Psychology Department

Health Psychology Home Page

Papers written by students providing scientific reviews of topics related to health and well being

Search HomeWeight LossAlternative Therapy | Supplements | Eating Disorders | Fitness | Links | Self-Assessment | About this Page |


Eating Disorders and Personality Disorders

A Study In Comorbidity

James Caldwell


The correlation between eating disorders and other psychological disorders is very important for our understanding of the causes and possible treatments for eating disorders. It is known that many people with eating disorders also fit the criteria for several DSM-IV psychological disorders. If researchers can find patterns of comorbidity between these two types of disorders they may be able to better diagnose and treat patients with both of these disorders. The question that I pose it what is the relationship between eating disorders and personality disorders(axis 11 disorders in DSM-IV)? It is important to look for comorbidity between the two disorders to determine the impact they have on each other. Once we understand their relationship we may be able to prevent one disorder by treating the other or maybe use the same type of therapy to treat both. In order to answer the question posed I have reviewed several major research articles on the prevalence and comorbidity of personality disorders and eating disorders.


Personality disorders fall under axis 11 of the DSM-IV. This section includes borderline, schizoid, paranoid, antisocial, narcissistic, obsessive-compulsive and avoidant disorders. There are several disorders included in axis 11 but for our purposes we will mostly be dealing with the disorders listed above. The most common personality disorder found among patients with eating disorders is borderline disorder. The majority of the research deals with borderline disorder, a disorder that is characterized by "vulnerability to a range of impulsive behaviors (overeating, shoplifting, substance abuse) and a history of self-destructive behavior, including suicidal gestures and self-mutilation. They show affective instability marked by rapid fluctuations of moods ranging from intense irritability and anger to profound feelings of depression, emptiness and boredom."(Johnson, 1989, p.9).


In a majority of the research articles studies were done on both patients with anorexia nervosa and patients with bulimia nervosa. During the research, as we will see, there was some difference in the comorbidity of personality disorders depending on whether the subjects were anorexic or bulimic.


Unfortunately there are no clear cut answers to the question posed at the beginning of this article. The main reason, which I will describe later, has to due with the broad range of analisation techniques used by researchers in defining and determining the rate and impact of comorbidity. There isn't any question that there is a great deal of comorbidity between eating disorders and personality disorders because most of the research done has found some correlation. The question remains as to how much of a correlation there is and what effect the personality disorders have on eating disorders. In order to examine this closer it is helpful to look at the results of several major experiments.

Johnson, Tobin, and Enright (1989).

In this experiment 94 consecutive patients seeking treatment for eating disorders were evaluated for several other psychological problems including the prevalence of borderline disorder. 25% of the patients met the criteria for borderline personality disorder. These patients had relatively few differences in the their patterns of eating behaviors and attitudes when compared with nonborderline patients. The borderline patients were more disturbed on several other relevant emotional factors. The only aspect of eating behavior that varied between the two groups was that significantly more borderline patients reported having used laxatives and abused them as a form of purging. There was no difference found in the frequency of binging or vomiting or in the onset or duration of these symptoms. The greatest discrepancy occurred in the eating attitudes of those patients with borderline personality disorder. These patients were more disturbed than nonborderline patients when it came to beliefs about eating. Borderline patients reported a greater drive for thinness, more body dissatisfaction, and greater difficulty in identifying inner states and feelings yet these feelings didn't seem to be manifested in more binges or purges. Finally the borderline patients were more likely to engage in more restrictive dieting like fasting and crash diets.

Steiger and Stotland (1995).

In this experiment the researchers had some similar findings to those in the first experiment. Those with borderline personality disorder showed more psychiatric symptoms but the difference in the eating symptoms were less dramatic. Borderline patients, like those studied previously, had only slightly poorer responses in selected areas of disturbance. The researchers found that borderline patients had more of a drive towards thinness without displaying more binging and purging behavior than their nonborderline counterparts."Results corroborate others linking Axis-11 comorbidity to unfavorable outcome in the eating disorder, but suggest the character disturbance is more linked to course of general psychiatric symptoms than it is to the course of eating disturbances."(p. 1 49)

Skodol, Oldham, Hyler et al (I 993)

The researchers in the experiment had some very interesting results that highlight the problems with this form of research and problems associated with the other experiments. They found that the association between personality disorders and eating disorders varied according to the diagnostic method they used. They also found that specifically those patients with bulimia nervosa were associated with borderline personality disorder and those with anorexia nervosa were associated with avoidant disorder. Regardless they had the same general conclusion that comorbidity is characterized by chronicity and low levels of functioning and not necessarily and increase in patterns of eating habits. The researchers do feel that their research showed that comorbidity made eating disorders more severe due to the greater chronicity and lower levels of functioning.

Herzog, Martin, Lavod, et al (1992)..

In this experiment 210 subjects with eating disorders were interviewed to determine the prevalence of personality disorders within that group. The researchers found the most common diagnosis to be borderline personality disorder present in 9% of all the patients. Like the previous experiment there were no cases of borderline personality disorder found in those patients with anorexia. They found that overall those with personality disorders had a longer mean duration of illness, had more comorbid Axis I disorders and were more likely to have attempted suicide.(p.150) They also discovered that the group with the highest rate of personality disorders, those with bulimia nervosa had the youngest mean age of onset of their eating disorder, the longest mean duration of illness. This suggested that a long duration of illness may lead to a personality disorder, and personality disorders may predispose an individual to develop an eating disorder early.(p.151).

Wonderlich, Fullerton, Swift, et al (1994)

Like several of the previous experiments the researchers in this case also noted that subjects with personality disorders did not differ from those without personality disorders in the amount of symptomatic changes, in their patterns of eating , overtime. Yet for obvious reasons their psychopathology generally remained more severe in the cases of borderline personality disorder.

Problems with the Experiments.

There are several things that bring into question that validity of these experiments. All of the experiments probably have their own faults but it would take to long to go over all of them. Fortunately there are several problems that encompass all of the experiments reviewed. The biggest problem is in the way that patients are diagnosed with their disorders. The are many inconsistencies within the studies due to the use of different methods for personality disorder assessment and as well as the treatment setting in which they are seen.(Skodol, 1993, p.415). Because of the different methods use to measure personality disorders there can be discrepancies between the experiments even though there goal may be relatively the same.

Problems reporting the experiments.

Unfortunately due to the number of variables involved in these experiments it can be very difficult to brief them with out leaving out important information. The statistics involved with these forms of experiment are long and can be very complex, yet they can be vital to understanding the experiment. The briefings listed above are about cover the most important information relating directly to the question at hand.



After examining the experiments there are several key factors in the relationship between eating disorders and personality disorders. The most common conclusion reached by our experimenters was that borderline personality disorder is found more often in cases of eating disorders than any other personality disorder. Bordedine personality disorder affected those patients with both bulimia nervosa and anorexia, in most of the experiments. Several experiments reported few to no cases of borderline personality disorder in patients with anorexia nervosa. Another cross experiment observation is that personality disorders tend not to change the patterns of binging or purging even though it may effect many of the mental aspects of eating disorders including the feeling of needing to be thinner. One experiment also noted that many of the borderline patients were more likely to fast than their nonborderline counterparts. Another experiment noted that borderline patients had a tendency to use and abuse laxatives more often than nonborderline patients. The experimenters suggest that this may be a manifestation of the self-inflicted physical abuse associated with people with borderline personality disorders Overall the majority of experimenters stated two things about the experiments in general. First of all they all, in some way, were surprised by the outcome of the experiments. Many felt that the difference between borderline patients and nonborderline patients with eating disorders would be much greater. Finally many of them agreed that more research needed to be conducted and for longer periods of time with more consistent assessment methods.


Carroll, Julie M. et al. Specific Comorbidity Between Bulimia Nervosa and Personality Disorders. International Journal of Eating Disorders, Voll9(2), 159-170.(1996).

Herzog, David B. et al. The Prevalence of Personality Disorders in 210 Women With Eating Disorders.Jounal of Clinical Psych8atry, Vol. 53(5), 147-152. 1992.

Johnson, Craig et al. Prevalence of Clinical Characteristics of Borderline Patients in and Eating-Disordered Population. Journal of Clinical Psychiatry. Vol 50(l), 9-15. (1989).

Skodol, Andrew E. et al. Comorbidity of DSM-111-R Eating Disorders and Personality Disorders. International Journal of Eating Disorders Vol 14(4), 403-416.(1993).

Steiger, Howard et al. Prospective Study of Outcome in Bulimics as a Function of Axis-11 Comorbidity: Long Term Responses on Eating and Psychiatric Symptoms. Internationgl Journal of Eating Disorders, Vol 20 (2), 149-161.(1996).

Wonderlich, Stephen A. et al. Five Year OUtcom from Eating Disorders: Relevance of Personality Disorders. International Journal of Eating Disorders. Vol 15(3), 233-243.(1994).


Psychology Department

The Health Psychology Home Page is produced and maintained by David Schlundt, PhD.

Vanderbilt Homepage | Introduction to Vanderbilt | Admissions | Colleges & Schools | Research Centers | News & Media Information | People at Vanderbilt | Libraries | Administrative Departments | Medical 

  Return to the Health Psychology Home Page
  Send E-mail comments or questions to Dr. Schlundt


Search: Vanderbilt University
the Internet
  Help  Advanced
Tip: You can refine your last query by searching only the results by clicking on the tab above the search box
Having Trouble Reading this Page?  Download Microsoft Internet Explorer.