Personality Disorders Among Subjects Recovered
from Eating Disorders
Hisato Matsunaga,1,2* Walter H. Kaye,1 Claire McConaha,1 Katherine Plotnicov,1
Christine Pollice,1 and Radhika Rao1
1
Western Psychiatric Institute and Clinic, University of Pittsburgh Medical Center,
Pittsburgh, Pennsylvania
2
Department of Neuropsychiatry, Osaka City University Medical School, Osaka, Japan
Accepted 8 October 1998
Abstract: Objective: Personality disorders are common in symptomatic eating disorders
subjects. Because personality symptoms could be exaggerated by malnutrition or Axis I
disorders, we studied women who had recovered from eating disorders for at least 1 year to
see if personality disorder symptoms persisted in the well state. Method: Personality disorders
were evaluated in 10 women recovered from anorexia nervosa (AN), 28 women recovered
from bulimia nervosa (BN), and 16 women recovered from AN and BN, using the Structured
Clinical Interview for DSM-III-R personality disorders. Results: Fourteen of 54 subjects (26%)
met the criteria for at least one personality disorder, such as self-defeating, obsessivecompulsive, or borderline personality disorder. Cluster B personality disorders were closely
associated with bulimic subtypes. Conclusions: While a recovery from eating disorders may
have an attenuating influence on the symptoms of personality disorders, such personality
disorder diagnoses persist after recovery in some recovered subjects. © 2000 by John Wiley
& Sons, Inc. Int J Eat Disord 27: 353–357, 2000.
Key words: eating disorder; personality disorder; recovery
INTRODUCTION
Personality disorders (PDs) have been extensively investigated in patients with eating
disorders (EDs; Gartner, Marcus, Halmi, & Loranger, 1989; Wonderlich, Swift, Slotnick, &
Goodman, 1990; Braun, Sunday, & Halmi, 1994; Kennedy et al., 1995; Matsunaga, Kiriike,
Nagata, & Yamagami, 1998). In spite of a close association between PDs and EDs, it has
been suggested that as one of the problems inherent to the categorical typology for PDs,
the presence of Axis I disorders, such as depression, influences the expression or the
measurement of PDs (Shea, Widiger, & Klein, 1992; Zimmerman, 1994; McDavid & Pil-
*Correspondence to: Hisato Matsunaga, M.D., Department of Neuropsychiatry, Osaka City University Medical
School, 1-4-3 Asahi-machi Abeno-ku, Osaka city, Osaka 545-8585, Japan.
© 2000 by John Wiley & Sons, Inc.
Prod. #1476
354
Matsunaga et al.
konis, 1996). Similarly, treatment for Axis I disorders may have an attenuating influence
on the course and stability of PD diagnoses (Ricciardi et al., 1992; McDavid & Pilkonis,
1996). Some studies on PD comorbidity among ED patients pointed out the significant
effect of the clinical course of EDs on the PD diagnoses, including borderline PD (AmesFrankel et al., 1992; Steiger, Stotland, & Houle, 1994). In addition, the presence of depression or anxiety in a majority of eating-disordered subjects may confound attempts to
assess personality functioning, as the acute disturbance may bias the judgments of both
subjects and their evaluators (Vitousek & Manke, 1994). It is thus possible that a variety
of clinical features typical of EDs might exert some influence on the assessment of the
personality pathology. It is uncertain as to whether the personality characteristics of ED
subjects, commonly described in the previous studies, may truly reflect their enduring
personality pathology, even under the possible state-biasing effect caused by the presence
of EDs. To examine these issues, we assessed PDs among subjects who were recovered for
at least 1 year from EDs.
METHOD
The subjects were 10 women recovered from anorexia nervosa (RAN), 16 women recovered from anorexia nervosa and bulimia nervosa (BAN), and 28 women recovered
from bulimia nervosa (NWB) diagnosed according to criteria outlined in the 3rd Rev. ed.
of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R; American Psychiatric Association, 1987). After receiving a complete description of this study, all subjects gave signed informed consent to participate in the study. All subjects had previously
received treatment for EDs at Western Psychiatric Institute and Clinic (WPIC) at the
University of Pittsburgh Medical Center Health System. Each subject had maintained a
stable normal weight and had not binge eaten, purged, or engaged in restrictive eating
patterns for at least 1 year prior to this study. To assess the DSM-III-R PDs, all subjects
were directly interviewed face-to-face using the Structured Clinical Interview for DSMIII-R PDs (SCID-II; Spitzer, Williams, & Gibbon, 1987). Interviewers were master’s or
doctoral-level psychologists with extensive experience in diagnostic assessment with
structured interviews.
Significant mean group differences for parametric variables were evaluated using oneway analysis of variance (ANOVA) followed by the Scheffe test. Nonparametric variables
were compared with chi-square tests with 1 df and Yate’s correction for discontinuity.
Fisher’s exact tests were also used for factors with expected cell frequencies of less than
5. Significance level was set at p < .05.
RESULTS
There were no significant differences in age and length of recovery among the three ED
groups. Subjects in the BAN group were significantly younger in age at onset than subjects in the NBW group (F = 2.54, df = 2,51, p < .05). Both subjects in the RAN and BAN
groups had significantly lower percentage of average body weight (ABW) in their lifetime
than subjects in the NWB group (F = 71.3, df = 2,51, p < .01). Subjects in the NWB group
were significantly more likely than the other groups to have had higher percent ABW in
their lifetime (F = 6.15, df = 2,51, NWB vs. RAN; p < .01, NWB vs. BAN; p < .05).
In the assessment of PDs using the SCID-II, 14 (26%) of the subjects, including 2 RAN
Personality Disorders
355
subjects (20%), 6 BAN subjects (38%), and 6 NBW (21%), met the threshold diagnoses for
at least one PD. When all of the subjects were considered together, self-defeating PD was
most commonly found (11%), followed by obsessive-compulsive PD (9%), borderline and
dependent PDs (7%,each), histrionic PD (6%), and then avoidant PD (4%). Among three
clusters of PDs, the anxious-fearful (cluster C) PDs, such as obsessive-compulsive, dependent, or avoidant PD were most frequently diagnosed, followed by the dramaticerratic (cluster B) PDs, such as borderline PD. With respect to the odd-eccentric (cluster A)
PDs, there were only 2 subjects in the BAN group who met the criteria for either paranoid
or schizotypal PD.
Table 1 compares the prevalence of three clusters of PDs, and borderline, obsessivecompulsive, avoidant, and self-defeating PDs among the three groups. In the comparison
of the threshold diagnoses of each cluster or PD, no significant differences were detected
among the three groups. However, after including the subthreshold diagnoses of each PD
in this comparison, both subjects in the BAN and NWB groups were significantly more
likely than subjects in the RAN group to have cluster B PDs (Fisher’s exact test, p < .05).
They also showed a trend toward a higher prevalence of borderline PD compared with the
subjects in the RAN group.
DISCUSSION
A considerable proportion (26%) of the subjects recovered from EDs met threshold
criteria for DSM-III-R PD diagnoses. This is lower than the rate of PD found in subjects
Table 1.
Comparison of personality disorders
Cluster A
Thr
Total
Cluster B
Thr
Total
Borderline
Thr
Total
Cluster C
Thr
Total
Obsessive-compulsive
Thr
Total
Avoidant
Thr
Total
Self-defeating
Thr
Total
RAN
(N = 10)
BAN
(N = 16)
NWB
(N = 28)
0
1 (10)
2 (13)
3 (19)
0
2 (7)
0
0
4 (25)
7 (44)*
4 (14)
11 (39)*
0
0
2 (13)
4 (25)
2 (7)
7 (25)
1 (10)
4 (40)
7 (44)
8 (50)
3 (11)
11 (39)
1 (10)
3 (30)
3 (19)
6 (38)
1 (4)
8 (29)
0
2 (20)
1 (6)
3 (19)
1 (4)
3 (11)
1 (10)
1 (10)
2 (13)
5 (31)
3 (11)
4 (14)
Note: Values are expressed as numbers (%). Thr = threshold diagnoses; Total = total numbers of threshold and
subthreshold diagnoses. RAN = recovered from anorexia nervosa; BAN = recovered from anorexia and bulimia
nervosa; NWB = recovered from bulimia nervosa.
*p < .05 compared to the anorexia nervosa group. Comparisons are made using chi-square tests with 1 df and
Yate’s correction for discontinuity. Fisher’s exact tests were also used for factors with expected cell frequencies
of less than 5.
356
Matsunaga et al.
with active EDs (51–74%; Gartner et al., 1989; Wonderlich et al., 1990; Braun et al., 1994;
Kennedy et al., 1995; Matsunaga et al., 1998). There may be several reasons for this
difference. First, the rate of PDs, which may predict poor ED outcome, may be higher in
a chronically ill sample (Wonderlich, Fullerton, Swift, & Klein, 1994; Steiger & Stotland,
1996). Second, malnutrition or an Axis I disorder may exaggerate behaviors such as
unstable mood which may inflate the rate of PD-like behaviors (Zimmerman, 1994; McDavid & Pilkonis, 1996; Ames-Frankel et al., 1992).
Because we studied subjects after recovery, the number of subjects studied was small.
Thus, we collapsed PDs in the three main subdivisions. The clearest finding in this study
was the higher rate of cluster B PDs in recovered bulimic subtypes and the absence of this
diagnosis in recovered RAN. Such a close association has been found in studies of symptomatic ED subjects (Wonderlich et al., 1990; Herzog, Keller, Lavori, Kenny, & Sacks, 1992;
Rossiter, Agras, Telch, & Schneider, 1993; Braun et al., 1994; Vitousek & Manke, 1994;
Matsunaga et al., 1998). In addition, compared to recovered RAN, these bulimic people
also showed a trend for a higher prevalence of borderline PD as found by others (Wonderlich et al., 1990; Herzog et al., 1992; Braun et al., 1994; Vitousek & Manke, 1994;
Matsunaga et al., 1998).
Similarly, as reported in patients with active EDs (Gartner et al., 1989; Braun et al., 1994;
Matsunaga et al., 1998), cluster C PDs tended to be more common compared to cluster A
or B PDs. In the studies of active EDs, avoidant PD was identified to be the most common
cluster C PD, whereas in this study obsessive-compulsive PD was most commonly found
in all ED subtypes. Cluster C PDs, especially avoidant PD, have been characterized as
being most susceptible to state effects of Axis I disorders among DSM-III-R PDs (Loranger
et al., 1991; Ricciardi et al., 1992). On the other hand, compulsivity and perfectionism,
which are major factors for obsessive-compulsive PD, have been considered to exert
significant effects on the emergence and maintenance of ED symptoms in anorexics
(Strober, 1980; Kaye, 1997; Pollice, Kaye, Greeno, & Weltzin, 1997) and bulimics (Johnson
& Holloway, 1988; Tobin, Johnson, Steinberg, Staats, & Dennis, 1991). The persistence of
elevated perfectionism was also reported in recovered ED subjects (Strober, 1980; Srinivasagam et al., 1995). Thus, it can be hypothesized that the close linkage between cluster C
PDs and EDs may be attributed not only to the inherent susceptibility of these PDs to the
state effects of Axis I disorders, but to the underlying trait closely related to the essential
basis for EDs.
In conclusion, a considerable proportion (26%) of the subjects recovered from EDs
showed significant severity in personality pathology, enough to qualify for PD diagnoses,
such as self-defeating, obsessive-compulsive, or borderline PD. The inclusion of subthreshold diagnoses of each PD to this study led us to find some similar relationships
between PDs and EDs to those reported in active ED subjects, such as a specific linkage
with cluster B PDs, especially borderline PD in bulimics. These findings suggest that
although treatment for EDs may have an attenuating influence on the assessment of PDs,
there may be an essential linkage between EDs and PDs. The cross-sectional design of this
study prevented us, however, from exploring how the state artifact caused by the presence of EDs effects the assessment of PDs.
REFERENCES
American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd Rev ed.).
Washington, DC: Author.
Personality Disorders
357
Ames-Frankel, J., Devlin, M.J., Walsh, B.T., Strasser, T.J., Sadik, C., Oldham, J.M., & Roose, S.P. (1992). Personality disorder diagnoses in patients with bulimia nervosa: Clinical correlates and changes with treatment.
Journal of Clinical Psychiatry, 53, 90–96.
Braun, D.L., Sunday, S.R., & Halmi, K.A. (1994). Psychiatric comorbidity in patients with eating disorders.
Psychological Medicine, 24, 859–867.
Gartner, A.F., Marcus, R.N., Halmi, K., & Loranger, A.W. (1989). DSM-III-R personality disorders in patients
with eating disorders. American Journal of Psychiatry, 146, 1585–1591.
Herzog, D.B., Keller, M.B., Lavori, P.W., Kenny, G.M., & Sacks, N.R. (1992). The prevalence of personality
disorders in 210 women with eating disorders. Journal of Clinical Psychiatry, 53, 147–152.
Johnson, N.S., & Holloway, E.L. (1988). Conceptual complexity and obsessionality in bulimic college women.
Journal of Counseling Psychology, 35, 251–257.
Kaye, W.H. (1997). Neurobiology and genetics: Anorexia nervosa, obessional behavior, and serotonin. Psychopharmacology Bulletin, 33, 335–344.
Kennedy, S.H., Katz, R., Rockert, W., Mendlowitz, S., Ralevski, E., & Clewes, J. (1995). Assessment of personality
disorders in anorexia nervosa and bulimia nervosa: A comparison of self-report and structured interview
methods. Journal of Nervous and Mental Disease, 183, 358–364.
Loranger, A.W., Lenzenweger, M.F., Gartner, A.F., Susman, V.L., Herzig, J., Zammit, G.K., Gartner, J.D.,
Abrams, R.C., & Young, R.C. (1991). Trait-state artifacts and the diagnosis of personality disorders. Archives
of General Psychiatry, 48, 720–728.
Matsunaga, H., Kiriike, N., Nagata, N., & Yamagami, S. (1998). Personality disorders in patients with eating
disorders in Japan. International Journal of Eating Disorders, 23, 399–408.
McDavid, J.D., & Pilkonis, P.A. (1996). The stability of personality disorder diagnosis. Journal of Personality
Disorders, 10, 1–15.
Pollice, C., Kaye, W.H., Greeno, C.G., & Weltzin, T.E. (1997). Relationship of depression, anxiety and obsessionality to state of illness in anorexia nervosa. International Journal of Eating Disorders, 21, 367–376.
Ricciardi, J.N., Baer, L., Jenike, M.A., Fischer, S.C., Sholtz, D., Buttolph, M.L. (1992). Changes in DSM-III-R Axis
II diagnoses following treatment of obsessive-compulsive disorder. American Journal of Psychiatry, 149,
829–831.
Rossiter, E.M., Agras, W.S., Telch, C.F., & Schneider, J.A. (1993). Cluster B personality disorder characteristics
predict outcome in the treatment of bulimia nervosa. International Journal of Eating Disorders, 13, 349–357.
Shea, M.T., Widiger, T.A., & Klein, M.H. (1992). Comorbidity of personality disorders and depression: Implications for treatment. Journal of Consulting and Clinical Psychology, 60, 857–868.
Spitzer, R.L., Williams, J.B., & Gibbon, M. (1987). The Structured Clinical Interview for DSM-III-R personality
disorders. New York: New York State Psychiatric Institute.
Srinivasagam, N.M., Kaye, W.H., Plotnicov, K.H., Greeno, C., Weltzin, T.E., & Rao, R. (1995). Persistent perfectionism, symmetry and exactness after long-term recovery from anorexia nervosa. American Journal of
Pschiatry, 152, 1630–1634.
Steiger, H., & Stotland, S. (1996). Prospective study of outcome in bulimics as a function of Axis II comorbidity:
Long-term responses on eating and psychiatric symptoms. International Journal of Eating Disorders, 20,
149–161.
Steiger, H., Stotland, S., & Houle, L. (1994). Prognostic implications of stable versus transient borderline features
in bulimic patients. Journal of Clinical Psychiatry, 55, 206–214.
Strober, M. (1980). Personality and symptomatological features in young, nonchronic anorexia nervosa patients.
Journal of Psychosomatic Research, 24, 353–359.
Tobin, D.L., Johnson, C., Steinberg, S., Staats, M., & Dennis, A.B. (1991). Multifactorial assessment of bulimia
nervosa. Journal of Abnormal Psychology, 100, 14–21.
Vitousek, K., & Manke, F. (1994). Personality variables and disorders in anorexia nervosa and bulimia nervosa.
Journal of Abnormal Psychology, 103, 137–147.
Wonderlich, S.A., Fullerton, D., Swift, W.J., & Klein, M.H. (1994). Five-year outcome from eating disorders:
Relevance of personality disorders. International Journal of Eating Disorders, 15, 233–243.
Wonderlich, S.A., Swift, W.J., Slotnick, H.B., & Goodman, S. (1990). DSM-III-R personality disorders in eatingdisorder subtypes. International Journal of Eating Disorders, 9, 607–616.
Zimmerman, M. (1994). Diagnosing personality disorders: A review of issues and research methods. Archives
of General Psychiatry, 51, 225–245.