III. Clinical Features Influencing the Treatment Plan

text A A A

A. Chronicity of Eating Disorders

In some patients with eating disorders, the illness course becomes chronic, lasting for a decade or more. The care of chronically ill patients is challenging, and modifications in treatment goals may be needed for these patients to benefit. For example, the goals of psychological interventions may be to make small, progressive gains and achieve fewer relapses. Throughout the outpatient care of such patients, communication among professionals is especially important. In addition, more frequent outpatient contact and other supports may sometimes help prevent hospitalizations.

Among patients with a chronic course of anorexia nervosa, many are unable to maintain a healthy weight and experience chronic depression, obsessionality, and social withdrawal. Treatment may require consultation with other specialists; repeated hospitalizations, partial hospitalizations, or residential care; individual or group therapy; other social therapies; trials of various medications; and, occasionally, ECT for patients with severe or treatment-resistant depression. During hospitalizations, expectations for weight gain may be more modest; achieving a safe weight compatible with the patient's life rather than a healthy weight may be all that is possible. The focus of treatment may be on addressing quality-of-life issues (rather than on weight changes or more normal eating habits) and providing compassionate care, with the recognition that patients can realistically achieve only limited goals (125, 304, 305).

Even for patients who have been ill for 20–30 years, there is some evidence that significant benefits can still be derived from treatment. Here too the clinician's approach may need to be modified by relying more on the emotional resources of the patient and much less on the resources of the family. Therapy may need to focus on patients' recognition that their eating disorder has cost them decades of their lives; their therapists may then help them use their energy to improve the remainder of their lives. Some older patients maintain accurate images of their body and recognize that they are too thin but still need significant help with actually gaining the needed weight or relinquishing a strongly established habit of binge eating or purging. The family work often revolves around helping the family adjust to the positive changes that occur with symptom and behavioral changes in the patient (306).


B. Other Psychiatric Factors

1. Substance use disorders

Substance use disorders are common among both women and men with eating disorders (106, 307, 308). Among individuals with bulimia nervosa, 22.9% meet criteria for alcohol abuse (309). Alcohol and other substance use disorders appear to be less common among patients with the restricting type of anorexia nervosa than among those with the binge eating/purging type (310, 311). Binge eating disorder is also associated with high rates of lifetime co-occurring substance abuse, which is more prevalent in male (57%) than in female (28%) patients (313). Patients with co-occurring substance abuse and anorexia or bulimia nervosa appear to have more severe problems with impulsivity in general, including greater risk of shoplifting, suicidal behaviors, self-injurious behaviors, and laxative abuse (83, 314–316). The relation between bulimia nervosa and alcohol abuse/dependence may be indirect and may be influenced by associations with major depression and posttraumatic stress disorder (PTSD) (317).

Available data indicate that patients with eating disorders who have a history of prior but currently inactive substance use disorder respond to standard therapies in the same manner as those without such a history (318–320) and do not appear to experience exacerbation of their substance use after their eating disorder is successfully treated (319). Furthermore, co-occurring alcohol abuse increases the risk of mortality in anorexia nervosa (321). The presence of an active substance use disorder does have implications for the treatment of eating disorders. Patients with co-occurring eating and substance use disorders require longer inpatient stays and are less adherent with treatment after hospitalization than those with substance use disorders alone (322). In everyday clinical practice, substance use shows a strong association with length of treatment required for remission (81). A study of 70 patients with co-occurring eating disorders and substance abuse found that the associated axis III medical disorders reflected complications of both eating and substance use disorders. Where treatment staff are skilled in treating both disorders, concurrent treatment should be attempted.

2. Mood and anxiety disorders

A high percentage of treatment-seeking patients with eating disorders report a lifetime history of major depression (313, 323–327). Nutritional insufficiency and weight loss often predispose these patients to symptoms of depression (279). Depressed individuals with an eating disorder experience higher levels of anxiety, guilt, and obsessionality but lower levels of social withdrawal and lack of interest than depressed individuals without eating disorders (328). The presence of co-occurring depression at initial presentation has minimal or no predictive value for treatment outcome (329). Although antidepressant medications may be relatively ineffective in treating depressive symptoms before a patient has gained weight (174) and are typically not helpful in weight restoration (see Section II.B.2.c.i. ), their use has sometimes been associated with reduced symptoms of depression and anxiety during the weight restoration phase (176). In addition, the experience of many clinicians suggests that severe depression can impair a patient's ability to become meaningfully involved in psychotherapy and may dictate the need for medication treatment for the mood symptoms from the beginning of treatment.

Lifetime prevalence rates for anxiety disorders also appear to be higher for patients with anorexia or bulimia nervosa. In a controlled study with a large sample, 71% of women with anorexia or bulimia nervosa had at least one anxiety disorder, a rate significantly higher than that found in the control group (330), although rates for specific anxiety disorders varied (331). Social phobia and obsessive-compulsive disorder (OCD) are the anxiety disorders most commonly described in patients with anorexia nervosa. For those with bulimia nervosa, co-occurring presentations of social phobia, OCD, PTSD, or simple phobia are most often described (332, 333). Patients with binge eating disorder also have high rates of lifetime anxiety disorders (29%), with simple phobia and panic disorder being the most commonly described (334–340). Overanxious disorders of childhood are also common in conjunction with anorexia and bulimia nervosa, and anxiety disorders often precede the onset of these eating disorders (82, 341). Youngsters with anorexia nervosa may have co-occurring OCD more frequently than adults with anorexia nervosa (342). Although there is no clear evidence that co-occurring anxiety disorders significantly affect eating disorder treatment outcome, such comorbid problems should be addressed in treatment planning.

Available data on the extent of PTSD among patients with eating disorders suggest an association between childhood trauma and bulimia nervosa (333, 343–346). Although specific causal links have not been demonstrated and the mechanisms of association and potential transmission remain unclear, it is thought that early trauma may sensitize some individuals to later traumatic experiences and an array of impulsive behaviors, including eating disorder symptoms. The lifetime rate of PTSD among women with bulimia nervosa reported in studies has varied between 6% and 45% (330, 332, 346, 347). Clinical consensus suggests that the extent of trauma history and the possible presence of PTSD should be taken into consideration in treatment planning.

3. Personality disorders

Temperament, genetics, family pathologies, and other dimensions of personality appear to be involved in the occurrence and manifestations of eating disorders (348–350). In all cases, personality types and the extent of personality pathology have implications for treatment (79, 351, 352). As with mood and anxiety disorders, the presence of comorbid personality pathology (whether or not it is severe enough to meet DSM criteria for a personality disorder) is the norm rather than the exception (353). Patients with eating disorders should routinely be assessed for concurrent personality disorders. The reported prevalence of personality disorders has varied widely across eating disorders and studies. Individuals with anorexia nervosa tend to have higher rates of Cluster C personality disorders, particularly obsessive-compulsive, perfectionistic, and avoidant and histrionic traits, whereas normal-weight patients with bulimia nervosa are more likely to display features of Cluster B disorders, particularly impulsive, affective, and narcissistic traits (354–361). The presence of borderline personality disorder seems to be associated with a greater disturbance in eating attitudes, a history of more frequent hospitalizations, and the presence of other problems such as suicidal behaviors and self-injurious behaviors (356, 360). The presence of borderline personality disorder is also associated with poorer treatment outcome and higher levels of psychopathology at follow-up (362, 363). The improvement of axis I eating disorder symptoms may produce concurrent improvement in traits associated with these personality disturbances (364). Although this approach has not yet been systematically studied, clinical consensus strongly suggests that the presence of a co-occurring personality disorder, particularly borderline personality disorder, dictates the need for longer-term therapy that focuses on the underlying personality structure, coping strategies, and interpersonal relationships in addition to the symptoms of the eating disorder. Recent naturalistic research documents the generally poorer outcome and longer treatment required for bulimia nervosa patients with borderline, avoidant, or more broadly emotionally dysregulated personality pathology, although many such patients do improve substantially with longer-term treatment (365).


C. Concurrent General Medical Conditions

1. Type 1 diabetes

Eating disorders may be more common among patients with type 1 or type 2 diabetes than among nondiabetic patients (366, 367), but this contention is not strongly supported by research (368). It has been suggested that type 1 diabetic patients are more likely to have anorexia nervosa or bulimia nervosa (368a) and type 2 diabetic patients are more likely to have binge eating disorder (367). Although patients with both an eating disorder and diabetes are more likely to be female, males with both disorders are also commonly found, particularly among type 2 diabetic patients. The presentation of eating disorders in the context of diabetes may be substantially more complex than that seen with an eating disorder alone. In addition, these concurrent diagnoses may present as numerous general medical crises before the presence of the eating disorder is diagnosed and treated, particularly among patients with type 1 diabetes. Evidence suggests that rates of diabetic complications are higher when insulin-dependent diabetes mellitus co-occurs with bulimia nervosa or EDNOS (369). Mortality rates are much higher with combined anorexia nervosa and type 1 diabetes than with either condition alone (370). Patients with uncontrolled diabetes who also have bulimia nervosa may require a period of inpatient treatment for stabilization of both illnesses (371, 372). Treatment of these concurrent disorders may also require more interaction with general medical specialists. Lengthy inpatient treatment for adult diabetic patients with eating disorders (primarily bulimia nervosa) is promising but has been studied in only a few patients (373).

Diabetic patients with eating disorders often underdose their insulin to lose weight. Throughout the studies undertaken to date, insulin omission has been found to be common (374). It has been suggested that insulin omission be considered a specific type of purging behavior in the next DSM revision.

2. Pregnancy

Eating disorders may begin de novo during pregnancy, but many patients become pregnant even while they are actively symptomatic with an eating disorder. Behaviors associated with eating disorders, including inadequate nutritional intake, binge eating, purging by various means, and the use or misuse of some teratogenic medications (e.g., lithium, benzodiazepines, divalproex, diet pills), can all result in fetal or maternal complications (375). The care of a pregnant patient with an eating disorder is difficult and usually requires the collaboration of a psychiatrist and an obstetrician who specializes in high-risk pregnancies (376–379).

In a recent study of 49 nulliparous women previously diagnosed with an eating disorder (24 with anorexia nervosa, 20 with bulimia nervosa, 5 with EDNOS), 22% had a verified relapse of eating disorder during pregnancy. Compared with a control group, women with past or current eating disorders were at increased risk of hyperemesis, infants with significantly lower birth weight, smaller head circumference (including microcephaly), and small-for-gestational-age infants (380). Women with active symptoms of anorexia or bulimia nervosa are at higher risk for Cesarean section and postpartum depression (381). Yet another study linked the presence of an eating disorder before pregnancy or an eating disorder (or dieting or fasting) during the first trimester with an increased risk to the infant, including a greater risk of low birth weight, preterm delivery, a small-for-gestational-age size, and neural tube defects (382).

Data on the impact of pregnancy on eating disorder symptoms are conflicting. One study has shown that among treated bulimia nervosa patients, childbirth is not associated with increased symptoms (383). Another study found that anorexia and bulimia nervosa patients had a decreased severity of eating disorder symptoms during pregnancy and that patients with bulimia nervosa, but not patients with anorexia nervosa, maintained this improvement through 9 months postpartum (384).

Prepregnancy counseling of eating disorder patients and their families should be detailed and cautionary. Although some patients may be able to eat normally and decrease binge eating and purging during their pregnancy, it is best for the eating disorder to be treated before the pregnancy if possible. Although women with a lifetime history of anorexia nervosa may not have reduced fertility, they do appear to be at risk for a greater number of birth complications than comparison subjects and of giving birth to babies of lower birth weight, whether or not the anorexia nervosa is active at the time of pregnancy (312). Mothers with eating disorders may have more difficulties than others in feeding their babies and young children than mothers without eating disorders and may need additional guidance, assistance, and monitoring of their mothering (385–388). In one study, primiparous mothers with an episode of eating disorder during the first postpartum year often expressed negative emotions toward their infants during meals, the infants' emotional tone was more negative, and the infants' mealtimes were more conflictual than those of control infants (388).

Active bulimia nervosa may be associated with polycystic ovaries or other follicular abnormalities. In a study of eight women with bulimia nervosa, six of the seven women with active bulimic symptoms showed polycystic ovaries; at follow-up, the five who continued to have bulimic symptoms all had polycystic ovaries, whereas normal ovaries were seen in the three whose bulimia nervosa had remitted (389).

Some patients with an eating disorder may present to fertility clinics and request medications to facilitate conception. Although little is known about the consequences of fertility drugs in patients with anorexia nervosa, the simultaneous multiple pregnancies that often occur with these drugs may be even more difficult for patients with an eating disorder to contend with than for women without an eating disorder.


D. Demographic Variables

1. Male gender

Although eating disorders are more prevalent in women, males with eating disorders are not rare, and case series often report on hundreds of male patients (390, 391). The stereotype that eating disorders are female illnesses may limit a full understanding of the scope and nature of problems faced by male patients with eating disorders.

There continues to be a paucity of information regarding males with eating disorders (392). However, more males may be seeking treatment (393). The most common form of eating disorders among males may be EDNOS (394).

In community samples, males with eating disorders have more psychiatric comorbidity and psychosocial morbidity than females with eating disorders (395). Studies of national samples of girls and boys exposed to physical and sexual abuse have shown that although binge and purge behaviors were nearly twice as prevalent among girls (13%) as boys (7%), boys who had experienced both physical and sexual abuse were nearly twice as likely as girls to report these behaviors (odds ratios 8.25 and 4.28, respectively) (396). Especially in the bulimia nervosa and binge eating disorder subgroups, males with eating disorders who present to tertiary care centers may have greater rates of co-occurring substance use disorders and more frequently have a history of having premorbid obesity or being overweight (397). As in women, there is a higher risk of osteoporosis in men with eating disorders (398, 399).

Although many similarities exist between males and females with different types of eating disorders, notable differences have been reported. Females with anorexia or bulimia nervosa score higher on the Drive for Thinness subscale than do males, and this appears to be a real difference; females also score higher on the body dissatisfaction subscale than do males, but this may result from a failure of these scales to address the specific ways in which males are dissatisfied with their bodies (e.g., males seem to be more concerned with upper torso and muscular development than are females). Studies of binge eating disorder patients demonstrate that women have greater body image dissatisfaction and are more likely to cope with negative affect by binge eating, whereas men have higher rates of drug and alcohol use disorder histories (400).

Body dysmorphic disorder (BDD), a severe form of body image disturbance seemingly related to OCD, is common among patients with anorexia nervosa (401) and appears to be increasing in prevalence. Although BDD is nearly as common in men as in women, the focus of body preoccupations may differ by gender (402, 403). Muscle dysmorphia, a distressing or impairing preoccupation with the idea that one's body is not sufficiently lean and muscular, is a form of BDD that occurs primarily in men, often involves abnormal eating behavior, and appears to overlap in other ways with eating disorders. The relation of BDD to eating disorders other than anorexia nervosa is poorly understood (404). In one small inpatient study, anorexia nervosa patients with BDD were found to have significantly poorer functioning, more episodes of psychiatric hospitalization, and three times the rate of lifetime suicide attempts compared with anorexia nervosa patients without BDD (401).

Although a patient's gender per se does not appear to influence the outcome of treatment, some aspects of treatment may need to be modified on the basis of gender. For example, with regard to personality traits, males with eating disorders have somewhat less perfectionism, harm avoidance and reward dependence behaviors, and cooperativeness than females (405). Open-blind studies suggest that bringing testosterone levels to within normal ranges in males during nutritional rehabilitation for anorexia nervosa may be helpful in increasing lean muscle mass, but definitive studies on this have not been completed.

Although studies in clinical samples have suggested that a higher prevalence of homosexuality may exist among males with eating disorders (390), this finding has not yet been confirmed epidemiologically. Nevertheless, because issues concerning sexual orientation are not uncommon among males with eating disorders seen in clinical settings, these issues should be considered in treatment (390).

Where possible, therapy groups restricted to male patients may address some of these patients' specific needs and help them deal with potential stigmatization of male patients by female patients in treatment. Male patients with anorexia nervosa may require higher energy intakes (up to 4,000–4,500 kcal/day) during nutritional rehabilitation because they normally have higher lean body mass and lower fat mass compared with female patients. Further, because they are generally taller and larger framed to begin with, males with anorexia nervosa often require much larger weight gains to return to a healthy weight (391, 399).

2. Age

Although most eating disorders start while patients are in their teens and 20s, earlier and later onsets have been observed as well. Among the youngest patients with early onset (i.e., ages 7–12 years), obsessional behavior and depression are common (406). Children often present with physical symptoms such as having nausea, experiencing abdominal pain, feeling full, or being unable to swallow (all conditions requiring evaluation to investigate a variety of potential etiologies) (407). Their weight loss can be rapid and dramatic. Similarly, their lack of expected weight gain is problematic. Food avoidance for primarily psychological reasons that results in weight loss (food avoidance emotional disorder [FAED]) has been delineated and differentiated from anorexia nervosa in young children. These young patients know that they are underweight, would like to be heavier, may not know why this goal is hard to achieve, and usually show more generalized anxiety unrelated to food. A small number of FAED patients go on to develop eating disorders, but direct continuity between the two types of disorders has not been demonstrated (408). Other syndromes to be considered are selective eating disorder, functional dysphasia, and pervasive refusal syndrome (407).

Children with early-onset anorexia nervosa may have delayed growth (3, 97, 98, 409–412) and be especially prone to osteopenia and osteoporosis (413, 414). In a few cases, exacerbations of anorexia nervosa and OCD-like symptoms have been associated with pediatric infection-triggered autoimmune neuropsychiatric disorders (415), but further research is needed to determine the nature of this association (416). Bulimia nervosa is rarely seen in children under age 12 years (417–420).

With respect to middle-aged patients, case reports and clinical consensus now suggest that as the baby boomer generation grows older, body image concerns and eating disorders are becoming more prevalent. One case report from an established residential program noted a shift in the age of its patients and reported treating more middle-aged women than a decade ago (417–420).

At the other end of the age spectrum, incidence rates for anorexia nervosa among women over age 50 years are low, accounting for <1% of patients with newly diagnosed anorexia nervosa (421). Anorexia nervosa has been reported in elderly patients in their 70s and 80s; these are generally women in whom the illness has been present for 40 or 50 years. In many cases, the illness started after age 25 (so-called anorexia tardive). In some case reports, adverse life events such as deaths, a marital crisis, or a divorce have been found to trigger these older-onset syndromes. The fear of aging has also been described as a major precipitating factor in some patients (142, 422). Rates of co-occurring depression have been reported to be higher among these patients in some studies but not in others (423). Regardless of the age at onset, concerns about comorbid medical conditions, especially osteopenia and osteoporosis, take on greater significance with older patients.

3. Cultural factors

Specific pressures and values concerning weight and shape vary among different cultures. The quest for beauty and acceptance in accordance with the stereotypes women perceive in contemporary media is leading increasing numbers of women around the world to develop attitudes and eating behaviors associated with eating disorders. Despite initially higher prevalence rates of this phenomenon in Western and postindustrial societies, disordered eating behaviors now appear to be globally distributed (424–427).

Popkin and Gordon-Larsen (428) have described a phenomenon called the "nutritional transition" in economically emerging countries. In some Asian and Middle Eastern countries, the decrease in activity levels and increase in access to high-fat, high-calorie foods have occurred much more rapidly than in the United States; likewise, this trend has led to a faster rate of increase in the number of individuals becoming overweight or obese than that seen in the United States. Consequently, the pressure to lose weight or remain slender may ultimately affect an even wider group of people more quickly in these regions than it has in the United States. Emerging population data suggest that transnational migration and modernization may increase the risk of disordered eating and body dissatisfaction across diverse ethnic and social contexts (427, 429–432).

Clinicians should engage female patients from non-Western cultures in informed and sensitive discussions regarding their struggles, experiences, and personal perceptions about what it means to be attractive and satisfied with their bodies (433). Clinicians should be sensitive to and inquire about how weight and shape concerns are experienced by patients, especially those who are from minority populations or non-Western or other cultural backgrounds or are transitioning and assimilating into Western societies. Transcultural differences in the meanings of terms and concepts are considerable (434). It is not uncommon to find Asian patients with atypical behaviors, such as denying a fear of weight gain, appraising their bodies as malnourished, denying distorted perceptions of their bodies, and specifically denying a drive for thinness (435). (Atypical patients who specifically deny a drive for thinness are also seen among North American and European populations [436].)

Although little information exists on cross-cultural differences among males with regard to eating disorders, male patients with anorexia nervosa, bulimia nervosa, and binge eating disorder have been described in non-Western populations. Some population differences across cultures among males have been found with regard to attitudes about eating, body shape, and weight (437, 438).

4. Eating disorders in athletes

The risk for eating disorders is greater among competitive athletes than in the general age-matched population (439–441). Female athletes in competitive sports that emphasize a thin body or appearance, such as gymnastics, figure skating, and distance running, are especially at risk, as are female ballet dancers. Male athletes in sports such as bodybuilding and wrestling are also at greater risk. Certain antecedent factors such as cultural preoccupation with thinness, performance anxiety, and athlete self-appraisal may predispose a female athlete to body dissatisfaction, which may mediate the development of eating disorder symptoms (442). Parents and coaches of young athletes may support distorted shape and eating attitudes in the service of guiding the athlete to be more competitive. Although competitive college athletes appear to be at greater risk for eating disorders than nonathlete peers, this has not been demonstrated among high school athletes (443). Girls as young as age 5 years who participate in aesthetic sports, such as ballet or figure skating, have exhibited greater weight concerns than girls who participate in nonaesthetic sports or who do not participate in sports (444). Participation in sports may protect some young athletes from developing eating disorders if, for example, they approach their sports in a sensible way and derive appropriate pride and self-esteem from their achievements (443, 445).

Physicians working with adolescent and young adult athletes, particularly competitive athletes participating in the at-risk sports mentioned above, must be alert to early symptoms of eating disorders. Simple screening questions about weight, possible dissatisfaction with appearance, amenorrhea, and nutritional intake on the day before a physical evaluation may help identify an athlete who is developing an eating disorder. Early general medical and psychiatric intervention is key to prompt recovery from the disorder.

Extreme exercise appears to be a risk factor for developing anorexia nervosa, especially when combined with dieting (446–448). A "female athlete triad" has been identified, consisting of disordered eating (including the full spectrum of eating-related problems from simple dieting to clinical eating disorders), amenorrhea, and osteoporosis (449). The exact relation between the triad and clinical eating disorders is not fully understood, but athletes with the triad would meet diagnostic criteria for EDNOS. The prevalence of the triad among collegiate athletes has been studied; menstrual irregularity has been found in nearly one-third of female athletes not using oral contraceptives, and athletes participating in aesthetic sports (e.g., gymnastics) have higher scores on the 26-item Eating Attitudes Test (31, 32) than athletes in endurance or team sports (450). Similarly, an "overtraining syndrome," described as a state of exhaustion, depression, and irritability in which athletes continue to train but their performance diminishes, has been observed (451, 452). Both the female athlete triad and the overtraining syndrome parallel the "activity anorexia" syndrome that has been observed in animal models (453, 454).

5. Eating disorders in high school and college students

Eating disorders are common among female high school and college students. From a primary prevention perspective, health professionals may be called on to provide information and education about eating disorders in classrooms, athletic programs, and other extracurricular venues. However, the efficacy of such educational programs in reducing eating disorders is still uncertain (455, 456). Health professionals who serve as trainers, coordinators, and professional supports for peer counseling efforts conducted at school, in dormitories, and through other campus institutions may help in early intervention. Through student health and psychological services, health professionals may serve as initial screeners and diagnosticians and help manage the treatment of students with eating disorders of varying levels of severity (457).

Psychiatrists may be occasionally called on as clinicians and agents of the school administration to offer guidance in the management of impaired students with serious eating disorders. In such situations, the suggested guidelines for levels of care described in Table 8 should be followed. According to the guidelines, students must be treatable as outpatients to stay in school. It is advisable that students be required to take a leave of absence if they are severely ill (457, 458). Students should be directed to inpatient hospital care if their weight is considerably below an expected healthy weight and they meet the other indications for hospitalization listed in Table 8.

For students with serious eating disorders who remain in school, it is useful for the psychiatrist and other health care professionals to work with the school's administration toward developing policies and programs that make student attendance contingent on participation in a suitable treatment program. When the psychiatrist and other health care professionals serve as "dual agents" for the school and whenever information must be shared among health care professionals, appropriate consents must be obtained and the requirements of the Health Insurance Portability and Accountability Act followed. For severely ill students, the clinical team must include a general medical clinician who can gauge patients' physical safety and monitor their weight, vital signs, and laboratory indicators. For the student to be permitted to continue in school, these clinicians may require a minimum weight and other physical, behavioral, or laboratory target measures to ensure basic medical safety. An explicit policy should be developed specifying that clinicians have the final say regarding the student's participation in physically demanding activities (e.g., organized athletics). Restrictions must be based on actual medical concerns. Procedures should be in compliance with the school's policies regarding management of students with psychiatric disabilities and the Americans With Disabilities Act (458).

6. Identification of risk and protective factors

Many efforts have been made to understand how eating disorders develop. Overall, our understanding of risk and vulnerability still outweighs our knowledge of protective factors and resilience. Temperamental factors, eating dysregulation, attachment issues, deficient self-regulation, childhood abuse in the case of bulimia nervosa, and sociocultural ideals of health and beauty may all contribute to risk and pathogenesis (367, 459, 460). High rates of childhood anxiety disorders precede eating disorders, especially overanxious disorder and OCD for anorexia nervosa and overanxious disorder and social phobia for bulimia nervosa; this could be of potential clinical relevance, especially when treating children and adolescents (341). A history of childhood obsessive-compulsive traits—notably perfectionism, rigidity, and rule-bound behavior—may also be associated with an increased risk for the development of an eating disorder (461). Richly documented clinical histories of patients with anorexia nervosa followed over 30 years from infancy to early midlife suggest several potential risk factors related to early perceived body image distortions, body regulatory problems, and academic and interpersonal problems (146).

Because it is well known that the risk of eating disorders is transmitted in families, it is important to offer particular help to patients with eating disorders who are themselves mothers. Attention should be paid to their mothering skills and attachment styles and to their offspring to minimize the risk of eating disorders being transmitted (386–388, 462, 463).

In some patients, increasingly compulsive exercise may precipitate anorexia and bulimia nervosa (447, 464). Female athletes in certain physical activities such as ballet and gymnastics are especially vulnerable (465). Unlike habitual runners, ballet dancers exhibit eating pathology similar to that of individuals with eating disorders (466). Male bodybuilders are also at risk, although the symptom picture often differs because bodybuilders may emphasize a wish to "get bigger" and may also abuse anabolic steroids to reach their goal (467, 468). This has been called the "Adonis complex" (469, 470).

Programs have been designed to influence these risk factors and thus prevent eating disorders. A recent meta-analysis (471) of prevention programs suggests that programs vary significantly in their impact, ranging from an absence of any effect to a reduction in current and future eating pathology. Some effects persist as long as 2 years and are superior to minimal-intervention control conditions. Larger effects occur for selected (versus universal), interactive (versus didactic), and multisession (versus single-session) programs; programs offered solely to female patients and to participants over age 15 years; programs without psychoeducational content; and trials that use validated measures. Several of these programs have resulted in enhanced knowledge about eating disorders and healthy eating. A few have resulted in improved attitudes toward size, shape, eating, and weight, and some have addressed self-esteem and "weightism" and stressed normal/healthy attitudes and behaviors. Other approaches have resulted in modest changes in eating- and weight-related behaviors (472–474), but these changes are not always sustained (455, 456, 475). Because some studies even suggest that certain preventive efforts actually increase the likelihood that maladaptive eating behaviors would be attempted, particularly among adolescents (455), caution is recommended in selecting target populations for such intervention and proper follow-up is necessary.


Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).
Related Content
Dulcan's Textbook of Child and Adolescent Psychiatry > Chapter 1.  >
Dulcan's Textbook of Child and Adolescent Psychiatry > Chapter 2.  >
Psychiatric News
PubMed Articles
Environment, human reproduction, menopause, and andropause. Environ Health Perspect 1993;101 Suppl 2():91-100.
  • Print
  • PDF
  • E-mail
  • Chapter Alerts
  • Get Citation