A salient feature at the core of anorexia nervosa (AN) is energy intake that is inadequate with respect to caloric requirements, resulting in significantly low body weight. Restrictive eating may be characterized by food rules, some of which are remarkably similar across individuals (e.g., recurrent selection of low-fat foods) and some of which are idiosyncratic (e.g., eating only at 10 minutes after the hour). Although present across eating disorders, restrictive eating poses particular challenges when it leads to undernourishment, nutrient deficiencies, or frank starvation, as in the cases of AN and avoidant/restrictive food intake disorder (ARFID). Restrictive eating behaviors in these disorders warrant particular attention because they contribute to the severity of illness, including medical morbidity, the need for hospitalization or other intensive treatment, and mortality.
Restrictive intake by individuals with AN has been documented in objective studies of eating behavior, showing caloric intake below caloric needs and a significantly reduced intake of fat specifically (Hadigan et al. 2000; Mayer et al. 2012). In a study of hospitalized patients with AN, measurement of eating in a laboratory setting revealed significantly reduced intake compared with that of healthy control subjects at the time of hospital admission. Although intake increased after normalization of weight, it remained significantly reduced compared with that observed in the control subjects. In addition to their restricted energy intake, patients with AN showed a specific reduction in calories from fat. Patterns of restriction have been shown to be related to longer-term outcome (Schebendach et al. 2008); these patterns include rigid rules, repetitive intake of the same few foods with little variety, and intake of low-energy foods and noncaloric beverages. Ecological momentary assessment studies and laboratory meal studies among individuals who restrict food have demonstrated a relationship between affective state and restriction (Lavender et al. 2013; Steinglass et al. 2010).
AN and ARFID differ in the psychological features that motivate food restriction, and fewer studies have been done of eating behavior in ARFID specifically. Nevertheless, treatments for both have similar components because the common goal is normalization of weight and eating behavior. In this chapter, we outline principles for treating feeding and eating disorders that are characterized by low weight and restrictive eating, including AN and ARFID. The principles of treatment in this chapter are derived from the treatment of AN, with applicability to ARFID. We consider features specific to ARFID at the end of the chapter.
Low-weight disorders are challenging to treat for two notable reasons: First, no treatment has emerged as the clear, empirically supported treatment of choice for all patients. Second, for many individuals, there is a reluctance to utilize treatment, including aspects that emphasize improved eating behavior and improved weight status.
AN and ARFID are different illnesses in their initiating and sustaining factors. AN is characterized by inadequate intake relative to requirements associated with fear of fatness or behavioral interference with weight gain, together with body weight or shape concerns. ARFID is an eating or feeding disturbance that is manifested by failure to meet appropriate nutritional and/or energy needs that is not associated with body weight or shape concern but may be associated with anxiety about eating and associated features (e.g., fear of vomiting, fear of choking) or with food avoidance due to the sensory characteristics of food.
Because ARFID is a newly described disorder in DSM-5 (American Psychiatric Association 2013), very little has been written about its treatment. In contrast, much has been described about approaches for AN, but small study sample sizes, high dropout rates, negative findings in randomized clinical trials using particular treatment strategies, and tiny numbers of studies examining more comprehensive, multimodal treatment approaches have limited the evidence base for treatments for AN. Most treatment information regarding AN appears in professional guideline and expert consensus documents, and this information suggests that behavioral management is a core strategy for the achievement of behavioral change in the treatment of AN (American Psychiatric Association 2006; Wilson and Shafran 2005).
Both AN and ARFID are psychiatric conditions with medical as well as psychological features; therefore, treatment needs to include a comprehensive assessment of medical and psychiatric symptoms and a specific assessment of the acute medical and psychiatric risks. This assessment will inform treatment goals and the selection of an appropriate treatment setting. Selection of treatment setting usually includes consideration of the least restrictive setting that is appropriate for the identified goals (Table 13–1). Treatment goals for individuals with restrictive eating and low weight include medical stabilization as needed, nutritional rehabilitation (reversal of nutritional deficiencies and restoration of normal weight), and interruption of eating-disordered behaviors. Strategies for achieving these goals are described in the remainder of this chapter.
Both AN and ARFID are associated with nutritional compromise and physiological changes, many of them severe and some potentially life-threatening. Height and weight assessments are a first step in assessing nutritional status. A clinician’s determination of underweight commonly includes assessment of weight in the context of an individual’s baseline or highest weight, as well as assessment of physiological disturbances that may be associated with weight status (see Chapter 2, “Eating Problems in Adults”). According to the National Heart, Lung, and Blood Institute (2000) and the World Health Organization (1995), the lower limit of a normal body mass index (BMI) is 18.5 kg/m2. Notably, the World Health Organization defines moderate thinness as BMI less than 17.0 kg/m2, severe thinness as BMI less than 16.0 kg/m2, and extreme thinness as less than 15.0 kg/m2.
| Setting | Description | Indications |
|---|---|---|
Outpatient |
Individual or group-based sessions are available, and patients often select or are recommended for several treatment components, often offered by clinicians from different clinical disciplines (e.g., psychology, nutrition, medicine). Outpatient treatment is optimized when providers communicate regularly in order to coordinate care, creating a treatment “team.” |
An outpatient program is the most commonly used setting for eating disorder treatment. Many patients utilize outpatient treatment as they begin engagement in eating disorder treatment or because they are unable to access higher levels of care because of geographic or other resource limitations. Outpatient treatment is most appropriate for individuals who are medically stable and are achieving or maintaining behavioral goals using this level of care. |
Intensive outpatient program (IOP) |
IOP refers to a routine of outpatient sessions in which visits include several hours of treatment per visit offered at a frequency of several visits (e.g., three) each week. Supervised meals are often available as part of an IOP visit. |
An IOP is appropriate for individuals who are medically stable and require small amounts of meal instruction or meal supervision without additional daily programmatic structure. An IOP is often used as a step-down from higher levels of care. |
Partial hospital program (PHP) |
Also known as day treatment programs, PHPs offer more hours of weekly treatment than do IOPs. A PHP may serve as a transition from inpatient to outpatient care or may help some individuals avoid the need for hospitalization. PHPs generally include 4–7 days of treatment with two or three supervised meals each treatment day. |
A PHP is appropriate when meal supervision is needed without requirement for 24-hour supervision. PHP admission generally requires that patients be at or above a minimally acceptable weight (e.g., ≥80% ideal body weight) and maintain other evidence of medical stability. A PHP is often used as a step-down from a higher level of care, or as a step-up from a lower level of care. |
Residential treatment center (RTC) |
RTCs offer specialized treatment delivered in a full-time setting; however, they are less structured than hospital programs. They include less medical monitoring and less staffing at night than do hospital-based programs. |
An RTC is appropriate for patients with low weight (e.g., <85% ideal body weight) and/or evidence of eating disorder behaviors who are in need of close supervision but not in need of daily medical attention. Patients in an RTC must demonstrate motivation needed for voluntary treatment. |
Inpatient program |
Psychiatric hospitalization represents the highest level of care and may be necessary for some individuals with eating disorders, especially those at low weights and those with comorbidities. Specialized inpatient programs include medical personnel, such as psychiatrists and nurses. |
An inpatient program is appropriate for patients with significantly low weight (e.g., <75% ideal body weight) and/or the presence of other signs of medical or psychiatric instability (e.g., vital sign or electrolyte disturbance; comorbidity, including behavioral dyscontrol and/or suicidality). Inpatient treatment is also appropriate for individuals who have failed to respond to treatment at an RTC. |
For children and adolescents, assessments of weight and height require comparison to reference standard data for age and sex. BMI varies greatly in growing children, and BMI-for-age reference standards are important for evaluating healthy and expected growth. BMI assessment for children should be examined in the context of individual growth curves (see Chapter 3, “Eating Problems in Children and Adolescents”). Failure to gain as expected may be as serious an indication of nutritional compromise in a child or adolescent as weight loss is for an adult with a restrictive eating disorder.
Many of the physical consequences of malnourishment that are commonly manifested in AN may be manifested in ARFID as well. Almost every system in the body is affected as part of the physiological responses of the body to being underweight, including cardiac, metabolic, endocrine, skeletal, hematopoietic, gastrointestinal, and dermatological (including skin and hair). Physiological responses to low weight commonly include bradycardia, decreased respiration rates, and low body temperature. Laboratory assessments commonly reflect abnormalities consistent with nutritional deficiencies, dehydration, and purging behaviors (see Chapters 2 and 3 for additional information about medical complications of low weight). Although many of these complications are chronic consequences of starvation and weight loss, others present acute management issues (Trent et al. 2013). As refeeding is initiated, vital signs and laboratory test results should be monitored closely and should improve as energy intake and hydration reach daily requirements.
Successful treatment of restrictive eating associated with AN or ARFID requires nutritional rehabilitation. Resumption of energy intake adequate for gaining weight to and then maintaining weight within a healthy range is essential. Psychological support can help with motivation to eat and making specific behavioral changes, but formal psychotherapy and other psychosocial interventions may be of limited utility in underweight and nutritionally deficient individuals. Patients are encouraged to restore weight fully (e.g., BMI=20-22 kg/m2; weight consistent with pre-illness weight range or growth curve, if patient had healthy baseline; or weight consistent with return of normal menstruation for the amenorrheic patient). Better long-term outcomes have been shown to be associated with full weight restoration (Baran et al. 1995; Kaplan et al. 2009).
Nutritional plans for initial weight gain involve reintroducing foods at modest caloric levels (e.g., 1,500-1,800 kcal/day); providing supervision, psychological support, and psychoeducation (Table 13–2); and medical monitoring. Macronutrient composition is prescribed consistent with the standard daily macronutrient requirements per the Institute of Medicine to ensure adequate dietary fat in particular (Marzola et al. 2013). Liver function should be monitored because abnormalities, including paradoxically elevated cholesterol, are common. In addition to the medical monitoring described above, patients may benefit from the nutritional information that their cholesterol will improve with a normal diet. Low-fat diets are not indicated.
Caloric prescription should increase steadily (e.g., by 400 kcal every 48-72 hours), with ongoing monitoring, until a weight gain rate of 1-2 kg per week is consistently achieved. Weight restoration at this rate commonly requires consumption of 3,500-4,000 kcal/day. In addition to food, meals and snacks, nutritional supplements are often needed during weight gain. Vitamin supplements (e.g., daily multivitamin, thiamine, folate) are commonly prescribed. Supportive acknowledgment of the physical discomfort associated with the gastrointestinal sequelae of starvation, including decreased motility and constipation, which may contribute to early fullness and related discomfort after eating, and the sequelae related to laxative use discontinuation (e.g., edema and constipation) is needed. Repeated reassurance that continued intake will lead to improvement in these physical symptoms is often required. Stool softeners (e.g., docusate sodium) and nonstimulant bulking agents (e.g., polyethylene glycol) may also be considered.
Low-weight patients may benefit from information about the physiological and psychological consequences of low weight and restrictive eating. For some low-weight individuals, it may be useful to emphasize the identified diagnosis (e.g., anorexia nervosa or avoidant/restrictive food intake disorder), whereas for others, especially those who insist that their eating disturbance has atypical features, it is preferable to begin with the identified risk of low weight or restrictive eating and the need for nutritional change, without one particular diagnostic label. Individuals with low weight should be informed that many of their symptoms (e.g., anxiety, depression, preoccupation with food) would be expected to improve with weight gain, despite their often strong beliefs that weight gain or changed eating behaviors would worsen their mood or anxiety symptoms. Obtaining a careful history of cognitive and psychological functioning prior to the onset of restrictive eating may identify baseline strengths that may be used as treatment targets and possible motivators for embarking on weight and eating change. For individuals with avoidant/restrictive food intake disorder, obtaining a careful history from patient and/or family of symptoms that interfere with normal eating and inform specific food choices is essential. Treatment needs to include individualized goals appropriate for specific symptoms. Discomfort with food sensations may require graded exposure to novel foods; restrictive eating due to fear of choking or vomiting may require exercises that target these concerns. |
Nutritional rehabilitation emphasizes normative eating, with structured meals and snacks that include adequate dietary variety and energy density. Feeding via nasogastric tube may be necessary for individuals resistant to eating voluntarily or for those prescribed exceptionally large doses of liquid intake. Individuals with ARFID may need meal plans that target specific nutrient deficiencies that have developed in the context of the eating disturbance.
Therapeutic meal plans should be designed to improve diet variety and increase energy density. For example, they should include items with higher kilocalories per gram, meals that moderate water consumption, and minimal noncaloric foods and beverages. Although meal plans initially may not offer patients much preference or choice, greater autonomy in food selection is given as patients improve in medical status, weight, and eating behavior. Following weight restoration, nutritional plans should adjust to help patients stay within a healthy weight range.
It is important to understand that while nutritional rehabilitation targets weight and physiological change, it also improves psychological symptoms associated with AN and ARFID. Many of the psychological features attributed to these conditions are, in fact, part of the natural sequelae of starvation and being underweight. In their landmark study of semistarvation in previously healthy male subjects who were given restrictive diets, Keys et al. (1950) described depressed mood, restricted affect, heightened anxiety, poor concentration, perfectionism, and obsessionality associated with the underweight and malnourished state. Even in these subjects without an eating disorder, the authors observed increased preoccupation with food, as well as unusual patterns of eating (e.g., eating quickly or dawdling over eating) in the setting of significant weight loss. These historical findings suggest that the psychological symptoms present in low-weight restrictive eaters with AN or ARFID may have developed or intensified as a result of the state of undernutrition.
Other common psychological sequelae of the underweight state in individuals with AN or ARFID include poor sleep, sadness, hopelessness, and anxiety. Anxiety symptoms may include social fears, generalized worry, and physical symptoms of anxiety, as well as eating-related and non-eating-related obsessions and compulsions. Cognitive disturbances, including poor attention, visuospatial deficits, and executive functioning deficits, have been well described in the underweight state among individuals with AN (Steinglass and Glasofer 2011). It is recommended that individuals with AN or ARFID receive psychoeducation about both the physical and psychological consequences of low weight as they are supported through nutritional rehabilitation (see Table 13–2).
In addition to experiencing psychological change secondary to nutritional depletion, patients with AN or ARFID may have co-occurring psychiatric diagnoses, most commonly mood and anxiety disorders, and these may not resolve with refeeding. In a sample of 172 individuals with AN presenting for treatment, 35% met criteria for comorbid mood disorders and 11% for anxiety disorders (Bühren et al. 2014). In their retrospective study of 173 children and adolescents receiving day treatment for a feeding or eating disorder, Nicely et al. (2014) reported that 72% of the 39 individuals with ARFID met criteria for an anxiety disorder, in contrast to the lower rate of 37% of the 93 individuals with AN. Additionally, the investigators found that 13% of those with ARFID and none of those with AN met criteria for autism spectrum disorder. Also, across many studies, suicidal ideation, suicide attempts and self-injury, and rates of completed suicides were consistently reported to be high in samples of patients with AN (Berkman et al. 2007).
Significant improvement in psychological symptoms is seen with nutritional rehabilitation and weight restoration (Attia et al. 1998; Sysko et al. 2005). Psychological improvement may lag behind physiological change, which can be a challenging situation for patients who seek relief of symptoms. Mood and anxiety symptoms, in particular, may continue to be outside the normal range at the time of acute weight normalization. These symptoms show continued improvement with long-term maintenance of healthy eating and healthy weight (Pollice et al. 1997).
Because both AN and ARFID may be associated with reluctance to normalize eating behaviors, behavioral management treatment is commonly employed to reverse or reduce many of the most worrisome features of these eating disorders (Attia and Walsh 2009). Behavioral management programs are those that encourage the achievement of normal weight and eating behavior through the use of reinforcements for healthy behavioral choices. Behavioral management may be delivered as part of inpatient or outpatient treatment; if it is offered as part of outpatient treatment, a frequency of more than once weekly is generally required. Commonly, these treatments include supervised meals and snacks, use multiple treatment modalities, and include clinicians across disciplines. The aim of the meal supervision is to address behaviors at meals that contribute to the perpetuation of eating restriction. Supervision is additionally included after meals to support “having eaten” and to prevent compensatory behaviors, including vomiting, standing, and exercising.
Behavioral programs reinforce healthy eating by offering privileges or activities following the successful completion of eating goals. As an example, the specialty eating disorders treatment program at Columbia University offers off-unit privileges to patients who consume all of prescribed food, offers additional groups and activities for those at healthier weight ranges, and offers opportunities for brief unaccompanied passes to patients once weight gain goals have been achieved. Part of the power of the behavioral treatment comes from the consistency of the program, which sets standards that all participants can achieve. The expectation that all program participants will aim to eat 100% of prescribed food contributes to the likelihood that the goal will be met and to the overall therapeutic effect. Additionally, the treatment milieu provides group reinforcement for healthy choices. Participants often report that they meet their treatment goals to avoid disappointing their peers as much as for any other motivation for change. In addition to the standard reinforcements of the program, individually tailored reinforcements can be introduced. For example, if an individual demonstrates lack of motivation for the privilege of an off-unit pass, contingencies may be adjusted to reinforce healthy eating and weight with opportunities on the unit (e.g., opportunities for food preparation and cooking on the unit).
Behavioral management for ARFID may require more attention to the individualized assessment of restrictive behaviors and a plan that specifically reinforces successful eating of the restricted foods or reversal of some of the avoidant or restrictive behaviors. Patients who avoid foods because of their sensory characteristics (e.g., smells, textures) may need treatments that expose patients to the specific sensations that are associated with their restrictive eating. Treatments for ARFID need to reinforce reversal of specific eating disturbances in addition to generally reinforcing any required increase in food intake, weight restoration, and improvements in identified nutritional deficiencies.
Behavioral management is incorporated into most intensive treatment programs for individuals with low weight and restrictive eating (see Table 13–1). Intensive treatment programs include inpatient, residential, day treatment, and other intensive outpatient programs that generally require several visits weekly. Less intensive outpatient treatment for individuals with these disorders should similarly include firm behavioral goals. Outpatient treatment may be offered by a team of clinicians, including an internist or pediatrician, a therapist, and a nutritionist. Patients in outpatient treatment—and sometimes their families—should participate in setting treatment goals that reinforce healthy and improved eating behaviors and weight change.
Several specific psychotherapeutic approaches that emphasize behavioral change, including family-based treatment, cognitive-behavioral therapy (CBT), and exposure and response-prevention treatment, have been studied in the treatment of AN and are commonly used in outpatient settings. There are no published treatment studies of ARFID, but behavioral strategies used in other eating disorders are commonly applied to the treatment of ARFID and other conditions that include avoidant or restrictive eating.
Also known as the Maudsley approach, family-based treatment for adolescents with AN is a psychological treatment with solid empirical support (Agras et al. 2014; Lock et al. 2010). Family-based treatment emphasizes participation by all family members and empowers parents to refeed their undernourished child. This outpatient approach aims to help adolescents achieve full weight restoration with normal eating behaviors. When successful, the 6- to 12-month treatment terminates with a transition back to developmentally appropriate autonomy regarding eating and food choice for the weight-restored adolescent.
With stronger support for its effectiveness for the treatment of BN than for AN, CBT is, nevertheless, used by eating disorder clinicians treating low-weight conditions (Fairburn 2008). CBT generally begins with education about the medical and psychological effects of being underweight (see Table 13–2). With attention to treatment alliance and goal setting, the clinician encourages the patient to examine and change behaviors that contribute to the restrictive state and to create a plan for regular, healthier eating. As behavioral change is made and nutritional status improves, the treatment examines and addresses the cognitive distortions that contribute to the individual’s illness. Food monitoring records are a mainstay of treatment, and close attention is paid to actual eating behavior. Over time, thought records and methods for challenging problematic beliefs (i.e., cognitive distortions) are introduced. These techniques have shown modest benefit, although they are not empirically superior to other outpatient psychotherapies that pay close attention to weight and eating behavior (McIntosh et al. 2005). For individuals who have already achieved full weight restoration, CBT has been shown to be more successful than nutritional counseling alone in preventing relapse of AN (Pike et al. 2003).
Exposure and response prevention is the cornerstone of treatment for many anxiety disorders and obsessive-compulsive disorder and is a behavioral treatment strategy that has shown some promise for individuals with AN. The premise of the technique is that individuals need to confront rather than avoid the anxiety-producing stimuli. With incremental exposure to feared stimuli, patients learn that anxiety dissipates over time and that feared consequences do not occur. With this behavioral learning as the mechanism of change, patients practice resisting avoidance behaviors (i.e., response prevention). In treatment of eating disorders specifically, exposure and response prevention targets eating-related anxiety and aims to support patients in confronting rather than avoiding eating-related fears. In inpatient settings, these techniques successfully supported healthier eating behavior and clinical improvement (Simpson et al. 2013; Steinglass et al. 2014a).
Because ARFID is newly described as a distinct diagnosis in DSM-5, no data specific to ARFID are yet available to provide empirical support for treatment. Clinical guidelines suggest that behavioral treatment approaches are likely to be beneficial, because the primary concern is the need to alter behavior. CBT principles are likely to be applicable and successful. CBT, however, is a general therapeutic approach that often needs to be specifically tailored to diagnoses with different features. For example, the CBT manuals for depression differ significantly from those for anxiety disorders. Even within the anxiety disorders category, each diagnosis has its own emphasis for helpful interventions. ARFID is likely a heterogeneous category, and behavioral strategies will need to be tailored differently, depending on the type of illness. The features specific to ARFID, however, suggest particular directions for the development of useful CBT interventions. For example, those individuals whose ARFID symptoms occur within the context of autism spectrum disorder will differ from those whose symptoms are more strongly associated with an anxious temperament or anxiety spectrum. For the anxious patient, interventions may focus on exposure to sensations, whereas for the individual with autism, interventions may focus on consistent meal schedules and positive reinforcement of adequate intake.
Exposure and response prevention, as described in the subsection “Psychotherapeutic Approaches,” may be particularly relevant for treating ARFID. Among some individuals with ARFID, symptoms develop as a conditioned negative response to an experience of eating. These cases of ARFID share many features with specific phobias; however, among individuals with ARFID, the avoidant or restrictive eating behavior has become the primary focus of treatment. Principles of exposure therapy suggest that for each individual, a hierarchy of feared situations can be created. Similar to individuals treated for fear of heights who are gradually exposed to higher and higher floors of a building, individuals with ARFID would begin treatment with eating-related activities that generate low levels of anxiety. As the individual becomes increasingly able to engage in these behaviors, the assignments move toward increasingly higher levels of anxiety. When ARFID includes avoidant or restrictive intake associated with heightened sensory awareness or sensitivity around aspects of eating, exposure to various eating sensations may be necessary. Additionally, techniques that promote awareness of internal bodily sensations may be useful for exposure.
Pharmacological trials in patients with AN are few in number. Owing to the lack of information about the neurobiological mechanisms underlying AN, approaches to medication management in patients with AN have by necessity relied on shared features with other psychiatric illnesses. The high levels of depressive and anxiety symptoms that accompany starvation led to consideration of antidepressants, in particular, as potentially helpful for weight restoration treatment. Many medications appeared promising in case reports or case series, only to prove disappointing when compared with placebo treatment in randomized controlled trials (Hay and Claudino 2012). Meta-analyses have attempted to use the available data from small studies to advance understanding of which strategies may or may not hold promise, and these have similarly shown limited utility of medications (de Vos et al. 2014). These studies highlight the need for rigorous testing of medications, including comparison with placebo, for both AN and ARFID. Although current treatment guidelines have emphasized the lack of utility of medications for AN (Aigner et al. 2011; Watson and Bulik 2013), medications continue to be frequently prescribed, contributing to the cost of treatment and the potential for unwanted complications from medication.
Individuals with AN often present with significant depressive symptoms, including sad mood, hopelessness, and/or anhedonia. Furthermore, early anti-depressants were also associated with weight gain as an unwelcome side effect in non-eating disorder populations. Together, these data suggested the potential utility of antidepressant medications for the treatment of AN. Unfortunately, these strategies proved disappointing. Early trials of tricyclic antidepressants showed no benefit for weight gain (indicating no significant impact on eating behavior) (Biederman et al. 1985; Halmi et al. 1986; Lacey and Crisp 1980), and these medications are associated with cardiac side effects that preclude their use in underweight individuals with AN.
Some of the most influential data have come from a study comparing fluoxetine with placebo in patients with AN receiving behavioral treatment for weight restoration (Attia et al. 1998). This study clearly indicated that fluoxetine offered no benefit over placebo when offered together with a comprehensive weight restoration program. Although all study participants showed improvement in weight as well as in mood and anxiety symptoms during the study period, there were no differences between the fluoxetine-treated and placebo-treated groups.
Underweight individuals with AN are associated with profoundly altered physiology. Therefore, the possibility that antidepressants may confer benefit only after nutritional rehabilitation has been accomplished has been studied separately. Unfortunately, these data have been similarly disappointing. The largest randomized clinical trial among weight-restored individuals with AN showed no benefit of fluoxetine compared with placebo (Walsh et al. 2006). These patients were studied for 1 year after hospital discharge, while receiving CBT aimed at relapse prevention. Fluoxetine again conferred no benefit in rate of relapse or in improvement of psychological symptoms.
Together, these data are very convincing that antidepressants do not significantly improve the treatment of AN.
Individuals with AN commonly struggle with anxiety, specifically around mealtimes. This may be a prominent feature of ARFID as well (Nicely et al. 2014). Because anxiety has been shown to be related to actual food intake (Engel et al. 2013; Steinglass et al. 2010), medications that may reduce anxiety acutely seem worth consideration, such as for individuals during structured treatment. Benzodiazepines are commonly considered as an option to relieve premeal anxiety, yet there are no randomized controlled trials of the clinical utility of benzodiazepines in restrictive eating. The only available data show no benefit of alprazolam compared with placebo in reducing premeal anxiety among a small group of hospitalized patients with AN or in improving their intake in a laboratory meal (Steinglass et al. 2014b). Similarly, in the treatment of obsessive-compulsive disorder, benzodiazepines have not been shown to reduce symptoms (Hollander et al. 2003).
Antipsychotic medications have been considered for the treatment of AN, both for the potential psychological benefits and to capitalize on the weight gain side effects seen in other populations. The concrete, rigid, and near-delusional thought processes seen in AN make this class of medications a compelling possibility for treatment. Early trials were uninformative, because the complications from these medications precluded their use (i.e., seizures, binge and purge symptoms) (Dally and Sargant 1960; Vandereycken 1984; Vandereycken and Pierloot 1982). With the innovation of second-generation antipsychotics and their broad-ranging pharmacology and improved side-effect profile, a new treatment possibility emerged. Initial studies have shown some weight gain benefit of olanzapine among adults with AN (Attia et al. 2011; Bissada et al. 2008), although not in adolescents (Kafantaris et al. 2011). Additionally, it may be that olanzapine relieves some of the obsessionality around eating seen in individuals with AN (Bissada et al. 2008) and thereby may contribute to improved eating. Larger trials will be informative as to whether olanzapine may be a useful treatment for outpatients with AN.
Bone health issues, osteopenia and osteoporosis, are well documented in individuals with AN, and reduced bone density may be the single medical complication that may not fully normalize with complete weight restoration. Bone issues occur in the context of a low-estrogen state, leading to increased bone resorption and poor nutrition, which, in turn, lead to decreased bone formation. Reductions in bone mineral density can be seen on dual-energy X-ray absorptiometry (DXA) as early as 6-12 months after onset of illness (Castro et al. 2000). Results from a more recent study (Faje et al. 2014) suggest that patients with AN carry an increased fracture risk, even in the absence of identifiable areal bone mineral density deficits. A number of pharmacological interventions have been studied, including oral and transdermal hormone replacement, growth factors (i.e., insulinlike growth factor 1), and bisphosphonates. Only one study in adolescents of transdermal estrogen with cyclic progesterone has shown significant promise (Misra et al. 2011). Studies of bisphosphonates suggest that these drugs may offer modest improvement; however, their long half-life and potential impact on a developing fetus make them inappropriate for use in women of reproductive potential.
No medication trials have been done specifically for the treatment of ARFID. Given the similarities in malnourishment between AN and ARFID, it is certainly plausible that medications will be similarly disappointing for ARFID as they have been for AN. However, there may be significant differences in the underlying psychological and neurobiological mechanisms that differentiate these illnesses. Pharmacological treatment studies for ARFID are needed. The prominence of anxiety symptoms and phobic-like traits among individuals with ARFID suggests that anxiolytic medications—and possibly selective serotonin reuptake inhibitors—may be more useful in this population than they have been for individuals with AN. Given the potential promise of second-generation antipsychotics in treating AN, these are worth studying for the treatment of ARFID as well.
Restrictive eating and low weight, associated with eating disorders such as AN and ARFID, require careful clinical evaluation and management. Low weight is associated with many physiological disturbances and substantial medical risk. Low weight is also associated with psychological symptoms that may worsen in the context of nutritional deficiencies. Nutritional rehabilitation and behavioral management, requiring multimodal treatment, are the core components for reversing low weight and normalizing disturbances in eating behavior. Empirical support for specific treatments is limited for AN and entirely absent for ARFID. Novel treatment approaches need to be developed for AN. Descriptive data as well as preliminary data regarding treatment efforts are sorely needed for the recently identified ARFID category.
Key Clinical Points
Agras WS, Lock J, Brandt H, et al: Comparison of 2 family therapies for adolescent anorexia nervosa: a randomized parallel trial. JAMA Psychiatry 71(11):1279-1286, 2014 25250660
Aigner M, Treasure J, Kaye W, et al: World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for the pharmacological treatment of eating disorders. World J Biol Psychiatry 12(6):400-443, 2011 21961502
American Psychiatric Association: Treatment of patients with eating disorders, third edition. Am J Psychiatry 163 (7 suppl):4-54, 2006 1692519
American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 5th Edition. Arlington, VA, American Psychiatric Association, 2013
Attia E, Walsh BT: Behavioral management for anorexia nervosa. N Engl J Med 360(5):500-506, 2009 19179317
Attia E, Haiman C, Walsh BT, et al: Does fluoxetine augment the inpatient treatment of anorexia nervosa? Am J Psychiatry 155(4):548-551, 1998 9546003
Attia E, Kaplan AS, Walsh BT, et al: Olanzapine versus placebo for out-patients with anorexia nervosa. Psychol Med 41(10):2177-2182, 2011 21426603
Baran SA, Weltzin TE, Kaye WH: Low discharge weight and outcome in anorexia nervosa. Am J Psychiatry 152(7):1070-1072, 1995 7793445
Berkman ND, Lohr KN, Bulik CM: Outcomes of eating disorders: a systematic review of the literature. Int J Eat Disord 40(4):293-309, 2007 17370291
Biederman J, Herzog DB, Rivinus TM, et al: Amitriptyline in the treatment of anorexia nervosa: a double-blind, placebo-controlled study. J Clin Psychopharmacol 5(1):10-16, 1985 3973067
Bissada H, Tasca GA, Barber AM, et al: Olanzapine in the treatment of low body weight and obsessive thinking in women with anorexia nervosa: a randomized, double-blind, placebo-controlled trial. Am J Psychiatry 165(10):1281-1288, 2008 18558642
Bühren K, Schwarte R, Fluck F, et al: Comorbid psychiatric disorders in female adolescents with first-onset anorexia nervosa. Eur Eat Disord Rev 22(1):39-44, 2014 24027221
Castro J, Lázaro L, Pons F, et al: Predictors of bone mineral density reduction in adolescents with anorexia nervosa. J Am Acad Child Adolesc Psychiatry 39(11):13 65-1370, 2000 11068 891
Dally PJ, Sargant W: A new treatment of anorexia nervosa. BMJ 1(5188) :1770-1773, 1960 13813846
de Vos J, Houtzager L, Katsaragaki G, et al: Meta analysis on the efficacy of pharmacotherapy versus placebo on anorexia nervosa. J Eat Disord 2(1):27, 2014 25379181
Engel SG, Wonderlich SA, Crosby RD, et al: The role of affect in the maintenance of anorexia nervosa: evidence from a naturalistic assessment of momentary behaviors and emotion. J Abnorm Psychol 122(3):709-719, 2013 24016011
Fairburn CG: Cognitive Behavior Therapy and Eating Disorders. New York, Guilford, 2008
Faje AT, Fazeli PK, Miller KK, et al: Fracture risk and areal bone mineral density in adolescent females with anorexia nervosa. Int J Eat Disord 47(5):458-466, 2014 24430890
Hadigan CM, Anderson EJ, Miller KK, et al: Assessment of macronutrient and mi-cronutrient intake in women with anorexia nervosa. Int J Eat Disord 28(3):284-292, 2000 10942914
Halmi KA, Eckert E, LaDu TJ, et al: Anorexia nervosa: treatment efficacy of cyproheptadine and amitriptyline. Arch Gen Psychiatry 43(2):177-181, 1986 3511877
Hay PJ, Claudino AM: Clinical psychopharmacology of eating disorders: a research update. Int J Neuropsychopharmacol 15(2):209-222, 2012 21439105
Hollander E, Kaplan A, Stahl SM: A double-blind, placebo-controlled trial of clonazepam in obsessive-compulsive disorder. World J Biol Psychiatry 4(1):30-34, 2003 12582975
Kafantaris V, Leigh E, Hertz S, et al: A placebo-controlled pilot study of adjunctive olanzapine for adolescents with anorexia nervosa. J Child Adolesc Psychopharmacol 21(3):207-212, 2011 21663423
Kaplan AS, Walsh BT, Olmsted M, et al: The slippery slope: prediction of successful weight maintenance in anorexia nervosa. Psychol Med 39(6):1037-1045, 2009 18845008
Keys A, Brozek J, Henschel A, et al: The Biology of Human Starvation. Minneapolis, University of Minnesota Press, 1950
Lacey JH, Crisp AH: Hunger, food intake and weight: the impact of clomipramine on a refeeding anorexia nervosa population. Postgrad Med J 56 (suppl 1):79-85, 1980 6994086
Lavender JM, De Young KP, Wonderlich SA, et al: Daily patterns of anxiety in anorexia nervosa: associations with eating disorder behaviors in the natural environment. J Abnorm Psychol 122(3):672-683, 2013 23647124
Lock J, Le Grange D, Agras WS, et al: Randomized clinical trial comparing family based treatment with adolescent-focused individual therapy for adolescents with anorexia nervosa. Arch Gen Psychiatry 67(10):1025-1032, 2010 20921118
Marzola E, Nasser JA, Hashim SA, et al: Nutritional rehabilitation in anorexia nervosa: review of the literature and implications for treatment. BMC Psychiatry 13:290, 2013 24200367
Mayer LE, Schebendach J, Bodell LP, et al: Eating behavior in anorexia nervosa: before and after treatment. Int J Eat Disord 45(2):290-293, 2012 21495053
McIntosh VV, Jordan J, Carter FA, et al: Three psychotherapies for anorexia nervosa: a randomized, controlled trial. Am J Psychiatry 162(4):741-747, 2005 15800147
Misra M, Katzman D, Miller KK, et al: Physiologic estrogen replacement increases bone density in adolescent girls with anorexia nervosa. J Bone Miner Res 26(10):2430-2438, 2011 21698665
National Heart, Lung, and Blood Institute: The Practical Guide: Identification, Evaluation, and Treatment of Overweight and Obesity in Adults (NIH Publ No 00-4084), Bethesda, MD, National Heart, Lung, and Blood Institute, 2000.
Nicely TA, Lane-Loney S, Masciulli E, et al: Prevalence and characteristics of avoidant/restrictive food intake disorder in a cohort of young patients in day treatment for eating disorders. J Eat Disord 2(1):21, 2014 25165558
Pike KM, Walsh BT, Vitousek K, et al: Cognitive behavior therapy in the posthospitalization treatment of anorexia nervosa. Am J Psychiatry 160(11):2046-2049, 2003 14594754
Pollice C, Kaye WH, Greeno CG, et al: Relationship of depression, anxiety, and obsessionality to state of illness in anorexia nervosa. Int J Eat Disord 21(4):367-376, 1997 9138049
Schebendach JE, Mayer LE, Devlin MJ, et al: Dietary energy density and diet variety as predictors of outcome in anorexia nervosa. Am J Clin Nutr 87(4):810-816, 2008 18400701
Simpson HB, Wetterneck CT, Cahill SP, et al: Treatment of obsessive-compulsive disorder complicated by comorbid eating disorders. Cogn Behav Ther 42(1):64-76, 2013 23316878
Steinglass JE, Glasofer DR: Neuropsychology, in Eating Disorders and the Brain. Edited by Lask B, Frampton I. Chichester, UK, Wiley, 2011, pp 106-121
Steinglass JE, Sysko R, Mayer L, et al: Pre-meal anxiety and food intake in anorexia nervosa. Appetite 55(2):214-218, 2010 20570701
Steinglass JE, Albano AM, Simpson HB, et al: Confronting fear using exposure and response prevention for anorexia nervosa: a randomized controlled pilot study. Int J Eat Disord 47(2):174-180, 2014a 24488838
Steinglass JE, Kaplan SC, Liu Y, et al: The (lack of) effect of alprazolam on eating behavior in anorexia nervosa: a preliminary report. Int J Eat Disord 47(8):901-904, 2014b 25139178
Sysko R, Walsh BT, Schebendach J, et al: Eating behavior among women with anorexia nervosa. Am J Clin Nutr 82(2):296-301, 2005 16087971
Trent SA, Moreira ME, Colwell CB, et al: ED management of patients with eating disorders. Am J Emerg Med 31(5):859-865, 2013 23623238
Vandereycken W: Neuroleptics in the short-term treatment of anorexia nervosa: a double-blind placebo-controlled study with sulpiride. Br J Psychiatry 144:288-292, 1984 6367876
Vandereycken W, Pierloot R: Pimozide combined with behavior therapy in the short-term treatment of anorexia nervosa: a double-blind placebo-controlled cross-over study. Acta Psychiatr Scand 66(6):445-450, 1982 6758492
Walsh BT, Kaplan AS, Attia E, et al: Fluoxetine after weight restoration in anorexia nervosa: a randomized controlled trial. JAMA 295(22):2605-2612, 2006 16772623
Watson HJ, Bulik CM: Update on the treatment of anorexia nervosa: review of clinical trials, practice guidelines and emerging interventions. Psychol Med 43(12):2477-2500, 2013 23217606
Wilson GT, Shafran R: Eating disorders guidelines from NICE. Lancet 365(9453):79-81, 2005 15639682
World Health Organization: The Use and Interpretation of Anthropometry (Report of a WHO Expert Committee, WHO Technical Report Series 854). Geneva, Switzerland, World Health Organization, 1995.