This is Web4Health's cache of
Web4Health's cache is the snapshot that we took of the page when we added it to Web4Health.
The page may have changed since that time, so this cached copy may be out-of-date.

Web4Health is not affiliated with the authors of this page nor responsible for its content.


Eating Disorders and Addiction

by Barton J. Blinder, M.D., Ph.D.; Mary C. Blinder, M.D.; and Visant A. Sanathara

Psychiatric Times - December 1998 - Vol. XV - Issue 12

Educational Objectives:

After reading this article, you will be familiar with:
  • Common and divergent causal factors of eating disorders and substance use disorders.
  • Comorbidity of eating disorders and substance use disorders.
  • Neurobiology of eating disorders and substance use disorders.
  • Assessment techniques and nutritional, pharmacological and psychotherapeutic interventions.

Who may benefit from reading this article?
Psychiatrists, primary care physicians, neurologists, psychologists, social workers, dietitians, psychiatric nurses and other mental health care professionals.

Clinical experience increasingly has suggested an association between eating disorders (EDs) and addiction to substances. Descriptively, subjective states of craving, repetitive/ritualistic behaviors, resistance to treatment and frequency of relapse have led to diagnostic inferences about similarities and differences in the two disorders.

Understanding common and divergent causal factors (genetic, neurobiological, developmental) of EDs and substance use disorders (SUDs) will contribute significantly to the specificity of diagnosis and efficacy of available treatment options, including pharmacotherapy, psychotherapy and rehabilitation (Table 1).

Dopamine receptors in the shell of the nucleus accumbens, associated with catecholamine and opioid neurotransmitters, initiate and modulate the reward reaction to abused substances (including marijuana), as well as drive the addictive behaviors and the craving that emerge from their use (Bloom, 1998). Rewarding and aversive withdrawal reactions to drugs of abuse are mediated separately by dopamine in the nucleus accumbens (reward) and corticotropin-releasing factor activation in the amygdala (substance withdrawal and stress-induced withdrawal).

The reward for food is independent of the drug abuse system (Bloom, 1998; Maldonado et al., 1997). Endogenous opioid peptides drive feeding, while serotonin and dopamine are primarily inhibitors of feeding in the paraventricular nucleus and lateral area of the hypothalamus.

Convergence and overlap of neurobiological determinants may be responsible for both similarities in clinical symptoms and responses to treatment in EDS and substance abuse.

Studies of Comorbidity

Disorders of substance use are associated more frequently with bulimia nervosa (BN) and "bulimic behaviors" (i.e., bingeing and >purging) than with anorexia nervosa (AN) (Holderness et al., 1994). AN patients, who exclusively restrict food intake, are less prone to substance use and abuse than those with bulimic behaviors. Furthermore, the incidence of BN is greater than AN among substance abusers.

Impulsive eating and >purging behaviors often precede substance use (Holderness et al., 1994). In a 10-year prospective follow-up study of 95 AN patients (77 restrictors and 18 binge-eating subtypes), Strober et al. (1996) reported that, where there had been no pre-existing diagnosis of SUDs, by the end of the study a new diagnosis of comorbid substance use was made in 50% of the binging subtype patients compared to only 12% for the restrictors. In addition, SUDs were more prevalent among the binge-eating subtype patients who remained underweight. Binge-eaters also were more likely to have at least one first-degree relative with SUD (55% for binge-eaters versus 14% for restrictors).

The complex interaction between EDs and SUDs may yield distinct clinical behavioral characteristics that may be absent within a single diagnosis. Impulsivity has been demonstrated as a key attribute that links bulimia with substance abuse (Holderness et al., 1994). ED patients with at least one personality disorder are significantly more likely to manifest substance dependence than those without a personality disorder (Braun et al., 1994). Bulik et al. (1997) reported that a subgroup of women with BN and comorbid alcohol dependence had greater Axis II psychopathology (most commonly borderline personality disorder). The women also had occurrence of increased frequency of symptomatic behavior, novelty seeking and impulsivity.

One may question if the majority of these symptoms are related to alcohol abuse rather than BN. Suzuki et al. (1993) studied clinical characteristics of established female alcoholics with and without EDs (primarily BN). Alcoholic women with EDs were younger, had lower body weight, and had diagnoses of borderline personality disorder and depression.

At the 1998 American Psychiatric Association's annual meeting, Lilenfeld et al. (1998) reported that women suffering from BN and substance dependence had more cluster B personality diagnoses (dramatic, emotional and erratic personality disorders) as compared to the BN and control groups. Furthermore, she suggested that impulsivity, mood instability and certain forms of anxiety (particularly social phobia and panic disorder) may represent familial vulnerability factors for substance use by these women. Dansky et al. (1998) suggested that comorbid diagnosis of major depression (MD) and posttraumatic stress disorder (PTSD) need to be considered as possible comorbid diagnoses when dealing with comorbid ED and SUD.

Comorbidity raises important concerns regarding the manner in which these disorders are transmitted. Is the link between EDs and SUDs a result of the same gene expressed differentially, or are the disorders transmitted independently? Kaye et al. (1996) studied alcohol dependence among first-degree relatives of three distinct proband groups with BN and lifetime alcohol and/or drug dependence, BN alone and a non-eating-disordered community control group.

If both disorders were transmitted by the same genetic component, then the rates of alcohol/drug dependence among the first- degree relatives should be significantly higher in both BN groups when compared to the control group. However, among the BN with alcohol dependence probands, 38% of the first-degree relatives qualified for lifetime alcohol/drug dependence as opposed to only 10% for the relatives of the BN probands and 18% for the relatives of the community controls. A similar pattern was confirmed when the data was segregated between male and female relatives.

This was less evident with expanded criteria, including both alcohol and other drug use or dependence. Bulik (1991) concluded that women suffering from BN who also had a comorbid diagnosis of alcohol abuse or dependence were more likely to have at least one family member with a SUD.

Schuckit et al. (1996), reversing the primary proband group studied, searched for lifetime rates of BN or AN among 4,265 alcohol-dependent men and women and their relatives. The study did not find any significant evidence of a strong familial crossover between AN or BN and alcohol dependence. These studies demonstrate that only probands with established ED or SUD and the comorbid presence (ED or SUD) had higher rates of this comorbid disorder among their relatives. Therefore, although ED and SUD may aggregate in individuals and families, it is probable that they are independently transmitted.

Research Design

Defining research design is critical to understanding the complex interactions between EDs and SUDs. Structured interviews using the criteria in the Diagnostic and Statistical Manual of Mental Disorders, 4th ed. (American Psychiatric Association, 1994), as opposed to local questionnaires and scales, are strongly recommended in order to make valid comparisons between studies as well as generalized inferences across groups.

In addition, different categories within eating disorders (e.g., AN, BN and eating disorders not otherwise specified) and SUDs (e.g., type of drug) should be identified in order to localize and differentiate associations within each subtype. Data analysis becomes more troublesome when these subtypes are clumped into their broader diagnoses (Holderness et al., 1994). The need for a control population of similar demographic characteristics to the index proband is always mandatory.

Future studies should aim to establish the sequence of onset of EDs and SUDs in individual patients. A suggested way of studying this onset-consequence would be to look for symptomatic behaviors of SUD (i.e., increased drug use) and ED (i.e., binge-eating and >purging) to determine when these symptoms emerge as a distinct disorder (Jonas et al., 1987). Studies that investigate ED among SUD patients are needed to balance more extensive data derived from studies of SUD in primary diagnosed ED patients (Table 2 and Table 3). With large population samples and structured standardized criteria, the overall rates of comorbidity may be lower but, nevertheless, clinically relevant.

Within the general population, EDs and SUDs have a prevalence rate of 2.5% and 3%, respectively (Hsu, 1996; Myers et al., 1984) (Table 4). However, the SUD rate among those with ED rises to 25.7%, over eight times higher than in the general population. In a like manner, the ED rate rises to 16.3% among those with SUD, over five times greater than the mainstream population. It is evident that the occurrence of ED or SUD as the primary diagnosis is associated with the increased liability for the lifetime occurrence of each other as a comorbid diagnosis.


The exact nature of the neurobiology of the EDs and SUDs is not known; however, a significant amount of research has been conducted in an effort to determine the relationship between EDs, SUDs and a variety of neurotransmitters. Table 5 compares neurotransmitter determinants in the eating and addiction disorders.

As in other psychiatric disorders, multiple neurotransmitters are thought to be involved in the EDs. Research has demonstrated that opioids increase feeding and opioid antagonists decrease feeding. Confronted with the compulsion to binge in bulimia and to starve in anorexia, researchers have entertained the possibility of a link between endogenous opioids and the EDs.

A number of studies have demonstrated opioid abnormalities in patients with EDs. For example, in patients with BN, low plasma and cerebrospinal fluid (CSF) beta-endorphin levels have been identified (Brewerton et al., 1992). In addition, anorexic patients have been noted to have decreased beta- endorphin levels in the CSF (Kaye et al., 1987).

Similarly, many of the same neurotransmitters involved in EDs appear to be involved in the SUDs. Hubbell et al. (1986, 1987) and Mercer and Holder (1997) showed that morphine, an opioid agonist, potentiates alcohol consumption. This finding suggested that central opioid transmission might be involved in alcohol abuse.

Another neurotransmitter thought to be involved in the neurobiology of EDs is dopamine. In an effort to determine the relationship between dopamine and feeding behaviors, Terry and Katz (1992) showed that both food-deprived rats and routinely fed rats decreased feeding when given D1 agonists. Dopamine inhibits the initiation and extent of feeding (Morley et al., 1988).

In a different study, Taber et al. (1998) suggest a connection between opioid receptors and the release of dopamine associated with feeding. When rats were given naltrexone (ReVia), an opioid antagonist, feeding was decreased, and the release of dopamine associated with feeding was blocked.

Dopamine is strongly associated with amphetamine and cocaine abuse. Extracellular dopamine has been shown to increase in the nucleus accumbens in rats when they press a lever for food, and when they are given amphetamines or cocaine (Hernandez and Hoebel, 1988). The increase in extracellular dopamine has been thought to be caused by cocaine blocking reuptake of dopamine via the dopamine transporter (London et al., 1996). Most recently, a mechanism independent of the dopamine transporter in cocaine abuse has been identified (Sora et al., 1998).

Dopamine is involved in addiction disorders in relation to its interaction with neuropeptide Y. Studies have shown that neuropeptide Y stimulates feeding at the perifornical hypothalamus (Clark et al., 1984). When rats are given amphetamines, neuropeptide Y's stimulatory effect on feeding is decreased, and is eliminated at higher dosages. When haloperidol (Haldol), a dopamine blocker, is administered, the effect of amphetamine on neuropeptide Y is blocked (Gillard et al., 1993).

Amphetamine inhibits feeding by increasing extracellular dopamine and decreasing the feeding stimulatory effect of neuropeptide Y. It is common for patients with EDs to abuse stimulants such as over-the-counter and prescription drugs, in addition to illegal drugs such as amphetamines and cocaine. It is often thought that patients use these drugs for the direct effect of stimulants on weight loss. These studies help to explain why amphetamine addicts and cocaine addicts usually lose weight.

Neuropeptide Y itself has been studied in alcohol and opioid addiction with implications for its role in basic mechanisms and its connection to feeding. Woldbye et al. (1998) demonstrated an opioid antiwithdrawal effect for neuropeptide Y, suggesting a potential role for neuropeptide Y and Y5 receptors in both basic mechanisms and therapeutic interventions in opioid dependence and withdrawal.

Neuropeptide Y also was found to have an inhibitory effect on the development of tolerance to alcohol, possibly related to its capacity to increase feeding behavior as a competitive motivation (Timofeeva et al., 1992). A controlled baseline study of CSF in male alcoholics found no significant elevation in CSF levels of neuropeptide Y (Roy et al., 1990). In an animal study, alcohol pretreatment significantly potentiated neuropeptide-Y-stimulated food intake, but did not by itself cause spontaneous food ingestion (Blackburn et al., 1994).

Finally, serotonin has been implicated in EDs. CSF levels of metabolites of serotonin and dopamine (5-hydroxyindoleacetic acid [5-HIAA] and homovanillic acid) have been found to be low in BN patients who binge frequently (Jimerson et al., 1992). In addition, AN patients have been found to have low CSF 5-HIAA levels that return to normal after the patients gain weight (Kaye et al., 1988).

The selective serotonin reuptake inhibitor (SSRI) sertraline (Zoloft) decreases alcohol consumption in monkeys that are conditioned to drink to intoxication; however, the sertraline effect is diminished in animals who are also subject to stress (Higley et al., 1998). Central nervous system and platelet serotonin have been found to be altered by alcohol. Schmidt et al. (1997) found that antisocial alcoholics have increased platelet serotonin in contrast to other subgroups of alcoholics. Alcohol ingestion and nicotine use were found to normalize platelet serotonin. Results of current research on platelet serotonin in alcoholics yields variable results with multiple potential determinants.

Potential treatment interventions are suggested from the foregoing studies. Research that demonstrates opioid-related increase of feeding behavior suggests that opioid antagonists may be a possible treatment option in EDs.

Naltrexone, an opioid antagonist, currently is used to treat alcoholism and narcotic addiction. Naltrexone significantly decreases bingeing and >purging in patients with BN and in patients with a bingeing subtype of AN (Jonas and Gold, 1988). Marrazzi et al. (1995a, b) reports an obese patient with bingeing disorder responding to naltrexone, experiencing a decrease in bingeing. Drewnowski et al. (1995), in a careful study of the effect of naloxone (Narcan, Talwin) infusion on the hedonic response to and consumption of sweet and high fat foods, demonstrated that food intake of non-bingeing obese women was not affected. In contrast, the naloxone infusion significantly reduced the consumption of sweet and high fat foods in BN women.

Other studies have shown low doses of naltrexone (50 to 100 mg/day) are ineffective in treating bulimia in contrast with high dosages (200 to 300 mg/day) that may decrease bingeing and >purging significantly (Mitchell et al., 1989).

Researchers have studied various SSRIs in treating eating disorders and addictions. In BN, higher doses of fluoxetine (Prozac) (60 mg/day) have been found to be effective (Fluoxetine Bulimia Nervosa Collaborative Study Group, 1992). Angelone et al. (1998) found that, after detoxification, alcoholic patients who received the SSRIs fluvoxamine (Luvox) or citalopram (Celexa) in combination with cognitive-behavioral therapy (CBT) had a more significant rate of abstinence than those treated with CBT alone.

Treatment Approaches

Impressive advances in the recognition and treatment of EDs have occurred in the past decade. Nutritional, pharmacological and psychotherapeutic interventions have addressed comorbidity (depression, obsessive-compulsive disorder, SUD, personality disorder); treatment resistance and chronicity; medical complications; and complexities of psychological development (American Psychiatric Association, 1996).

In BN, combined CBT and pharmacotherapy have proven to have an impressive positive outcome, although long-term maintenance of improvement and methodological factors in designing clinical research studies remain problematic (Abbott and Mitchell, 1993; Mitchell, 1996).

Fairburn et al. (1993, 1995) demonstrated the superiority of CBT over simplified behavioral treatment addressed only to the patient's eating habits, and attitudes about shape and weight. Standard CBT is delivered in 16 to 20 sessions initially with twice weekly meetings. Agras (1991) noted that the cognitive-behavioral model asserts that societal pressures for a thin body shape lead to dietary restraint and, eventually, to binge-eating with compensatory >purging.

A major aim of treatment is to restore normal dietary intake, resulting in decreased bingeing and diminished >purging. Cognitive elements involve correcting irrational food rules, challenging distorted appraisals of body shape and weight, and clarifying precipitating adverse environmental factors. Self-monitoring, self-regulation and self-perceptive awareness is emphasized.

More recently, interpersonal therapy (IPT) has shown efficacy as a nonpharmacological intervention for BN and for bingeing disorders without compensatory >purging. This model emphasizes resolving conflictual interpersonal issues, allowing control over food intake to be reestablished (Wilfley and Cohen, 1997).

Evidence from controlled clinical trials suggests that IPT may be a reasonable alternative or certainly an adjunct to medication as an acute, continuation and/or maintenance treatment for BN. An additional advantage with IPT is its efficacy in patients with dysthymia and in adolescents with depression diagnostic subgroups frequently associated with BN (Weissman, 1997).

Treatment of AN is often of longer duration, requiring a combination of cognitive and analytic interpretative techniques to explore the past, identifying maladaptive internal and relational schemata. Family therapy is especially helpful for adolescent patients (Murray et al., 1997).

Nutritional education and rehabilitation is essential in the treatment of EDs. Step-by-step, carefully conceived, cooperative approaches are necessary to modify appetite regulation, food selection and consumatory behavior (meal pattern, rate of eating, foraging and hoarding behavior, and compensatory >purging) (Reiff and Reiff, 1992).

Overeaters Anonymous and Alcoholics Anonymous emphasize a 12-step addiction model applied with some modifications to EDs. Carefully controlled studies to evaluate the efficacy of this approach are not available. Additionally, there may be hazards due to misinformation and opinions that label food as an "addictive substance."

Since 1979, a number of placebo- controlled studies have documented superior short-term outcome in patients with BN receiving antidepressant medication (Mitchell et al., 1993; Walsh et al., 1997). Binge reduction has been demonstrated at 47% for the tricyclic antidepressant desipramine (Norpramin). The SSRIs (fluoxetine, fluvoxamine) decreased binge rates and >purging between 45% and 67% (Crow and Mitchell, 1994). Fluoxetine also significantly improved depression, carbohydrate craving, and pathological eating attitudes and behavior at higher dosage levels (60 mg/day) (Fluoxetine Bulimia Nervosa Collaborative Study Group, 1992).

Tricyclic antidepressants have been administered in their usual antidepressant dosages to achieve an antibulimic effect (adverse cardiac effects may occur especially in the presence of hypokalemia). However, significantly higher doses of fluoxetine (60 mg/day) are required for the same antibulimic effect (Fluoxetine Bulimia Nervosa Collaborative Study Group, 1992). Other antidepressants studied include monamine oxidase inhibitors (which require dietary restrictions), bupropion (Wellbutrin) (not recommended due to reports of increased risk of seizure in BN patients), nomifensine (withdrawn due to immuno-hemolytic anemia) and mianserin (Crow and Mitchell, 1994; Mitchell et al., 1993).

Placebo-controlled studies with tricyclics have shown no significant effect in treating AN. Fluoxetine has been reported to have a possible benefit; primarily maintaining improvement in weight-recovered patients. But this effect needs further clarification and study from the perspective of the intensity and longitudinal course of comprehensive follow-up treatment. Strober et al. (1997) concluded that adjunctive treatment with fluoxetine "may not have additive long-term therapeutic benefit when measured against the effects of sustained and intensive follow-up treatment."

Additionally, high dosages of serotonin and histamine antagonist cyproheptadine (Periactin) given at 32 mg/day and the use of lithium (Eskalith, Lithobid) have shown benefit in some studies as an adjunctive therapy in restoring body weight (Crow and Mitchell, 1994).

A small number of comparison studies with a singular approach and limited severity and outcome measures have found CBT to be superior to pharmacological treatments in BN (Abbott and Mitchell, 1993). However, recent studies have shown that combining medication and CBT is superior to offering either treatment individually. A two-stage psychiatric medical intervention with SSRIs that synergistically maximizes the benefits of psychological therapy has been found to be optimal (Walsh et al., 1997).

Special Treatment Considerations

Comorbidity can pose a major treatment problem for settings primarily designed to handle only one of these disorders. Confronting comorbidity of EDs and SUDs is increasingly common (Marsh et al., 1997). Treatment difficulties may arise due to the need for adjusting to different treatment approaches, lack of familiarity or expertise in attending staff, and negative countertransference in reaction to manipulative behaviors (Katz, 1990).

Significant rates of comorbidity suggest routine screening for EDs in individuals with SUDs, and vice versa (Katz, 1990). Clinicians who treat patients with SUDs, for example, should assess these patients carefully for the comorbid presence of an ED and not attribute abnormal eating behavior to symptomatic expression of drug use alone (Jonas et al., 1987).

Physicians also should be aware that "disordered eating" patterns do not constitute the presence of an ED. Additionally, clinicians who attend to ED patients should be aware of the potential uses (abuses) of substances to compensate for ED consequences such as weight gain and bloating. For instance, the use of cocaine and other stimulants as a substance for weight control among patients with EDs has been documented (Cochrane et al., 1998). A thorough evaluation of these coexisting clinical features is necessary for effective treatment.

If comorbidity is confirmed, a clearly defined protocol should be activated to concurrently treat both disorders, taking into consideration the strength and limitations of the inpatient/partial/outpatient program and the clinical status and motivation of the patient. Neither condition should be used as a basis of treatment for the other disorder.

Thorough medical evaluation is crucial in assessing cardiac, hematological, electrolyte and thyroid status. Additional medical concerns involve a "rogues' gallery" of abnormalities complicating the patient's clinical status. Iron deficiency and other subtle nutritional deficits further compromise cognitive function and impair mood, resulting in diminished effectiveness of both psychotherapy and pharmacotherapy.

Certain clinical findings are of interest in ED/SUD comorbidity:

  • Opioids contribute to disturbances in bowel motility, prolonging gastric emptying and worsening constipation.
  • Alcohol use may facilitate regurgitation and dehydration.
  • Cocaine withdrawal worsens bingeing and may promote weight gain leading to reactive anxiety.
  • Laxative abuse may result in bloating, contributing to heightened concerns regarding body image.

Nutritional consultation should be obtained to provide a target range of weight, counseling and a psychoeducational approach to meal planning in ED/SUD programs. Patients should be carefully monitored to prevent substance access, hoarding of food and >purging (Katz, 1990).

Psychological treatment for substance abuse should include family work, relapse prevention through CBT, group and motivation techniques, stress management, and self-control and self-help group therapies. Specialized interventions for detoxification (including tapering of laxative overuse) and the use of transitional supportive medication should take into account the patient's nutritional status and specific electrolyte and cardiac vulnerabilities.

Among individual psychotherapy options, longer-term, psychodynamic psychotherapy should be considered. A broader array of anxiety, mood and personality disturbance is frequently encountered.

Psychoanalytic theory and developmental observations emphasize the significance of early attachment deficits that may determine later states of craving (substances, food) and self-regulation disturbances (mood, anxiety). Where developmental history indicates complexity and adverse events (abandonment, neglect and abuse), more intensive and longer-term psychotherapy is an essential component of treatment in both EDs and SUDs.

Support, clarification, insight and the working through of repeated relational conflicts help to create a basis for adaptation and mechanisms for coping with life stressors, which are often absent in patients after years of defeat and low morale. Common psychological defense structures such as denial, compartmentalization, impulsive action and avoidance in ED and SUD require a strong treatment alliance to bring about change (Blinder, 1980; Smolak and Levine, 1994; Wilson et al., 1992; Wurmser 1987, 1985, 1982).


We gratefully acknowledge the assistance of the following: Suvarna Bhat, M.D.; David Wall, Ph.D.; Edward Cummella, Ph.D.; Michael Strober, Ph.D.; Floyd Bloom, M.D.; Edith London, Ph.D.; Holly Applebaum; Michael Argosino; Denise A. Manandik; and Darren Brown. We hope this contribution will be a small step toward alleviating the casualty and suffering borne by patients and families.

Dr. Barton Blinder is a clinical professor and director of eating disorders research in the department of psychiatry and human behavior at the University of California Irvine.

Dr. Mary Blinder is an attending psychiatrist at the Royale Treatment and Residential Center in Santa Ana, Calif., and maintains a private practice.

Mr. Sanathara is a premedical student at UCI.


Abbott DW, Mitchell JE (1993), Antidepressants versus psychotherapy in the treatment of bulimia nervosa. Psychopharmacol Bull 29(1):115-119.

Agras S (1991), Nonpharmacologic treatments of bulimia nervosa. J Clin Psychiatry 52(suppl):29-33.

American Psychiatric Association (1996), Practice guideline for eating disorders. In: Practice Guidelines. Washington, D.C.: American Psychiatric Association, pp 36-77.

American Psychiatric Association (1994), Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, D.C.: American Psychiatric Association.

Angelone SM, Bellini L, Di Bella D, Catalano M (1998), Effects of fluvoxamine and citalopram in maintaining abstinence in a sample of Italian detoxified alcoholics. Alcohol 33(2):151-156.

Blackburn RE, Stricker EM, Verbalis JG (1994), Acute effects of ethanol on ingestive behavior in rats. Alcohol Clin Exp Res 18(4):924-930.

Blinder BJ (1980), Developmental antecedents of eating disorders: a reconsideration. Psychiatr Clin North Am 3:579-592.

Bloom F (1998), Substance abuse and reward systems in the brain (Lecture 4). Presented at the 151st Annual Meeting of the American Psychiatric Association. June 1; Toronto.

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

Brewerton TD, Lydiard RB, Laraia MT et al. (1992), CSF beta-endorphin and dynorphin in bulimia nervosa. Am J Psychiatry 149(8):1086-1090.

Bulik CM (1991), Family histories of bulimic women with or without comorbid alcohol abuse or dependence. Am J Psychiatry 148(9):1267-1268. Letter.

Bulik CM, Sullivan PF, Carter FA, Joyce PR (1997), Lifetime comorbidity of alcohol dependence in women with bulimia nervosa. Addict Behav 22(4):437-446.

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.

Clark JT, Kalra PS, Crowley WR, Kalra SP (1984), Neuropeptide Y and human pancreatic polypeptide stimulate feeding behavior in rats. Endocrinology 115(1):427-429.

Cochrane C, Malcolm R, Brewerton T (1998), The role of weight control as a motivation for cocaine abuse. Addict Behav 23(2):201-207.

Crow SJ, Mitchell JF (1994), Rational therapy of eating disorders. Drugs 48(3):372-379.

Dansky BS (1998), Victimization and substance abuse: risk factors for bulimia nervosa (Symposium 16D). Presented at the 151st Annual Meeting of the American Psychiatric Association. June 1; Toronto.

Drewnowski A, Krahn DD, Demitrack MA (1995), Naloxone, an opiate blocker, reduces the consumption of sweet and high-fat foods in obese and lean female binge eaters. Am J Clin Nutr 61(6):1206-1212.

Fairburn CG, Jones R, Peveler RC et al. (1993), Psychotherapy and bulimia nervosa. Longer-term effects of interpersonal psychotherapy, behavior therapy and cognitive-behavior therapy. Arch Gen Psychiatry 50(6):419-428.

Fairburn CG, Norman PA, Welch SL et al. (1995), A prospective study of outcome in bulimia nervosa and the long-term effects of three psychological treatments. Arch Gen Psychiatry 52(4):304-312.

Fluoxetine Bulimia Nervosa Collaborative Study Group (1992), Fluoxetine in treatment of bulimia nervosa. A multicenter, placebo-controlled double-blind trial. Arch Gen Psychiatry 49(2):139-147.

Gillard ER, Dang DQ, Stanley BG (1993), Evidence that neuropeptide Y and dopamine in the perifornical hypothalamus interact antagonistically in the control of food intake. Brain Res 628(1-2):128-136.

Grilo CM, Levy KN, Becker DF et al. (1995), Eating disorders in female inpatients with versus without substance abuse disorders. Addict Behav 20(2):255-260.

Hernandez L, Hoebel BG (1988), Food reward and cocaine increase extracellular dopamine in the nucleus accumbens as measured by microdialysis. Life Science 42(18):1705-1712.

Higley J, Hasert M, Suomi S, Linnoila M (1998), The serotonin reuptake inhibitor sertraline reduces excessive alcohol consumption in nonhuman primates: effect of stress. Neuropsychopharmacology 18(6):431-443.

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

Hsu LK (1996), Epidemiology of the eating disorders. Psychiatr Clin North Am 19(4):681-700.

Hubbell CL, Czirr SA, Hunter GA et al. (1986), Consumption of ethanol solution is potentiated by morphine and attenuated by naloxone persistently across repeated daily administrations. Alcohol 3(1):39-54.

Hubbell CL, Czirr SA, Reid LD (1987), Persistence and specificity of small doses of morphine on intake of alcoholic beverages. Alcohol 4(3):149-156.

Hudson JI, Pope HG Jr., Jonas JM, Yurgelun-Todd D (1983), Phenomenologic relationship of eating disorders to major affective disorder. Psychiatry Res 9(4):345-354.

Jimerson DC, Lesem MD, Kaye WH, Brewerton TD (1992), Low serotonin and dopamine metabolite concentrations in cerebrospinal fluid from bulimic patients with frequent binge episodes. Arch Gen Psychiatry 49(2):132-138.

Jonas JM, Gold MS (1988), The use of opiate antagonists in treating bulimia: a study of low-dose versus high-dose naltrexone. Psychiatry Res 24(2):195-199.

Jonas JM, Gold MS, Sweeney D, Pottash AL (1987), Eating disorders and cocaine abuse: a survey of 259 cocaine abusers. J Clin Psychiatry 48(2):47-50.

Katz JL (1990), Eating disorders: a primer for the substance abuse specialist. Theories of etiology, treatment approaches and considerations during comorbidity with substance abuse. J Subst Abuse Treat 7(4):211-217.

Kaye WH, Berretini WH, Gwirtsman HE et al. (1987), Reduced cerebrospinal fluid levels of immunoreactive pro-opiomelanocortin related peptides (including beta-endorphin) in anorexia nervosa. Life Sci 41(18):2147-2155.

Kaye WH, Ebert MH, Lake CR (1988), Disturbances in brain neurotransmitter systems in anorexia nervosa: a review of CSF studies. In: The Eating Disorders: Medical and Psychological Bases of Diagnosis and Treatment, Blinder BJ, Chaitin BF, Goldstein RS, eds. New York: PMA, pp 215-225.

Kaye WH, Gwirtsman HE, George DT et al. (1988), CSF 5-HIAA concentrations in anorexia nervosa: reduced values in underweight subject normalize after weight gain. Biol Psychiatry 23(1):102-105.

Kaye WH, Lilenfeld LR, Plotnicov K et al. (1996), Bulimia nervosa and substance dependence: association and family transmission. Alcohol Clin Exp Res 20(5):878-881.

Keck PE Jr., Pope HG Jr., Hudson JI et al. (1990), A controlled study of phenomenology and family history in outpatients with bulimia nervosa. Compr Psychiatry 31(4):275-283.

Lilenfeld L, Kaye WH, Greeno CG et al. (1998), Familial association of bulimia nervosa and substance dependence (Symposium 16A). Presented at the 151st Annual Meeting of the American Psychiatric Association. June 1; Toronto.

London ED, Cascella NG, Wong DF et al. (1990), Cocaine-induced reduction of glucose utilization in human brain. A study using positron emission tomography and [fluorine 18]-fluorodeoxyglucose. Arch Gen Psychiatry 47(6):567-574.

London ED, Grant SJ, Morgan MJ, Zukin SR (1996), Neurobiology of drug abuse. In: Neuropsychiatry, Fogel BS, Schiffer RB, Rao SM, eds. Baltimore: Williams and Wilkins, pp 635-678.

Maldonado R, Saiardi A, Valverde O et al. (1997), Absence of opioid rewarding effects in mice lacking dopamine D2 receptors. Nature 388(6642):586-589.

Marrazzi MA, Bacon JP, Kinzie J, Luby ED (1995b), Naltrexone use in the treatment of anorexia nervosa and bulimia nervosa. Int Clin Psychopharmacol 10(3):163-172.

Marrazzi MA, Markham KM, Kinzie J, Luby ED (1995a), Binge eating disorder: response to naltrexone. Int J Obes Relat Metab Disord 19(2):143-145.

Marsh LD, Key JD, Spratt E (1997), Bulimia and dextromethorphan abuse A case study. J Subst Abuse Treat 14(4):373-376.

Mercer MF, Holder MD (1997), Food cravings, endogenous opioid peptides and food intake: a review. Appetite 29(3):325-352.

Mitchell JE, Christenson G, Jennings J et al. (1989), A placebo-controlled, double-blind crossover study of naltrexone hydrochloride in outpatients with normal weight bulimia. J Clin Psychopharmacol 9(2):94-97.

Mitchell JE, Hoberman HN, Peterson CB et al. (1996), Research on the psychotherapy of bulimia nervosa: half empty or half full. Int J Eat Disord 20(3):219-229.

Mitchell JE, Raymond N,Specker S(1993), A review of the controlled trials of pharmacotherapy and treatment of bulimia nervosa. Int. J Eat Disord 14(3) 229-247.

Morley JE, Levine AS, Krahn DD (1988), Neurotransmitter regulation of appetite and eating. In: The Eating Disorders: Medical and Psychological Bases of Diagnosis and Treatment, Blinder BJ, Chaitin BF, Goldstein RS, eds. New York: PMA, pp 11-19.

Murray R, Hill P, McGuffin P (1997), The Essentials of Postgraduate Psychiatry, 3rd ed. New York: Cambridge University Press.

Myers JK, Weissman MM, Tischler GL et al. (1984), Six-month prevalence of psychiatric disorders in three communities 1980 to 1982. Arch Gen Psychiatry 41(10):959-967.

Reiff DW, Reiff KKL (1992), Eating Disorders: Nutrition Therapy in the Recovery Process. Gaithersburg, Md.: Aspen Publishers.

Roy A, Berrettini W, DeJong J et al. (1990), CSF neuropeptide Y in alcoholics and normal controls. Psychiatry Res 33(3):215-219.

Schmidt LG, Dufeu P, Heinz A et al. (1997), Serotonergic dysfunction in addiction: effects of alcohol, cigarette smoking and heroin on platelet 5-HT content. Psychiatry Res 72(3):177-185.

Schuckit MA, Tipp JE, Anthenelli RM et al. (1996), Anorexia nervosa and bulimia nervosa in alcohol-dependent men and women and their relatives. Am J Psychiatry 153(1):74-82.

Smolak L, Levine MP (1994), Critical issues in the developmental psychopathology of eating disorders. In: Understanding Eating Disorders: Anorexia Nervosa, Bulimia Nervosa and Obesity, Alexander-Mott L, Lumsden DB, eds. Washington, D.C.: Taylor and Francis, pp 37-60.

Sora I, Wichems C, Takahashi N et al. (1998), Cocaine reward models: conditioned place preference can be established in dopamine- and in serotonin-transporter knockout mice. Proc Natl Acad Sci U S A 95(13):7699-7704.

Strober M, Freeman R, Bower S, Rigall J (1996), Binge eating in anorexia nervosa predicts later onset of substance use disorder: a 10-year prospective longitudinal follow-up of 95 adolescents. Journal of Youth and Adolescents 25(4):519-532.

Strober M, Freeman R, DeAntonio M et al. (1997), Does adjunctive fluoxetine influence the post-hospital course of restrictor-type anorexia nervosa? A 24-month prospective, longitudinal follow-up and comparison with historical controls. Psychopharmacol Bull 33(3):425-431.

Suzuki K, Higuchi S, Yamada K et al. (1993), Young female alcoholics with and without eating disorders: a comparative study in Japan. Am J Psychiatry 150(7):1053-1058.

Taber MT, Zernig G, Fibiger HC (1998), Opioid receptor modulation of feeding-evoked dopamine release in the rat nucleus accumbens. Brain Res 785(1):24-30.

Terry P, Katz JL (1992), Differential antagonism of the effects of dopamine D1-receptor agonists on feeding behavior in the rat. Psychopharmacology (Berl) 109(4):403-409.

Timofeeva LV, Sabo G, Kelesheva LF, Telegdi G (1992), [Effects of neuropeptide Y on rat body temperature in normal conditions and after ethanol administration]. Biull Eksp Biol Med 113(4):344-347.

Toner BB, Garfinkel PE, Garner DM (1986), Long-term follow-up of anorexia nervosa. Psychosom Med 48(7):520-529.

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

Weissman MM (1997), Interpersonal psychotherapy: current status. Keio J Med 46(3):105-110.

Wilfley DE, Cohen LR (1997), Psychological treatment of bulimia nervosa and binge-eating disorder. Psychopharmacol Bull 33(3):437-454.

Wilson CP, Hogan CC, Mintz IR (1992), Psychodynamic Technique in the Treatment of the Eating Disorders. New York: Jason Aronson.

Woldbye DP, Klemp K, Madsen TM (1998), Neuropeptide Y attenuates naloxone-precipitated morphine withdrawal via Y5-like receptors. J Pharmacol Exp Ther 284(2):633-636.

Wurmser L (1987), Flight from conscience: experiences with psychoanalytic treatment of compulsive drug abusers. Part one: dynamic sequences underlying compulsive drug use. J Subst Abuse Treat 4(3-4):157-168.

Wurmser L (1985), Denial and split identity: timely issues in the psychoanalytic psychotherapy of compulsive drug use at impulsive enactment to defenses. J Subst Abuse Treat 2(2):89-96.

Wurmser L (1982), The question of specific psychopathology in compulsive drug use. Ann N Y Acad Sci 398:33-43.