A Complicated Process: Diagnosing and Treating Anorexia Nervosa and Bulimia


Diagnosis of the two main major eating disorders, anorexia nervosa and bulimia nervosa, can be difficult because of denial of symptomatology by the patients and problems with some of the diagnostic criteria. Although CBT has been the most effective, there are no treatments available that can guarantee a cure for either disorder. Medication is only a helpful adjunct to the treatment of anorexia, while many controlled studies that show antidepressant medications are effective in reducing binge/purge behavior in bulimia.

Psychiatric Times

May 2005


Issue 6

Sponsored by CME LLC for 1.5

Category 1 credits.

Original release date 6/05. Approved for CME credit through 5/31/06.

Educational Objectives:

Upon completion of this educational activity, the reader will be familiar with:

  • Diagnostic criteria of anorexia nervosa, bulimia nervosa and partial eating disorder syndromes.
  • Other psychiatric disorders that are comorbid with AN and BN.
  • Pharmacological and behavioral treatments for AN and BN.

Who will benefit from reading this article?

Psychiatrists, primary care physicians, neurologists, nurse practitioners, psychiatric nurses and other mental health care professionals. Continuing medical education credit is available for most specialties. To determine if this article meets the CE requirements for your specialty, please contact your state licensing board.

Anorexia nervosa (AN) and bulimia nervosa (BN) are the two major eating disorders. Anorexia nervosa is characterized by weight loss, an intense fear of gaining weight, a distorted body image and amenorrhea. Bulimia nervosa is a disorder in which binge eating is the predominant behavior. People with BN engage in some sort of compensatory behavior to counteract the potential weight gain from calories ingested during bingeing. They are frequently overconcerned about their physical appearance.

Why are these disorders difficult to diagnose? Patients with AN do not wish to be diagnosed because they do not wish to be treated. They deny the core symptomatology of their disorder and will often try to mislead their primary care provider into believing there are other medical issues causing their weight loss. Thus, it is very helpful to obtain further information from family members about the patient's behavior, if possible. Furthermore, there are problems with the DSM-IV diagnostic criteria for AN and BN (see Table 1 for an abbreviated version). The DSM-IV criteria are undergoing scrutiny because many patients suffering from eating disorders do not quite fit.

Diagnosis of Anorexia Nervosa

For AN, the criterion of weight loss would seem to be noncontroversial. However, there is no consensus as to how weight loss should be calculated. Some investigators emphasize a total weight loss from an original high weight, and others emphasize weight loss below a normal weight for age and height. Surprisingly, the degree of weight loss does not differentiate patients with AN on other characteristic clinical variables (Halmi, 1974). The psychological criteria for AN are also a problem. Through the Internet, patients with AN have learned not only diagnostic criteria but how to fool doctors both in diagnosis and in treatment. For example, they will deny the fear of gaining weight even though their behavior-such as eating very small amounts of food, fasting for long periods of time and excessive exercise-indicates they have such a fear. Thus, a positive evaluation on this criterion is often based on an inferred judgment from reportable observable behaviors.

Psychological assessments have shown that patients with AN have overwhelming maturity fears, lack of confidence in coping with life's problems and a pervasive sense of inadequacy (Kleifield et al., 1996). Staying thin is one behavior anorectic patients engage in better than anyone else, and, thus, they achieve some feeling of accomplishment by evaluating themselves in terms of their thinness. The constant preoccupation with dieting, food, weight and body image is a distraction that may well be a defense against having to face other life problems. Thus, to admit to the seriousness of their low body weight would mean acknowledgement of the necessity to change their behavior, which is an overwhelming and terrifying notion for the patient with AN.

The entire issue of amenorrhea is complicated by the fact that it is often difficult to get an accurate history of menstrual patterns. The use of birth control medication also makes evaluation of menstruation difficult. Because of this and a few examples of women who still menstruate at a low weight, there is current support for excluding the criterion of amenorrhea for diagnosis of AN.

The restricting and binge-eating/purging subtypes of AN were included in the diagnostic criteria after many studies showed both behavioral and medical differences (Halmi, 2002). Patients with the bingeing and purging subtype have a much higher association of impulsive behavior such as suicide attempts, self-mutilation, stealing and substance abuse--including alcohol abuse--compared to those with the restricting subtype. Impulsive personality disorders such as borderline personality disorder and histrionic personality disorder are far more prevalent in the bingeing and purging subtype compared to the restricting subtype. Also, the medical problems associated with bingeing and purging behavior are distinctly different from those of restrictors (Table 2).

Diagnosis of Bulimia Nervosa

Problems also exist with the diagnostic criteria for BN. The term bulimia merely means binge eating, which is a behavior that may occur on occasion in otherwise healthy people. It is difficult to define exactly how much food constitutes a binge. Obviously, a very small person eating a certain amount of food is proportionately going to be greater than a very tall person eating that same amount of food. Therefore, the definition of binge eating remains rather vague; that is, a greater than expected amount of food in a discrete period of time, such as under two hours. Earlier definitions of BN did not contain a frequency and chronicity criterion, which resulted in early population prevalence studies producing a rather high prevalence for BN, especially in vulnerable groups such as college students (Healy et al., 1985; Zuckerman et al., 1986). It then became necessary to separate those students who binged occasionally from those binge eaters who had a distinct impairment of function.

Subsequently, rather arbitrary criteria based on clinical experience for frequency and chronicity were placed in the DSM-IV criteria for BN. There is some question as to whether a person who binges and purges once a week is equally impaired as a person who binges and purges twice a week. Important clinical features of BN are the feeling of lack of control over eating and the regularity of binge eating. Many of these patients are very concerned about body shape and weight, which is another criterion for the diagnosis. However, there appears to be a subgroup for whom bingeing and purging is an addictive phenomena to relieve anxiety (Lynch et al., 2000; Powell and Thelen, 1996). This latter group is not well studied or defined.

It is assumed that patients with a diagnosis of BN are within 10% of normal weight range. If they are below 15% of a normal weight range, they usually meet criteria for AN, binge/purge subtype. To further clarify the diagnosis of BN, two subtypes were defined: the purging subtype, who self-induces vomiting and abuses laxatives, diuretics or enemas and the non-purging subtype, who uses compensatory behaviors to combat caloric intake by exclusively fasting or excessive exercise. Most patients with the non-purging type of BN are in the upper part of a normal weight range or are actually overweight. The criteria for BN will continue to be revised until ongoing studies of the biology of the disorder, including neuroimaging and genetic studies as well as longitudinal course studies, are completed.

Partial Eating Disorder Syndromes

The current DSM-IV diagnostic system recognizes that there are many variants of AN and BN that do not meet the specific criteria for these disorders but involve significant impairment of function. Thus, the diagnostic system simply states that the category of eating disorder not otherwise specified (EDNOS) is for disorders of eating that do not meet those criteria. An example is binge-eating disorder, which is defined as recurrent episodes of binge eating in the absence of the regular use of the inappropriate compensatory behaviors characteristic of BN. The major problem with binge-eating disorder is distinguishing it from the non-purging subtype of BN. It is emphasized that the diagnosis should be considered only when the individual reports a subjective sense of impaired control during episodes of overeating. Many of these individuals are obese and eat throughout the day in addition to their binge-eating episodes. Most have long histories of repeated efforts to diet, and others have given up all effort to diet. According to the DSM-IV, about 30% of individuals in weight control programs will have binge-eating disorder. Other examples of EDNOS are individuals who are within a normal weight range, eat normal size meals and then regularly purge.

Studies of the partial syndrome cases have produced discrepant results. In one study, the severity of eating disorder symptoms in people with partial syndrome cases was as great as or greater than those with full syndrome cases (Martin et al., 2000). In another study, people with partial syndromes or subclinical eating disorders had lower severity scores than those with full syndrome eating disorders (Cotrufo et al., 1998). A third study showed that some individuals, over time, progressed from the less to the more severe disturbances in eating behavior (Shisslak et al., 1995).

Psychiatric Comorbidities

Depression is common in both AN and BN. Two-thirds to three-fourths of these patients will have a lifetime history of depressive disorder (Braun et al., 1994; Ivarsson et al., 2000; Wade et al., 2000). In both AN and BN, one-half to two-thirds of patients will have a lifetime history of an anxiety disorder (Braun et al., 1994). Obsessive-compulsive behaviors often develop or become worse as AN progresses in severity. An obsession with cleanliness, an increase in cleaning activities and compulsive studying are commonly observed in these patients. The actual occurrence of lifetime obsessive-compulsive disorder as defined by DSM-IV criteria is only about 20% of patients with AN.

Several studies have shown a greater prevalence of OCD in the first-degree relatives of patients with eating disorders compared to controls (Cavallini et al., 2000; Halmi et al., 1991). A higher morbidity risk for obsessive-compulsive spectrum disorders in the first-degree relatives of patients with eating disorders compared to those of controls has led to the proposal that OCD and eating disorders are phenotype expressions of a common liability. In another family study, the risk of obsessive-compulsive personality disorder was elevated only among relatives of anorectic probands, evidence that these two disorders may have shared familial risk factors (Lilenfeld et al., 1998). Therefore, obsessional personality traits may be specific familial risk factors for AN. One of the common features of obsessive-compulsive personality disorder is perfectionism. In another study, patients with AN had significantly higher perfectionism ratings compared with controls (Halmi et al., 2000). The authors concluded, "Perfectionism is likely to be one of a cluster of phenotypic trait variables associated with a genetic diathesis for anorexia nervosa."

About one-third of patients with BN will have a lifetime history of alcohol or drug abuse (Bushnell et al., 1994). In another study, the relationship between substance abuse in patients with BN was found to be far stronger than in patients with AN (Holderness et al., 1994).

Although many of the studies of personality disorders and eating disorders are contradictory, almost all have shown a high preponderance of Cluster B (impulsive) personality disorders associated with bulimic subtypes compared with the anorectic restrictors (Braun et al., 1994; Herzog et al., 1992). In the 1994 study by Braun et al., borderline personality disorder was present in 26% of the bulimic subgroups and was the most common Cluster B condition. In that study, Cluster C (anxious) personality disorders were present in 26% of the sample and did not vary according to eating disorder subtype. Steiger et al. (1994) found that personality disorder classification did not predict the severity of bulimic symptoms or the responsiveness to treatment of bulimic symptoms.

It is necessary to diagnose comorbid psychiatric disorders since they may complicate treatment strategies.

Treatment Options

The intensity of treatment for the patient with an eating disorder will depend upon the severity of illness. The latter is determined by the patient's weight, medical status and other psychiatric comorbid problems. Seriously ill patients will require a specialized eating disorder inpatient unit where intensive medical management and/or monitoring for suicidal and impulsive behaviors can be conducted. Medical management requires weight restoration; nutritional rehabilitation; rehydration and correction of serum electrolytes; and daily monitoring of weight, food and calorie intake, as well as urine output. Patients must be closely monitored for vomiting and drug abuse behavior.

Less severe patients can be treated in a partial-hospitalization or day program, and those who are not in medical danger and are functioning fairly well can be treated in outpatient care. Table 3 and Table 4 contain guidelines for levels of care for patients with eating disorders.

Anorexia nervosa. For AN, a multifaceted treatment approach is the most effective and includes medical management, psychoeducation and individual therapy utilizing both cognitive and behavior therapy principles. Controlled studies have shown that children under age 18 do better if they also have family therapy (Russell et al., 1987). Nutritional counseling and pharmacological intervention can be useful components to the treatment plan.

There are fewer than 10 randomized, controlled treatment studies in AN. This is most likely because these patients are resistant to treatment and are prone to develop serious medical complications that require withdrawal from research treatment protocols. Cognitive-behavioral therapy (CBT) can be used in both inpatient and outpatient settings. Cognitive-behavioral therapy for patients with AN was carefully developed by Garner and Bemis (1982) and later expanded into a manual for a collaborative multicenter study. This type of treatment must be conducted by well-trained, experienced therapists. Monitoring is an essential component of CBT. Patients are taught to monitor their food intake, their feelings and emotions, their bingeing and purging behaviors, and their problems in interpersonal relationships. Cognitive restructuring is a method in which patients are taught to identify autonomic thoughts and challenge core beliefs. Problem solving, which is also part of CBT, is a specific method whereby patients learn how to think through and devise strategies to cope with their food-related and/or interpersonal problems.

Behavior therapy consists mainly of positive reinforcements for weight gain. The outpatient needs to be weighed weekly and the inpatient daily. Response prevention techniques are used to stop bingeing and purging. For example, in the inpatient setting, the patients may be required to sit together in a living room for one hour after eating. An in-depth discussion of CBT for AN can be found in Kleifield et al. (1996). Two studies comparing CBT with nutritional counseling have shown the former to be significantly superior in treating AN (Pike et al., 2003; Serfaty et al., 1999).

British studies have shown that adolescents who receive family counseling for treatment of AN do significantly better than those who do not (Dare and Eisler, 2001; LeGrange et al., 1992; Russell et al., 1987). These studies demonstrated that brief therapy separately involving parents and their daughters can be as effective as conjoint family therapy in which the whole family is treated together. In both forms of family therapy, the responsibility is placed on the family for changing eating behavior and weight gain in the patient.

Medications should be considered as an adjunct treatment for AN. Cyproheptadine (Periactin), a serotonin antagonist and antihistamine drug, has been shown to facilitate weight gain and reduce depressive symptomatology in patients with AN in a double-blind, placebo-controlled trial (Halmi et al., 1986). Cyproheptadine has the advantage of not having the tricyclic antidepressant side effects of reducing blood pressure and increasing heart rate. This makes it especially attractive to use in emaciated anorectic patients. It can be used in high doses such as 24 mg/day to 28 mg/day with safety.

There is some evidence that fluoxetine (Prozac) is useful in preventing weight relapse in AN and may specifically target the obsessive-compulsive behaviors that are seen with food and weight control (Kaye et al., 1991). There is no evidence that fluoxetine is effective in treating low-weight anorectics. Although chlorpromazine (Thorazine) was the first drug used to treat patients with AN, there are no double-blind, controlled studies to definitely prove its effectiveness for reducing core anorectic symptomatology and for inducing weight gain. Clinical experience has shown this medication to be particularly helpful in the severely ill patient who is overwhelmed with constant thoughts of losing weight and has uncontrollable behavioral rituals. It is better to start chlorpromazine at low doses of 10 mg tid and increase the dose gradually, monitoring blood pressure and side effects.

More recently, olanzapine (Zyprexa) has been demonstrated in several cases to be effective in inducing weight gain and reducing anxiety in patients with AN (Boachie et al., 2003; Malina et al., 2003; Powers et al., 2002). Controlled and open studies with this olanzapine are currently being conducted. Again, it is wise to start the medication at a low dose of 2.5 mg/day and gradually increase the dosage, monitoring side effects. Many patients are aware of the weight gain side effect of olanzapine and thus refuse to take this drug.

Unfortunately insurance coverage for psychiatric inpatient treatment for AN is almost nonexistent in many areas of the United States. This limits care to brief medical hospitalization and outpatient care. The impact of these policies on the long-term outcome of AN needs to be studied.

Bulimia nervosa. Cognitive-behavioral therapy is a most effective treatment, proven in 35 controlled studies of BN. About 40% to 50% of patients are abstinent from both bingeing and purging at the end of treatment (16 weeks to 20 weeks). Improvement by reducing bingeing and purging occurred in a range from 70% to 95% of patients. Another 30% who did not show improvement immediately posttreatment showed improvement to full recovery one year after treatment (Walsh et al., 1997; Wilson and Fairburn, 2002). In patients with BN, CBT interrupts the self-maintaining cycle of bingeing and purging and alters the individual's dysfunctional cognitions and beliefs about food, weight, body image and overall self-concept.

Three studies have combined CBT with antidepressant medication: One of those showed no additional benefit, and the other two showed the combination was superior to CBT or medication alone (Wilson and Fairburn, 2002).

Cognitive-behavioral group therapy is especially cost effective for both patients and treatment centers. Group therapy is appealing for patients with BN since they often keep their eating disorder secret and feel isolated, ashamed and embarrassed about their symptoms. There are no controlled studies to assess the efficacy of CBT group therapy in comparison with individual CBT for the treatment of BN.

Over a dozen double-blind, placebo-controlled trials of antidepressants, including amitriptyline, desipramine (Norpramin), fluoxetine, imipramine (Tofranil), nortriptyline (Aventyl, Pamelor) and phenelzine (Nardil), have been conducted in outpatients with BN with normal weight. The dosage of antidepressant medication was similar to that used for the treatment of depression. In all trials, antidepressants were significantly more effective than placebo in reducing binge eating. These medications also improved mood and reduced eating disorder symptoms such as preoccupations with shape and weight. However, the abstinence rate from bingeing and purging was only 22% overall (de Zwaan and Roerig, 2003). Fluoxetine is the only drug approved by the U.S. Food and Drug Administration for the treatment of BN in adults. In general, selective serotonin reuptake inhibitors are preferred for treating BN because of their more benign side-effect profile. There still remains the question of how long BN patients should remain on medication.


There are no treatments that can guarantee a cure for either anorexia nervosa or bulimia nervosa. Continuing research on innovative treatment techniques for both disorders is necessary. It is necessary to recognize that these eating disorders come about via an interaction of environmental influences with a biological or genetic vulnerability. Thus, a better understanding of the biological and genetic factors of these disorders may be helpful in devising more effective treatment strategies.

Dr. Halmi is professor of psychiatry at Cornell University Medical College and director of the eating disorder program at Weill Cornell Medical Center.

Dr. Halmi has indicated she has nothing to disclose regarding the nature of this article.


Boachie A, Goldfield GS, Spettigue W (2003), Olanzapine use as an adjunctive treatment for hospitalized children with anorexia nervosa: case reports. Int J Eat Disord 33(1):98-103.

Braun DL, Sunday SR, Halmi KA (1994), Psychiatric comorbidity in patients with eating disorders. Psychol Med 24(4):859-867.

Bushnell JA, Wells JE, McKenzie JM et al. (1994), Bulimia comorbidity in the general population and in the clinic. Psychol Med 24(3):605-611.

Cavallini MC, Bertelli S, Chiapparino D et al. (2000), Complex segregation analysis of obsessive-compulsive disorder in 141 families of eating disorder probands, with and without obsessive-compulsive disorder. Am J Med Genet 96(3):384-391.

Cotrufo P, Barretta V, Monteleone P, Maj M (1998), Full-syndrome, partial-syndrome and subclinical eating disorders: an epidemiological study of female students in Southern Italy. Acta Psychiatr Scand 98(2):112-115.

Dare C, Eisler I (2001), Family therapy and eating disorders. In: Eating Disorders and Obesity, 2nd ed., Fairburn CG, Brownell KD, eds. New York: Guilford Press, pp314-319.

de Zwaan M, Roerig J (2003), Pharmacological treatment of eating disorders: a review. In: Eating Disorders, Maj M, Halmi K, Lopez-Ibor JJ, Sartorius N, eds. New York: J. Wiley.

Garner DM, Bemis KM (1982), A cognitive-behavioral approach to anorexia nervosa. Cognit Ther Res 6:1223-1250.

Halmi KA (2002), Eating disorders. In: Pediatric Pharmacology: Principles and Practice, Martin A, Scahill L, Charney DS, Leckman JF, eds. New York: Oxford University Press, pp592-602.

Halmi KA (1974), Comparison of demographic and clinical features in patient groups with different ages and weights at onset of anorexia nervosa. J Nerv Ment Dis 158(3):222-225.

Halmi KA, Eckert E, LaDu TJ, Cohen J (1986), Anorexia nervosa. Treatment efficacy of cyproheptadine and amitriptyline. Arch Gen Psychiatry 43(2):177-181.

Halmi KA, Eckert E, Marchi P et al. (1991), Comorbidity of psychiatric diagnoses in anorexia nervosa. Arch Gen Psychiatry 48(8):712-718 [see comment].

Halmi KA, Sunday SR, Strober M et al. (2000), Perfectionism in anorexia nervosa: variation by clinical subtype, obsessionality, and pathological eating behavior. Am J Psychiatry 157(11):1799-1805.

Healy K, Conroy RM, Walsh N (1985), The prevalence of binge-eating and bulimia in 1063 college students. J Psychiatr Res 19(2-3):161-166.

Herzog DB, Keller MB, Lavori PW et al. (1992), The prevalence of personality disorders in 210 women with eating disorders. J Clin Psychiatry 53(5):147-152.

Holderness CE, Brooks-Gunn J, Warren MP (1994), Co-morbidity of eating disorders and substance abuse review of the literature. Int J Eat Disord 16(1):1-34.

Ivarsson T, Rastam M, Wentz E et al. (2000), Depressive disorders in teenage-onset anorexia nervosa: a controlled longitudinal, partly community-based study. Compr Psychiatry 41(5):398-403.

Kaye WH, Weltzin TE, Hsu LK, Bulik CM (1991), An open trial of fluoxetine in patients with anorexia nervosa. J Clin Psychiatry 52(11):464-471.

Kleifield EI, Wagner S, Halmi KA (1996), Cognitive-behavioral treatment of anorexia nervosa. Psychiatr Clin North Am 19(4):715-737.

LeGrange D, Eisler I, Dare C et al. (1992), Evaluation of family therapy in anorexia nervosa: a pilot study. Int J Eat Disord 12:347-357.

Lilenfeld LR, Kaye WH, Greeno CG et al. (1998), A controlled family study of anorexia nervosa and bulimia nervosa: psychiatric disorders in first-degree relatives and effects of proband comorbidity. Arch Gen Psychiatry 55(7):603-610.

Lynch WC, Everingham A, Dubitzky J et al. (2000), Does binge eating play a role in the self-regulation of moods? Integr Physiol Behav Sci 36(4):298-313.

Malina A, Gaskill J, McConaha C et al. (2003), Olanzapine treatment of anorexia nervosa: a retrospective study. Int J Eat Disord 33(2):234-237.

Martin CK, Williamson DA, Thaw JM (2000), Criterion validity of the multiaxial assessment of eating disorders symptoms. Int J Eat Disord 28(3):303-310.

Pike KM, Walsh BT, Vitousek K et al. (2003), Cognitive behavior therapy in the posthospitalization treatment of anorexia nervosa. Am J Psychiatry 160(11):2046-2049.

Powell AL, Thelen MH (1996), Emotions and cognitions associated with bingeing and weight control behavior in bulimia. J Psychosom Res 40(3):317-328.

Powers PS, Santana CA, Bannon YS (2002), Olanzapine in the treatment of anorexia nervosa: an open label trial. Int J Eat Disord 32(2):146-154.

Russell GF, Szmukler GI, Dare C, Eisler I (1987), An evaluation of family therapy in anorexia nervosa and bulimia nervosa. Arch Gen Psychiatry 44(12):1047-1056.

Serfaty MA, Turkington D, Heap M et al. (1999), Cognitive therapy versus dietary counseling in the outpatient of anorexia nervosa: effects of the treatment phase. European Eating Disorders Review 7(5):334-350.

Shisslak CM, Crago M, Estes LS (1995), The spectrum of eating disturbances. Int J Eat Disord 18(3):209-219.

Steiger H, Thibaudeau J, Leung F et al. (1994), Eating and psychiatric symptoms as a function of Axis II comorbidity in bulimic patients. Three-month and six-month response after therapy. Psychomatics 35(1):41-49.

Wade TD, Bulik CM, Neale M, Kendler KS (2000), Anorexia nervosa and major depression: shared genetic and environmental risk factors. Am J Psychiatry 157(3):469-471.

Walsh BT, Wilson GT, Loeb KL et al. (1997), Medication and psychotherapy in the treatment of bulimia nervosa. Am J Psychiatry 154(4):523-531.

Wilson GT, Fairburn CG (2002), Treatments for eating disorders. In: A Guide to Treatments That Work, 2nd ed., Nathan P, Gorman JM, eds. New York: Oxford University Press, pp559-592.

Zuckerman DM, Colby A, Ware NC, Lazerson JS (1986), The prevalence of bulimia among college students. Am J Public Health 76(9):1135-1137.

Psychiatric Times - Category 1 Credit

If you are not on the Lifelong Learning site, click here.

You must keep your own records of this activity. Copy this information and include it in your continuing education file for reporting purposes.

CME LLC is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. CME LLC designates this educational activity for a maximum of 1.5 category 1 credits toward the AMA Physician's Recognition Award. Each physician should claim only those credits actually spent on the educational activity.

CME LLC is an approved provider of continuing medical education for physicians, nurses and physician assistants in the State of Florida and is registered with CE Broker.

CME LLC is approved by the California Board of Registered Nursing, Provider No. CEP12748, and designates this educational activity for 1.5 contact hours for nurses. The American Nurses Credentialing Center (ANCC) accepts AMA category 1 credit toward recertification requirements.

Recent Videos
brain depression
nicotine use
eating disorder brain
© 2024 MJH Life Sciences

All rights reserved.