and other substances, as well as any misuse of prescribed or over-the-counter (OTC) medications or supplements (see Guideline II, “Substance Use Assessment,” in the Practice Guidelines for the Psychiatric Evaluation of Adults; American Psychiatric Association 2016). Substance use disorders are frequently comorbid with eating disorders (Bahji et al. 2019; Harrop and Marlatt 2010; Javaras et al. 2008; Krug et al. 2008); thus, a comprehensive substance use history is essential in a patient with a potential eating disorder. Cigarette smoking (including electronic cigarettes or vaping) can be used to suppress appetite (Mason et al. 2021; Naveed et al. 2021), and smoking can affect the rate of weight restoration during treatment (Van Wymelbeke et al. 2004). A specific inquiry should also be made about the use or misuse of prescribed or non-prescribed medications that suppress appetite (e.g., OTC weight loss products, stimulants) or enhance muscularity (e.g., supplements, androgens).

Among individuals with ARFID, comorbidity with GI disease (e.g., achalasia, eosinophilic esophagitis, celiac disease, inflammatory bowel disease) is common (Day et al. 2022; Fink et al. 2022; Gibson et al. 2021; Murray et al. 2021; J.E. Peters et al. 2022; Robson et al. 2019; Yelencich et al. 2022) and evaluation for GI abnormalities may be warranted, particularly in individuals whose symptoms are not lifelong. Autism spectrum disorder and ADHD also appear to be more frequent in individuals with ARFID (Farag et al. 2022; Keery et al. 2019; Yule et al. 2021), emphasizing the importance of taking a history for developmental, learning, and sensory issues. Anxiety symptoms and diagnoses (Fisher et al. 2014; Kambanis et al. 2020; Katzman et al. 2021; Keery et al. 2019; Norris et al. 2014, 2021) and depressive symptoms (Katzman et al. 2021) are also reported frequently by individuals with ARFID. More detailed recommendations about screening for co-occurring conditions can be found in the Practice Guidelines for the Psychiatric Evaluation of Adults (American Psychiatric Association 2016).

Implementation

The effects of malnutrition, binge eating, and purging can affect every organ system in the body (Academy for Eating Disorders Medical Care Standards Committee 2021; Cass et al. 2020; Sachs and Mehler 2016). In addition to the recommendations for a review of systems as found in the Practice Guidelines for the Psychiatric Evaluation of Adults (American Psychiatric Association 2016), a focus on issues that are common in patients with eating disorders can help to identify preexisting or co-occurring conditions (as discussed in Statement 4) as well as elicit symptoms of concern to the patient (see Table 3). Although some of these symptoms or conditions may improve or resolve with treatment of the eating disorder, others will require additional evaluation and treatment in addition to treatment of the eating disorder.

Patients with eating disorders commonly report symptoms such as abdominal discomfort or pain with eating, constipation, early satiety or fullness, bloating, nausea, and gastroesophageal reflux (Riedlinger et al. 2020). These symptoms do not necessarily reflect evidence of a structural GI disorder but may be a consequence of starvation and disordered eating patterns that result in functional GI disorders and problems with GI motility (e.g., delayed gastric emptying [Hetterich et al. 2019]). Patients who induce vomiting should be asked about hematemesis.

Cardiovascular issues are also common as described further in Statement 8. Changes in cardiac rhythm include bradycardia, whereas other arrhythmias can present with palpitations (Giovinazzo et al. 2019; Sachs et al. 2016). Low blood pressure, often in association with orthostatic hypotension, can result in dizziness on standing or syncope. Rates of mitral valve prolapse, pericardial effusion, and myocardial atrophy also appear to be increased in individuals with AN (Giovinazzo et al. 2019; Olivares et al. 2005; Sachs et al. 2016; Smythe et al. 2021).

It is similarly important to inquire about past or current neurological signs or symptoms, such as seizures and headache (including migraine headaches). Osteoporosis and fractures (including stress fractures) occur at an increased frequency in individuals with an eating disorder (Frølich et al. 2020; Robinson et al. 2016, 2019; Solmi et al. 2016) and should also be identified as part of the review of systems.

Assessment should include taking a menstrual history, when relevant, including age of menarche and date of last menstrual period. It is also important to ask about use of oral contraceptives or other hormonal therapies that may affect menses. Menstrual cycle abnormalities, including irregular periods and amenorrhea, occur in AN (Misra and Klibanski 2014), atypical AN (Garber et al. 2019; Lebow et al. 2015; Rastogi et al. 2020), BN (Gendall et al. 2000), and BED (Olguin et al. 2017). In addition, polycystic ovary syndrome also appears to be associated with an increased likelihood of having disordered eating (Pirotta et al. 2019) or an eating disorder, particularly BN or BED (Thannickal et al. 2020). Dietary restriction with significant weight reduction or a low BMI can also be associated with increased rates of pregnancy complications and neonatal difficulties. Patients should be asked about sexual (e.g., decrease in libido, erectile dysfunction) and reproductive (e.g., infertility, obstetrical complications) issues that may arise in the setting of altered hypothalamic-pituitary-gonadal axis functioning.

Implementation

A complete physical examination is strongly recommended in addition to assessment of vital signs and physical appearance. As described in the Practice Guidelines for the Psychiatric Evaluation of Adults (American Psychiatric Association 2016), it may be performed by a psychiatrist, another physician, or a medically trained clinician but is best performed by a clinician familiar with common findings in patients with eating disorders. Diagnosis should rely on a comprehensive assessment of psychiatric and medical status and history because a physical examination with normal results may not exclude an eating disorder. Furthermore, physical findings such as low blood pressure or low resting heart rate, which can be seen in healthy individuals, may not be a marker of health in an individual with other evidence of an eating disorder. Following the initial evaluation, the frequency of physical assessment will depend on the individual’s clinical status.

Table 2 describes physical signs that may occur in individuals with an eating disorder. The physical examination should give particular attention to vital signs because abnormalities may indicate medical instability, which would warrant a higher level of care (see Statement 9). Abnormalities of potential concern include a heart rate < 50 beats per minute (bpm), a systolic blood pressure < 90 mmHg, or a temperature < 36°C (96.8°F). A sustained decrease of systolic blood pressure of at least 20 mmHg or pulse increases of > 30 bpm in adults or > 40 bpm in adolescents ages 12–19 years within 3 minutes from lying to standing may also indicate medical instability (Freeman et al. 2011; Raj et al. 2020; Singer et al. 2012). All patients should be evaluated for evidence of self-injurious behaviors because individuals with eating disorders experience elevated rates of self-injury compared to the general population (Cucchi et al. 2016; Forrest et al. 2021; Kostro et al. 2014). Physical examination of children and adolescents with a possible eating disorder should also include assessment of growth and pubertal development (e.g., as indicated by the Tanner stage of sexual maturity).

Patient height, weight, and BMI should be evaluated initially, with weight obtained, ideally, at all visits (see www.cdc.gov/healthyweight/assessing/bmi/adult_bmi/english_bmi_calculator/bmi_calculator.html). The frequency at which the patient’s height needs to be measured will vary, with adults requiring height determinations less frequently than adolescents. Some patients may prefer to be weighed in a blinded fashion (i.e., turn around to remain unaware of their weight; Froreich et al. 2020; Wagner et al. 2013). The decision to weigh patients in this manner as compared to an open fashion is controversial and often depends on the philosophy of the eating disorders treatment program. Other factors that may influence the choice of blinded as compared to open weighing can include the type of setting (inpatient vs. outpatient), the type of psychotherapy (e.g., family-based therapy [FBT], cognitive-behavioral therapy [CBT]), the frequency of weight checks, and patient characteristics or preferences (Forbush et al. 2015). Even when weighing does occur in a blinded fashion, patients may become aware of their weight through electronic health record notes.

When interpreting a patient’s BMI and related measures, it is important to be aware of the limitations of this parameter. In particular, it does not distinguish between fat and muscle mass, reflect differences in fat distribution, or incorporate variations in BMI due to age, sex, race, or ethnicity (Kesztyüs et al. 2021; Lee et al. 2017; Liu et al. 2021; Tinsley et al. 2020). Nevertheless, it is readily measurable and in frequent clinical and research use. In children and adolescents, percent median BMI (current BMI/50th percentile BMI for age and sex × 100; www.cdc.gov/healthyweight/bmi/calculator.html), BMI percentile, or BMI Z-score should be determined (Golden et al. 2015a). Longitudinal growth charts should be documented directly or obtained from the patient’s pediatrician or family physician to assess for deviations from individual growth trajectories and to guide determination of a target weight. Such an approach can also help to identify growth retardation associated with AN (Marion et al. 2020; Modan-Moses et al. 2003, 2021; Swenne and Thurfjell 2003), weight suppression in association with atypical AN, or consistently low height and weight percentiles as can be seen in some youth with ARFID (Yule et al. 2021).

Patients should be evaluated for physical manifestations of malnutrition, which may include proximal muscle and temporal wasting, ankle and pedal edema, and dermatological changes such as lanugo (fine downy hair), hair loss, and dry skin. In addition, some individuals with severe dietary restriction may become immunocompromised (Brown et al. 2008). Signs of malnutrition improve with normalization of eating behavior and weight; targeted treatment (e.g., use of diuretics for malnutrition-related peripheral edema) is rarely needed for these signs.

Vitamin deficiencies (e.g., vitamin A, thiamine, vitamin B12, vitamin C, vitamin D, zinc) can also develop due to dietary restrictions in individuals with AN, atypical AN, or ARFID (Achamrah et al. 2017; Hanachi et al. 2019; Yule et al. 2021). Risk of vitamin deficiencies can be compounded by co-occurring conditions (e.g., thiamine deficiency with co-occurring alcohol use disorder). Physical findings may include angular stomatitis, glossitis, bleeding gums, and dermatological, ocular, or neurological findings (Suter and Russell 2018).

Clinicians should also assess for any signs of purging, such as parotid gland enlargement, dental enamel erosion, and calluses on the knuckles or dorsum of the hand (Russell’s sign) from scraping against the teeth during attempts to induce vomiting. If purging behavior is present, referral for a dental evaluation is indicated. Although seemingly paradoxical, patients should be instructed not to brush teeth after vomiting (Meurman and ten Cate 1996; Otsu et al. 2014). Oral rinsing with water after vomiting and avoiding ingestion of carbonated beverages or citrus fruits may also help to reduce effects on dentition (Otsu et al. 2014).

Implementation

Laboratory assessments can be helpful in the initial assessment of a patient with a possible eating disorder in detecting abnormalities that may require intervention (see Table 4), including a higher level of care. Abnormalities are more frequent in individuals with severe or chronic illness, frequent purging behaviors, or rapid recent weight loss, independent of the individual’s current weight. On the other hand, abnormal laboratory values do not occur in all individuals with an eating disorder, and normal laboratory values do not rule out a potential eating disorder.

Patients with eating disorders, particularly those who are at low weight, may present with anemia, leukopenia, and/or thrombocytopenia (Cleary et al. 2010; Hütter et al. 2009; Peebles and Sieke 2019). In individuals who purge or restrict fluids, hemoconcentration resulting from dehydration may initially mask anemia. These hematological abnormalities are typically reversible with restoration to a normal weight. Individuals with AN may show evidence of hepatic dysfunction (Rosen et al. 2016), reflected by elevations in liver enzymes (alanine aminotransferase [ALT] and aspartate aminotransferase [AST]). A rise in aminotransferases may also occur in conjunction with renourishment due to hepatic steatosis (Rosen et al. 2016). Patients with AN may also develop hypoglycemia in the setting of reduced glycogen stores and impaired gluconeogenesis (Gaudiani et al. 2012). In addition, postprandial hypoglycemia can occur in individuals with a low BMI (Hart et al. 2011; Kinzig et al. 2007). Electrolyte disturbances are common and can result from restrictive eating, purging, or laxative or diuretic use. Individuals who vomit regularly can develop hypokalemia and hypochloremic metabolic alkalosis, whereas patients who misuse laxatives may develop a hyperchloremic metabolic acidosis (Peebles and Sieke 2019). Although less common, patients with eating disorders who drink excessive amounts of water may present with hyponatremia, which poses a risk for seizures (Miller et al. 2005). The risk of hyponatremia may also be increased by concurrent use of medications that can cause it (e.g., SSRIs). Measurement of urinary specific gravity can help to identify individuals who are consuming excess water or, conversely, are at risk of dehydration. When volume depletion is severe, such as in individuals with AN who also purge, increases in blood urea nitrogen (BUN) and creatinine can be seen and, rarely, renal failure may occur.

The need for additional laboratory analyses should be determined on an individual basis depending on the patient’s condition or the laboratory tests’ relevance to making treatment decisions. Serum magnesium and phosphorus levels are commonly measured and may need to be ordered separately from a comprehensive metabolic panel. They should be considered depending on the patient’s clinical picture (e.g., risk of refeeding complications, low BMI, rapid recent weight loss, significant medical comorbidities, severe malnourishment). The risk of abnormal magnesium and phosphorus levels is not limited to low-weight individuals with eating disorders; hypomagnesemia can develop in patients with purging behavior (Raj et al. 2012), whereas hypophosphatemia can emerge in individuals with erratic eating patterns (e.g., periods of severe restriction or fasting behavior). Serum amylase levels, specifically levels of salivary amylase, may be elevated in patients who self-induce vomiting. With starvation and with renourishment, elevations in serum lipase can be seen but generally do not require intervention.

Individuals with AN have reduced bone mineral density (BMD) and increased fracture risk (Faje et al. 2014; Lucas et al. 1999; Misra et al. 2008a, 2008b; Nagata et al. 2017; Vestergaard et al. 2003). Prolonged amenorrhea is associated with reduced BMD. Thus, in patients with menstrual irregularities and primary or secondary amenorrhea, gonadotropin (e.g., follicle stimulating hormone, luteinizing hormone), estradiol, and prolactin levels can be measured, as well as a urine pregnancy test (American College of Obstetricians and Gynecologists 2018). Testosterone levels also appear to influence bone density in individuals of all genders (Khosla and Monroe 2018). Some groups have recommended based on consensus that, after 6 months of amenorrhea, bone densitometry (dual-energy X-ray absorptiometry [DXA] scan) may be warranted (Golden et al. 2014, 2015b; Gordon et al. 2017). Other clinicians obtain a DXA scan as a baseline measure even in patients with regular menses. In non-menstruating individuals with AN, there are no data to inform decisions about when to order a DXA scan. In all patients, evidence of low BMD can be useful in providing education about the health impact of AN and motivating them to gain weight.

Measurement of thyroid stimulating hormone can serve as a screening test for possible misuse of thyroid hormone (e.g., OTC thyroid supplements or levothyroxine) and can help rule out other medical conditions such as hyperthyroidism, which can lead to weight loss. On the other hand, individuals who have had significant weight loss and malnutrition can exhibit a nonthyroidal illness syndrome in which levels of total T3 are low and levels of thyroid stimulating hormone may be normal or reduced (Schorr and Miller 2017).

Other potentially useful laboratory assessments include a urine toxicology screen to assist in identifying stimulant misuse, measurement of the erythrocyte sedimentation rate to help distinguish an eating disorder from other inflammatory conditions in patients who report abdominal discomfort after eating, and serum tests to assess for nutritional status or vitamin deficiency (e.g., vitamin D, calcium, iron, thiamine).

Implementation

The appropriateness of an electrocardiogram (ECG) depends on the diagnosis, illness severity, and vital signs and need not be obtained in every patient or in those with mild symptoms. However, an ECG should be obtained in certain circumstances, including for individuals taking medications known to prolong QTc intervals and for those with a restrictive eating disorder, including AN, atypical AN, and ARFID. In addition, patients often underreport purging severity, and obtaining an ECG can identify concerning cardiac changes that may point to an underlying eating disorder.

Individuals with AN are at elevated risk for a number of structural and functional cardiac abnormalities, including bradycardia, myocardial atrophy, pericardial and valvular pathology, conduction abnormalities, and sudden cardiac death (Olivares et al. 2005; Sachs et al. 2016). Less is known about cardiac effects in individuals with atypical AN and ARFID, but bradycardia does occur (Sawyer et al. 2016; Strandjord et al. 2015; Whitelaw et al. 2014, 2018) and may be associated with greater amounts of recent and total weight loss (Whitelaw et al. 2018). In individuals with BN or the binge/purge subtype of AN, self-induced vomiting and/or laxative abuse contribute to an increased risk for prolonged QTc intervals and cardiac arrhythmias due to electrolyte abnormalities (e.g., hypokalemia, metabolic alkalosis; Gibson et al. 2019; Peebles et al. 2010). The risk of QTc prolongation can be increased by concurrent use of medications known to prolong the QT interval. Drug-drug interactions that increase serum levels of these medications can further increase risk. Such medications include, but are not limited to, antidepressants, antipsychotics, antiarrhythmics, some classes of antibiotics (e.g., macrolides, fluoroquinolones), antiviral medications (e.g., for HIV), antiemetics (e.g., ondansetron), antihistamines (e.g., hydroxyzine, diphenhydramine), and some cancer therapeutic agents (Funk et al. 2018; Woosley et al. 2022).

Implementation

In treating individuals with an eating disorder, a person-centered treatment plan should be developed, documented in the medical record (e.g., as part of a progress note), and updated at appropriate intervals. The aim of person-centered care, which is sometimes referred to as patient-centered care, is to provide care that is respectful of and responsive to individual preferences, needs, and values and ensures that an individual’s values guide clinical decisions (Institute of Medicine Committee on Quality of Health Care in America 2001). Person-centered care is achieved through a dynamic and collaborative relationship among individuals, families, other persons of support, and treating clinicians that helps achieve the individual’s realistic health and life goals and informs decision-making to the extent that the individual desires (American Geriatrics Society Expert Panel on Person-Centered Care 2016). With person-centered care, patients, families, and other persons of support are provided with information that allows them to make informed decisions (Institute of Medicine 2006). Evidence-based interventions should be adapted to meet individual needs and preferences when possible (van Dulmen et al. 2015). Self-management approaches and shared decision-making are encouraged (Institute of Medicine 2006), with the recognition that shared decision-making may not be possible if an individual lacks awareness of their illness or the need for treatment.

A person-centered treatment plan can be recorded as part of an evaluation note or progress note and does not need to adhere to a defined development process (e.g., face-to-face multidisciplinary team meeting) or format (e.g., time-specified goals and objectives). However, it should give an overview of the identified clinical and psychosocial issues, along with a specific plan for addressing factors such as food avoidance, restrictive eating, binge eating, purging, or other compensatory behaviors (if present) and related social avoidance and/or isolation. The plan should also discuss whether there is a need for further history and mental status examination; physical examination (by either the evaluating clinician or another health professional); laboratory testing; ongoing monitoring; and pharmacological, psychotherapeutic, and other nonpharmacological interventions, as indicated. In addition, the clinical evaluation should include discussion of the patient’s gender and their individual strengths, vulnerabilities, personality traits, developmental stage, and motivation for treatment, each of which can inform treatment planning and help anticipate possible issues that may arise during treatment. Collateral informants such as family members, friends, or other treating health professionals may express specific concerns about the individual’s eating disorder symptoms. If present, such concerns should be documented and addressed as part of the treatment plan. Treatment plans can also include elements such as collaborating with other treating clinicians, providing integrated care, educating patients about treatment options, discussing the potential impact of social media use on symptoms and eating-related behaviors, engaging family members, exploring family attitudes toward eating, and addressing these attitudes, if indicated.

An understanding of the individual’s cultural identity is essential to appreciating the ways in which the patient defines key concerns and values, interacts with family members, receives support from their social network, copes with stressors, and engages in help-seeking behaviors. Cultural and religious beliefs can also be relevant to the patient’s dietary choices. The DSM-5-TR Cultural Formulation Interview (American Psychiatric Association 2022) provides a framework for eliciting such information.

Depending on the urgency of the initial clinical presentation, the availability of laboratory results, or receipt of history from collateral informants, the initial treatment plan may need to be augmented over several visits and as more details of the history and treatment response are obtained. The patient’s goals and readiness to change eating patterns and behaviors will likely evolve over time. Changes to the treatment plan will also be needed if a patient has not tolerated or responded to a specific treatment or chooses to switch treatment approaches. Symptoms of the eating disorder or of co-occurring conditions may also shift with time and can require a reassessment of the diagnosis or treatment plan.

In determining a patient’s initial level of care or whether a change to a different level of care is appropriate, it is important to consider a constellation of factors, including the patient’s overall physical condition, behaviors, affective state, cognitions, and social circumstances (see Table 5 and Table 6). Services for the treatment of eating disorders can range from intensive inpatient programs (in which general medical care is readily available), to residential and partial hospitalization programs, to varying levels of outpatient care (in which the patient receives general medical treatment, nutritional counseling, and/or individual, group, and family psychotherapy). Characteristics of such services are described in Table 7.

As an initial treatment setting, outpatient care will be appropriate for the majority of patients, although some individuals will need initial treatment in a higher level of care. Evidence-based outpatient treatment is effective and will commonly be provided by a coordinated multidisciplinary team (Golden et al. 2015b). The processes by which care coordination occur will differ with the setting and with the expertise and responsibilities of multidisciplinary team members but should be discussed and agreed upon in advance to ensure optimal care.

Outpatient treatment has the advantage of allowing patients to remain with their families and continue to attend school or work; however, patients and their families should understand that a higher level of care may be necessary if weight control behaviors or eating disorder symptoms are worsening or if progress is not seen over 6 weeks (e.g., as evidenced by an average weight gain of 0.5–1 lb/week in individuals with AN, 50% decrease in purging behaviors for individuals with AN or BN). Thus, for individuals treated in an outpatient setting, careful monitoring is essential and includes at least weekly weight determinations done directly after the patient voids, with shoes and outerwear removed. Depending on the patient’s clinical presentation and symptoms, orthostatic pulse and orthostatic blood pressure may also need to be measured on a regular basis. Additionally, in patients who purge, it is important to monitor serum electrolytes with a monitoring frequency determined by prior electrolyte values, purging frequency, and other aspects of the patient’s clinical condition.

A number of factors can suggest that a higher level of care is needed, either initially or following a trial of outpatient treatment. These include low weight in relation to estimated individually determined target weight, rate of recent weight loss, medical complications of purging, evidence of medical instability (e.g., vital sign abnormalities, severe electrolyte disturbances), approaching a weight at which instability previously occurred in the patient, additional stressors that affect the patient’s eating disorder behaviors, the degree of the patient’s difficulties in collaborating in their care, and co-occurring psychiatric symptoms or diagnoses that suggest a need for a higher level of care or that merit inpatient admission in their own right. Insufficient weight gain or worsening eating disorder symptoms despite treatment can also suggest a need for a higher level of care. Each patient will differ in the degree to which these factors may influence decisions about the most appropriate care setting, and individuals will not necessarily move in a sequential fashion from one level on the care continuum to the next. Furthermore, there is no algorithmic approach that can determine the optimal care setting with certainty. For example, individuals may have had prior experiences in a particular treatment program or level of care that will influence current decision-making. In addition, consequences of nutritional restriction in children and adolescents can contribute to negative effects on neuronal development, impairments in concentration leading to poorer educational outcomes, loss of BMD, and arrested physical growth and development (Bang et al. 2021; Hemmingsen et al. 2021; Modan-Moses et al. 2021; Workman et al. 2020). A child or adolescent with a rapid loss of weight may also become medically compromised more rapidly than an adult with a similar amount or rate of weight loss. As compared to adults, resting vital signs also differ in children, who will typically have a higher resting heart rate and lower systolic blood pressure. Children and adolescents may also be more affected by stressful family dynamics or school-related stressors, including bullying or cyberbullying. On the other hand, older patients or those with chronically low weight may be at heightened risk as compared to younger individuals with shorter illness durations because of long-term malnutrition or multiple medical comorbidities. Older patients may also have additional stressors (e.g., financial, occupational, social) or fewer psychosocial supports as compared to children and adolescents who are residing with family.

For individuals who require inpatient care, the choice of a specific program should be made based on the patient’s psychiatric and medical status, the skills and abilities of local psychiatric and medical staff, and the availability of suitable programs to care for the patient’s psychiatric and other medical problems. In general, however, outcomes are likely to be better when patients are treated on inpatient units staffed with experts in treating eating disorders than when treated in general medical or psychiatric inpatient settings where staff lack expertise and experience with eating disorders. Furthermore, expert eating disorders behavioral specialty programs can improve eating disorder symptomatology and restore weight in the majority of underweight patients using a multidisciplinary approach that includes supervised meals, behavioral contingency management strategies, and individual, group, and family-based treatments (Attia and Walsh 2009). Patient and program characteristics may also play a role in determining the pluses and minuses of program choice. For example, many programs will only accept voluntary patients, which can influence treatment site access and selection. Insurance-related considerations are another common source of difficulty in accessing an appropriate level of care (Guarda et al. 2018; Walker et al. 2020).

In patients of all ages, legal interventions, including involuntary hospitalization and legal guardianship, may be necessary to address the safety of treatment-reluctant patients whose medical conditions are life-threatening. In such circumstances, involuntary treatment is ethically and clinically justified only when a patient’s decision-making capacity regarding appropriate treatment for their eating disorder is impaired, the risk of death or serious morbidity is high, and the likelihood of benefit from involuntary treatment outweighs the risk of harm. There is a limited amount of data on treatment outcomes with involuntary treatment of AN; however, in general, rates of mortality with long-term follow-up do not appear to differ for those who have received involuntary treatment as compared to voluntary treatment (Atti et al. 2021; Ward et al. 2015). Such findings are encouraging because involuntarily treated patients with AN have a higher severity of overall symptoms, a larger number of prior hospitalizations, a lower BMI at the time of admission, and a higher likelihood of having comorbid conditions than patients who are admitted voluntarily (Atti et al. 2021).

When shifts are made in the setting or location of care, continuity of care is essential. Transition planning requires that the care team in the new setting or locale be identified and that specific patient appointments be made. It is preferable that a specific clinician on the team be designated as the primary coordinator of care to ensure continuity and attention to important aspects of treatment.

Weight status, per se, or other physiological or behavioral markers should never be used as a sole criterion for transition to a less intensive setting. For example, patients who are physiologically stabilized on acute medical units will still require specific inpatient treatment for eating disorders if they do not meet medical, psychiatric, and behavioral criteria for less intensive levels of care and/or if no suitable, less intensive levels of care are accessible because of geographical or other reasons. Individuals with atypical AN may still be malnourished and be at risk for complications regardless of their current weight or BMI (Whitelaw et al. 2018). Assisting patients in determining and practicing appropriate food intake at a healthy weight is likely to decrease the chances of their relapsing after transitioning to a less intensive setting. If transitions between settings of care occur too frequently or after only brief periods of time, it can disrupt therapeutic relationships, sabotage patient progress, and lead to poorer outcomes.

Peer-support programs can supplement but should not replace professionally provided treatment for an eating disorder. Programs that provide peer support can reduce feelings of shame or stigma, decrease social isolation, assist with sustaining of recovery, and help families understand eating disorders. Such programs differ in their emphasis but focus on self-acceptance, improved body image, increased physical movement, and better nutrition and health. On the other hand, programs that focus on abstaining from specific food groups are nutritionally problematic and can interfere with recovery. Many patients and families are also accessing helpful information through websites, newsgroups, chat rooms, and social media. In some instances, however, the lack of professional supervision within these resources may result in unhealthy dynamics among users or in perpetration of misinformation. Thus, it is recommended that clinicians inquire about a patient’s or family’s use of peer- or internet-based support and openly discuss the information, ideas, and approaches to eating that have been gathered from these sources.

Implementation

Medical stabilization, nutritional rehabilitation, weight restoration, and maintenance of weight gain are critical components of treatment for AN that focus on helping the patient achieve and maintain a healthy and medically appropriate weight for their age and height. The same principles will apply to individuals with other restrictive eating disorders, including ARFID and atypical AN, and may also apply to individuals with other eating disorder diagnoses who require weight restoration. The goals of nutritional rehabilitation for seriously underweight, malnourished, or medically unstable patients are to restore medical stability (e.g., normalize vital signs, electrolytes, and fluid balance), restore weight, correct biological and psychological sequelae of malnutrition, normalize eating patterns, and achieve normal perceptions of hunger and satiety.

Nutritional rehabilitation may be achieved in a variety of settings (e.g., outpatient, day treatment program, residential, hospital) and will depend on the patient’s medical and psychiatric stability. The setting where a patient is engaged in weight restoration will vary according to a number of factors, including age, severity of illness, available psychosocial support networks, and available treatment options (see Table 7). For individuals who are markedly underweight, hospital-based programs for nutritional rehabilitation should be considered. Weight restoration for adolescents and emerging adults with AN will often occur as an outpatient under the supervision of their parents/caregivers (e.g., with FBT). Outpatient weight restoration may also be appropriate for some adult patients, provided they are able to demonstrate consistent increases in weight; however, a higher level of care may be needed if weight control behaviors or eating disorder symptoms are worsening or if progress is not seen over 6 weeks (e.g., as evidenced by an average weight gain of 0.5–1 lb/week). For those in inpatient or residential settings, the weight at which it is appropriate to discharge a patient may vary in relation to the patient’s individually determined target weight and will depend on the patient’s ability to feed themself, the patient’s motivation and ability to participate in treatment, and the availability and adequacy of programs at a lower level of care. In general, the closer a patient is to their individually determined target weight before discharge, the less risk they will have of relapsing and being readmitted. Having patients maintain their weight for a period of time before they are discharged from inpatient or residential treatment likely decreases the risk of relapse as well.

To help patients normalize eating and weight control behaviors, most specialty inpatient and residential programs employ supervised meals and group therapies as well as some level of behavioral contingency management as part of a structured behavioral treatment protocol. With such an approach, positive reinforcements (e.g., privileges) and negative consequences (e.g., required bed rest, exercise restrictions, restrictions of off-unit privileges) are built into the program; negative consequences can then be reduced or terminated and positive reinforcements accelerated as target weights and other goals are achieved.

Renourishment should be implemented in nurturing emotional contexts. Staff should convey to patients their intention to take care of them and not let them die even when the illness prevents the patients from taking care of themselves. If the patient experiences an element of a structured treatment program as aversive, the staff should clearly communicate the rationale for programmatic protocols—that the aim is to help shape and reinforce behaviors and choices aligned with health. Ongoing staff training and peer support models may be useful to support staff members in providing empathic care. As discussed in Statement 9, compulsory treatment is ethically and clinically justified only when a patient’s decision-making capacity regarding appropriate treatment for their eating disorder is impaired, the risk of death or serious morbidity is high, and the likelihood of benefit from involuntary treatment outweighs the risk of harm.

Implementation

Psychotherapy is appropriate as an initial intervention in all age groups. It is also appropriate in the weight restoration and relapse prevention stages of treatment. In addition to a focus on normalizing eating and weight control behaviors and restoring weight, psychotherapy should include consideration of other factors, such as body image normalization and eating-related cognitions.

During acute renourishment, it is beneficial to provide patients with AN with individual psychotherapeutic management that provides empathic understanding, explanations, praise for positive efforts, coaching, support, encouragement, and other positive behavioral reinforcement. Initiation of an eating disorder–focused psychotherapy is also integral to the treatment of AN, although the timing should be individualized based on the patient’s medical stability and readiness to engage in psychotherapy. For example, with severely malnourished patients, attempts to conduct formal psychotherapy may be ineffective. The goals of psychotherapeutic interventions include helping patients with AN 1) discuss their experience of their illness; 2) cooperate with their nutritional and physical rehabilitation; 3) change the behaviors and dysfunctional attitudes (e.g., cognitive distortions) related to their eating disorder; 4) identify developmental, familial, and cultural antecedents of their illness; 5) address comorbid psychopathology, psychological conflicts, adaptive benefits of symptoms, and family or cultural factors that reinforce or maintain eating disorder behaviors; 6) improve their coping skills and their interpersonal and social functioning; 7) resume age-appropriate life roles (e.g., school, work, relationships); and 8) address other quality of life concerns. During weight maintenance phases of treatment, psychotherapy can also help patients address residual concerns relating to body image, body shape, and weight acceptance. During this phase, psychotherapy can help patients identify areas for continued progress (e.g., further normalization of eating and exercise behavior, resumption of functional life roles) and learn how to avoid or minimize the risk of relapse, including specific concerns about possible abuse, neglect, or developmental traumas and approaches to better cope with salient developmental and other important life issues or stressors in the future. Patients’ responses to the therapist, including transference, can be influenced by the characteristics of the therapist (e.g., age, gender, ethnicity, body size), and it is important to attend to these issues, if present. In addition, clinicians need to attend to their countertransference reactions to patients with a chronic eating disorder, which often include beleaguerment, demoralization, and excessive need to change the patient. Consultation with other clinicians can be helpful in managing these responses.

In terms of specific psychotherapies, research studies often describe these interventions as distinct, but features of psychotherapies are often shared (see Table 8), and there is frequent overlap of the psychotherapeutic interventions used in clinical practice.

Some approaches, such as FBT, emphasize a time-limited model of treatment; however, for individuals with AN of long duration, psychotherapeutic treatment is frequently required for at least 1 year and may take many years. For patients whose illness is resistant to treatment, more extensive psychotherapeutic measures may be undertaken to engage and help motivate patients, because patients can experience substantial remission even after many years of illness (Dobrescu et al. 2020; Eddy et al. 2017; Eielsen et al. 2021).

Among the psychotherapies that have been examined, those with modest efficacy in treating AN in adults include CBT (eating focused and broadly focused), focal psychodynamic psychotherapy (FPT), interpersonal therapy (IPT), Maudsley Model of Anorexia Nervosa Treatment for Adults (MANTRA), and Specialist Supportive Clinical Management (SSCM). Experienced Carers Helping Others (ECHO) is aimed at supporting carers of patients with AN but can also contribute to improved outcomes. Psychotherapies that appear to be effective in treating adolescents and emerging adults are discussed in Statement 12.

Implementation

FBTs that include caregiver education are effective as a treatment for AN in adolescents. They are less well studied but likely to also be helpful in emerging adults ages 18–26 years of age. Family-based interventions are not limited to family members, per se, but could involve other non-family caregivers with whom the patient resides.

FBT for AN is a manual-based approach that focuses on the effects of severe weight loss as being central to the core psychology of AN (Le Grange 1999; Lock and Le Grange 2013). The focus of FBT is to enlist parents as experts on parenting their child and have them oversee and take responsibility for nourishing the child or adolescent back to an optimal weight range. One of the central tenets of FBT is to take a non-ideological stance to disorder onset. One purpose of this emphasis is to alleviate the significant blame that parents with a child with AN have historically experienced. The therapist acts as knowledge expert and facilitator; however, it is the parents who come up with solutions on how best to nourish their child. This process is facilitated by an in-session family meal to which parents bring a meal that they would feed their child. This in-session meal is a crucial part of therapy and allows the therapist to observe the approaches the parents use. For example, if parents’ own eating behaviors are abnormal or restrictive, this can influence their approach to engaging their child in eating. Through gentle questioning, the therapist helps the parents arrive at strategies that would be most optimal for the child within their family context. For example, cultural beliefs and practices related to eating may be explored and used in developing the parents’ approach to the child’s eating.

At the beginning of each FBT session, the therapist weighs the patient and discusses the patient’s current concerns, emotions, and thoughts. These data are presented to the family for use in discussion about strategies at home that are working or need to be enhanced. Other elements of the treatment can include providing information about nutrition and addressing eating-related cognitions and body image normalization. As treatment progresses to Phase 2, more responsibility for independent eating is given to the adolescent and, in Phase 3, the transition to typical adolescent development is discussed. When the patient has not gained sufficient weight after 1 month of treatment, a variation of FBT has been tested in which parents engage in more in-session family meals with more intense parental coaching (Lock et al. 2015b). These sessions are intended to help the parents engage in problem-solving to address barriers to implementation and reinvigorate the family around the severity of illness and the need for intense support and supervision. Treatment may also need to be adapted for higher levels of care (Freizinger et al. 2021; Halvorsen et al. 2018; Huryk et al. 2021; Spettigue et al. 2019).

The treatment may need to be adapted depending on the developmental needs of the patient. For example, with emerging adults as compared to young teens, it can be helpful to have the young adult give input on the type of mealtime support they prefer and to specifically consider age-appropriate situations such as attending college, beginning work, or living apart from family (Chen et al. 2016; Dimitropoulos et al. 2018; Gorrell et al. 2019). Adaptations or augmentative approaches to FBT have been developed (Gorrell et al. 2019; Lock et al. 2015b; Richards et al. 2018), including multifamily formats (Baudinet et al. 2021; Eisler et al. 2016) and a parent-focused format in which the therapist meets primarily with the parent and patient monitoring is conducted by another member of the treatment team (Le Grange et al. 2016).

For some adolescents and emerging adults, FBT may not be readily accessible due to geographical or other constraints. Small studies and significant experience during the COVID-19 pandemic suggest that FBT can be delivered using a telehealth platform to make care more readily available (Andersen et al. 2017; Hellner et al. 2021; Matheson et al. 2020). For other patients, FBT may not be feasible due to patient or family preferences or a lack of involved family members or non-family caregivers who are able and willing to engage in treatment. Other individuals may have been treated with FBT without achieving a significant response. In such circumstances, other family or individual approaches to psychotherapy may be helpful.

Adolescent-focused therapy (AFT), originally referred to as ego-oriented individual therapy, is aimed at helping patients identify their emotions and learn to tolerate negative affective states (Fitzpatrick et al. 2010; Lock et al. 2010; Robin et al. 1999). The therapist interprets the patient’s behavior and emotions to help them distinguish emotional states from bodily needs, such as eating. Other themes in treatment include a focus on separation and individuation. Psychoeducation related to nutrition and effects of malnutrition is incorporated into treatment. Weight restoration occurs during the 32–40 sessions of AFT; however, in contrast to FBT, in which the parents or other carers assume responsibility for the patient’s eating, AFT encourages the patient to change their own eating behaviors and gain weight. Meetings with family can occur with AFT but are used to assess parental functioning and provide updates on progress. In studies of this approach, AFT had comparable outcomes to FBT at the end of treatment but was less likely to lead to full remission of AN at follow-up assessments. In terms of other psychotherapies, systemic family therapy has been studied (Agras et al. 2014), as has CBT-E (Le Grange et al. 2020); both appear to have outcomes comparable to those of FBT.

Implementation

The aims of treatment for patients with BN are to 1) reduce and, where possible, eliminate binge eating and purging; 2) treat physical complications of BN; 3) enhance patients’ motivation to cooperate in the restoration of healthy eating patterns and participate in treatment; 4) provide education regarding healthy nutrition and eating patterns; 5) encourage increased food variety and minimization of food restriction; 6) encourage healthy but not compulsive exercise patterns; 7) help patients reassess and change core dysfunctional thoughts, attitudes, motives, conflicts, and feelings related to the eating disorder; 8) address other themes that may underlie eating disorder behaviors (e.g., developmental issues, identity formation, body image concerns, self-esteem, sexual and aggressive difficulties, affect regulation, sex role expectations, family dysfunction, coping styles, problem solving); 9) treat associated psychiatric conditions, including deficits in mood and impulse regulation, self-esteem, and behavior; 10) enlist family support and provide family counseling and therapy when appropriate; and 11) prevent relapse.

Nutritional intake is important to assess for all patients with BN regardless of their body weight or BMI because normal weight does not imply appropriate nutritional intake. Adequate nutritional intake and structured eating can prevent cravings and promote satiety. In addition, helping the patient develop a structured meal plan can aid in reducing episodes of dietary restriction and urges to binge and purge. Nutritional interventions also are valuable in increasing the variety of foods eaten. Nutrition counseling with a registered dietitian nutritionist is often needed to implement the eating plan and to make adjustments to address challenges that arise (e.g., co-occurring physical health conditions, frequent work-related schedule changes or travel) (Academy for Eating Disorders Nutrition Working Group 2021; Hackert et al. 2020).

Several approaches are possible for the initial treatment of an adult with BN. Many patients show a reduction in binge eating and purging behaviors with CBT alone; however, the combination of CBT plus high-dose fluoxetine (60 mg/day) is associated with somewhat better responses than fluoxetine alone (Goldbloom et al. 1997). For this reason, initial treatment could also include a combination of CBT and an SSRI (e.g., high-dose fluoxetine). Decisions about initiating treatment with psychotherapy alone as compared to combination treatment will depend on factors such as symptom severity, co-occurring disorders, and patient preferences. Nevertheless, the best long-term outcomes occur when the initial treatment response is relatively rapid (Mitchell et al. 1993). Thus, if there is minimal or no response to psychotherapy alone by 6 weeks of treatment, addition of an SSRI will typically be warranted. Antidepressants or other psychopharmacological agents may also be used to treat specific co-occurring disorders, such as depressive, anxiety, obsessive-compulsive, or posttraumatic stress disorders.

Implementation

As in the treatment of AN, FBT has evidence of benefits in the treatment of BN for adolescents or emerging adults who reside with family or other carers who are able to participate in treatment (Le Grange et al. 2007, 2015; Schmidt et al. 2007).

FBT for BN is similar to the manual-based approach for AN (Le Grange 1999; Lock and Le Grange 2013) as described in Statement 12; however, it focuses on addressing the secrecy, shame, and dysfunctional eating patterns of BN by developing a more collaborative relationship with parents or other carers. As a result, adolescents and emerging adults are assisted in resuming a typical developmental trajectory.

In some instances, FBT may not be feasible due to geographical or access constraints, patient or family preferences, or a lack of involved family members or non-family caregivers who are able and willing to engage in treatment. In other circumstances, a patient may have been treated with FBT without a complete response. For such individuals, CBT either adapted for adolescents or using a therapist-led GSH approach can be considered (Dalle Grave et al. 2021; Le Grange et al. 2015; Schmidt et al. 2007).

Use of fluoxetine or other SSRIs has not been well studied to treat BN in adolescents, although emerging adults have been included in some of the adult research studies. If antidepressant treatment is otherwise indicated for a co-occurring disorder, fluoxetine has been well studied in treatment of depression and anxiety disorders in this age group (Cipriani et al. 2016; Wang et al. 2017) and has the best evidence for efficacy in BN in adults (see Statement 13). If an antidepressant is considered, however, the potential benefits and risks of treatment should be discussed with patients (and parents or guardians as appropriate), and clinicians should attend to the boxed warnings relating to antidepressants in adolescents and young adults (U.S. Food and Drug Administration 2018).

Implementation

Psychotherapy with CBT or IPT shows short- and long-term benefits for BED outcomes (Hilbert et al. 2019). The aims of treatment for patients with BED are to 1) reduce and, where possible, eliminate binge eating; 2) enhance patients’ motivation to participate in treatment and cooperate in the restoration of healthy eating patterns; 3) provide education regarding healthy nutrition and eating patterns; 4) encourage increased food variety and minimization of food restriction, if present; 5) help patients reassess and change core dysfunctional thoughts, attitudes, motives, conflicts, and feelings related to binge eating; 6) address other themes that may underlie eating disorder behaviors (e.g., developmental issues, identity formation, body image concerns, self-esteem, sexual and aggressive difficulties, affect regulation, sex role expectations, family dysfunction, coping styles, problem solving); 7) treat associated psychiatric conditions, including deficits in mood and impulse regulation, self-esteem, and behavior; 8) enlist family support and provide family counseling and therapy when appropriate; and 9) prevent relapse.

Nutritional intake is important to assess for all patients with BED regardless of their body weight or BMI, because normal weight does not imply appropriate nutritional intake. Dietary restriction or weight suppression can be present but may not be identified without a detailed nutritional history. In addition, restrictive interventions aimed at weight loss may fuel binge eating. A history of food insecurity has also been associated with disordered eating, including binge eating (Hazzard et al. 2020). Helping the patient develop a structured meal plan with adequate nutritional intake can reduce dietary restriction, if present, and can also promote satiety, prevent craving, and reduce urges to binge eat. Nutritional interventions also are valuable in increasing the variety of foods eaten. Nutrition counseling with a registered dietitian nutritionist is often needed to implement the eating plan and make adjustments to address challenges that arise (e.g., co-occurring physical health conditions, frequent work-related schedule changes or travel) (Academy for Eating Disorders Nutrition Working Group 2021; Hackert et al. 2020).

Implementation

Psychotherapy, either CBT or IPT, is recommended for the treatment of BED. However, some adults with BED may prefer medication to psychotherapy, whereas other individuals will not respond to psychotherapy alone or will have moderate to severe BED that may benefit from adjunctive pharmacotherapy. In such circumstances, two approaches to medication therapy for BED can be considered: an antidepressant medication or lisdexamfetamine.

Although CBT alone is generally associated with greater effects than antidepressant medications alone in the treatment of BED (Devlin et al. 2005, 2007; Grilo et al. 2005b, 2012b; Ricca et al. 2001), antidepressant therapy has been shown to be beneficial in reducing binge eating, independent of whether a co-occurring depressive or anxiety disorder is present (Guerdjikova et al. 2008; Hudson et al. 1998; Leombruni et al. 2008; Pearlstein et al. 2003). In addition, many individuals with BED will have a co-occurring disorder that would warrant antidepressant treatment (Devlin et al. 2005). Based on the available research evidence, there is insufficient information to recommend one antidepressant or class of antidepressants over another. Consequently, selection of an antidepressant is usually made through shared decision-making based on tolerability, side effect profile, and potential for drug-drug interactions. For this reason, tricyclic antidepressants and monoamine oxidase inhibitors are less likely to be used than other antidepressant medications. In addition, for patients who have purging behaviors or a history of purging behaviors, bupropion is contraindicated given the increased risk of seizures observed in individuals with BN in early clinical trials of high-dose immediate-release bupropion (Horne et al. 1988; Pesola and Avasarala 2002). If patients have concerns about weight gain with antidepressants, studies typically do not show changes of weight in either direction when individuals with BED receive treatment with an antidepressant.

Lisdexamfetamine has also been associated with modest short-term effects in BED (Guerdjikova et al. 2016; McElroy et al. 2015b, 2016a, 2016b, 2017). In addition, continued treatment with lisdexamfetamine was associated with less risk of relapse than when lisdexamfetamine was discontinued (Hudson et al. 2017). However, lisdexamfetamine has primarily been studied in obese patients in primary care settings, and its benefits in other patients with BED are unclear. When used in individuals with BED, the initial dosage is 30 mg once daily in the morning, with increases in dosage of 20 mg/week to a therapeutic dosage of 50–70 mg once daily (Guerdjikova et al. 2016; McElroy et al. 2015b, 2016a, 2016b, 2017). No dosage adjustments appear to be needed for individuals with hepatic dysfunction, although drug-drug interactions can occur with other medications that are metabolized through CYP2D6 hepatic enzymes (Lexicomp 2021; Takeda Pharmaceuticals et al. 2021). For individuals with renal impairment, lower dosages are indicated (i.e., 50 mg/day maximum dosage for glomerular filtration rate [GFR] 15 to < 30 mL/minute/1.73 m²; 30 mg/day maximum dosage for GFR < 15 mL/minute/1.73 m² or in end-stage renal disease requiring hemodialysis) (Lexicomp 2021; Takeda Pharmaceuticals et al. 2021). Common side effects of lisdexamfetamine include insomnia, reduced appetite, upper abdominal pain, and xerostomia; however, these effects are typically well-tolerated with comparable rates of study withdrawal with lisdexamfetamine as with placebo (Guerdjikova et al. 2016; Lexicomp 2021; McElroy et al. 2015b, 2016a; Takeda Pharmaceuticals et al. 2021). Nevertheless, as with other stimulant medication, lisdexamfetamine treatment can be associated with modest decreases in weight and with increases in heart rate, blood pressure, anxiety, or jitteriness. Caution is needed if it is used in individuals with hypertension or cardiac disease, and more frequent monitoring of vital signs may be warranted. Individuals with psychotic symptoms or bipolar disorder or those who have risk factors for these conditions may experience a worsening of symptoms with stimulant treatment. The possibility of stimulant misuse or dependence should also be considered before deciding on treatment with lisdexamfetamine as well as during treatment.

Topiramate, either alone or in combination with CBT, has also been studied in individuals with BED and obesity (Claudino et al. 2007; McElroy et al. 2003, 2007b; Nourredine et al. 2021). Although topiramate treatment was associated with reductions in binge eating, attrition in these randomized trials was high, and adverse effects were more common in participants who received topiramate as compared to placebo. Cognitive dysfunction is commonly reported with topiramate, even at relatively low doses (Lexicomp 2021). Other adverse effects include increased risks of hyperchloremic metabolic acidosis, nephrolithiasis, and ocular problems (Lexicomp 2021). Dosage adjustment may be needed in geriatric patients and in individuals with renal or hepatic impairment. For individuals with childbearing potential, use of effective contraception is important; in utero exposure to topiramate has been associated with an increased risk in offspring of oral clefts or of being small for gestational age (Lexicomp 2021).