Personality and Individual Differences 53 (2012) 169–174

Contents lists available at SciVerse ScienceDirect

Personality and Individual Differences

journal homepage: www.elsevier.com/locate/paid

Interpersonal problems in anorexia nervosa: Social inhibition as defining

and detrimental
Jacqueline C. Carter a,b,⇑, Allison C. Kelly a,b, Sarah Jane Norwood c
a
Eating Disorders Program, Toronto General Hospital, University Health Network, Toronto, Canada
b
Department of Psychiatry, University of Toronto, Toronto, Canada
c
Department of Psychology, York University, Toronto, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Interpersonal difficulties are thought to play a central role in both the development and maintenance of
Received 5 August 2011 anorexia nervosa (AN). The primary aims of this study were to examine the nature of interpersonal prob-
Received in revised form 24 February 2012 lems in AN and to determine whether interpersonal problems are related to AN psychopathology and
Accepted 28 February 2012
treatment outcome. The participants were 218 individuals with AN admitted to a specialized treatment
Available online 24 March 2012
program. Overall, in comparison with a normative community sample, a pattern of difficulties with sub-
missiveness, nonassertiveness and social inhibition emerged among patients with AN. Results indicated a
Keywords:
positive association between interpersonal problems and eating disorder psychopathology at baseline.
Interpersonal problems
Anorexia nervosa
The overall level of interpersonal problems decreased from baseline to post-treatment and higher levels
Treatment outcome of social inhibition at baseline predicted treatment noncompletion. Our findings suggest that AN is asso-
ciated with a pattern of submissive and socially inhibited interpersonal behavior which contributes to the
maintenance of eating disorder pathology and interferes with treatment completion. The theoretical and
clinical implications of the findings are discussed.
Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction
Anorexia nervosa (AN) is a serious psychiatric disorder charac-
terized by extreme food restriction, maintenance of an abnormally
low body weight, intense fear of weight gain, and body image dis-
turbance (American Psychiatric Association, 2000). The focal
importance of interpersonal problems in AN has long been recog-
nized in the clinical literature (e.g., Bruch, 1973; Selvini-Palazzoli,
1974). Disturbances in family functioning have been highlighted
in models of both the etiology and maintenance of the disorder
(e.g., Lock, Le Grange, Agras, & Dare, 2001), and there is a growing
evidence base for the effectiveness of family therapy for children
and adolescents with AN (e.g., Le Grange, Binford, & Loeb, 2005).
Recently, researchers have begun to focus on the role of marital
functioning in the maintenance of AN in adults, and the use of cou-
ple therapy in the treatment of the disorder (Bulik, Baucom, Kirby,
& Pisetsky, 2011). AN is also associated with significant interper-
sonal dysfunction outside of the family (Schmidt, Tiller, & Morgan,
1995). A number of studies have found high rates of social anxiety
disorder (Godart, Flament, Lecrubier, & Jeammet, 2000) and avoid-
ant personality traits (Díaz-Marsá, Carrasco, & Sáiz, 2000; Skodol,
⇑ Corresponding author at: Eating Disorders Program, Toronto General Hospital,
200 Elizabeth Street, EN8-241, Toronto, Ontario, Canada M5G 2C4. Tel.: +1 416 340
4831; fax: +1 416 3404 198.
E-mail address: jacqueline.carter@uhn.ca (J.C. Carter).
Oldham, Hyler, & Kellman, 1993) among individuals with AN, as
well as difficulties with social withdrawal and social isolation
(e.g., Beaumont, 2002). To some extent, these interpersonal prob-
lems may be a consequence of the illness and, at the same time,
they may contribute to the maintenance of the disorder.
A growing interest in the role of interpersonal difficulties in the
maintenance of AN is reflected by recent research on interpersonal
treatment approaches for the disorder. Interpersonal models of AN
view eating disorder symptoms as ‘‘inextricably intertwined with-
in interpersonal relationships’’ and contend that interpersonal dif-
ficulties are often the trigger for symptoms (McIntosh, Bulik,
McKenzie, Luty, & Jordan, 2000). Thus, interpersonal treatments
for AN aim to reduce eating disorder pathology by improving inter-
personal functioning. Very little research has been conducted on
the effectiveness of interpersonal treatments for AN and the results
are mixed. McIntosh et al. (2000) adapted interpersonal psycho-
therapy (IPT), originally developed as a treatment for depression
(Klerman, Weissman, Rounseville, & Chevron, 1984), for the treat-
ment of AN. They found that IPT was less effective than cognitive
behavior therapy (CBT) and ‘‘specific supportive clinical manage-
ment’’ (SSCM) as a first-line intervention for acute AN in a random-
ized controlled study. SSCM was designed to mimic outpatient care
for AN in usual clinical practice and involved a combination of clin-
ical management and supportive psychotherapy. However, the
sample size in this study was small and the effect sizes in all three
conditions were relatively trivial.

0191-8869/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.paid.2012.02.020
170 J.C. Carter et al. / Personality and Individual Differences 53 (2012) 169–174
Schmidt et al. (2011) recently evaluated an interpersonal treat-
ment approach for AN (i.e., MANTRA). MANTRA is based on the
idea that AN symptoms facilitate the avoidance of emotions, and
that individuals with AN tend to be socially avoidant because close
interpersonal relationships may trigger the experience and expres-
sion of emotions. No difference was found between MANTRA and
SSCM for acute AN (Schmidt et al., 2011). However, the sample size
in this study was also small and the effect sizes obtained in both
conditions were fairly minor.
In addition to being a potential mechanism through which AN
symptoms can change, there is preliminary evidence that interper-
sonal problems at the start of therapy might moderate response to
treatment. Tasca, Taylor, Bissada, Ritchie, and Balfour (2004) found
that attachment avoidance predicted attrition while attachment
anxiety predicted treatment completion among patients with the
binge–purge subtype of AN (AN-BP), but not among patients with
the restricting subtype (AN-R) (Tasca et al., 2004). Thus, AN-BP pa-
tients high in attachment avoidance may find it difficult to main-
tain therapeutic bonds and may be more likely to disengage from
helping relationships, while those who worry about losing close
relationships may be more likely to remain in treatment.
According to Horowitz, Rosenberg, and Bartholomew (1993),
specific attachment styles are associated with particular patterns
of interpersonal problems. The Inventory of Interpersonal Prob-
lems (IIP) was developed to measure distress arising from interper-
sonal difficulties (Horowitz, Alden, Wiggins, & Pincus, 2000). The
IIP is based on a two-dimensional circumplex model that views
every interpersonal behavior along two dimensions—an affiliation
dimension ranging from cold (hostile) behavior to overly nurturant
(warm) behavior; and a dominance dimension that ranges from
nonassertive (submissive) to domineering (controlling) behavior
(see Fig. 1). According to this model, interpersonal problems can
be defined in terms of different combinations of affiliation or dom-
inance. For example, social avoidance is viewed as a combination
of cold and nonassertive, whereas intrusiveness is seen as a combi-
nation of overly nurturant and domineering.
The IIP appears to be a useful tool for measuring interpersonal
problems in AN since, clinically, individuals with AN tend to report
difficulties along these two dimensions—assertiveness and social
avoidance. To date, only one study has examined interpersonal
problems in AN using the IIP. Hartmann, Zeeck, and Barrett
(2010) found that patients with eating disorders report particularly
pronounced interpersonal difficulties with nonassertiveness and
with putting others’ needs before one’s own. Patients with AN-BP
reported more difficulties with social avoidance and lack of close-
ness to others than patients with AN-R, and these difficulties did
not improve with inpatient or day hospital treatment (Hartmann
et al., 2010). Taken together, preliminary studies suggest that
interpersonal problems may vary according to AN subtype and
may affect treatment outcome.
Fig. 1. Diagram of the interpersonal circumplex from Horowitz et al. (2000).
The overall goals of the current study were to characterize
interpersonal difficulties in AN and to determine whether certain
interpersonal problems are associated with AN psychopathology
and treatment outcome. In order to address these goals, the current
study had five specific aims. The first aim was to examine whether
patients with AN-R and AN-BP differ on IIP subscales. The second
aim was to compare the IIP profiles of AN patients to a normative
community sample as described in Horowitz et al. (2000). The
third aim was to examine the association between interpersonal
functioning and eating disorder psychopathology. The fourth aim
was to examine whether interpersonal problems in AN would im-
prove with remission of the eating disorder. The final aim was to
examine whether interpersonal problems at baseline would pre-
dict treatment noncompletion.
2. Methods
2.1. Participants
The participants were a consecutive series of 218 individuals
who met Diagnostic and Statistical Manual of Mental Disorders,
Fourth Edition Revised (DSM-IV-Revised) criteria for AN based on
the Eating Disorder Examination (EDE; Fairburn & Cooper, 1993)
interview. All were admitted to the hybrid inpatient/day treatment
unit of the Eating Disorders Program of the Toronto General Hospi-
tal between 2000 and 2008. This is a specialized program for se-
vere AN operated by an interdisciplinary team. Program goals
include medical stabilization, weight gain to a healthy level, erad-
ication of binge eating and purging symptoms, normalized eating,
and intensive group psychotherapy (Olmsted et al., 2010). The pro-
gram accommodates up to 12 patients in the group at a time who
are together for 30–40 h weekly, which creates an intensive thera-
peutic milieu.
At admission to the inpatient unit, the 218 participants had a
mean age of 26.0 years (SD = 7.6) and a mean BMI of 14.8
(SD = 1.8). Three percent were male and 97% were female. The
mean duration of AN was 7.1 years (SD = 6.8) and the mean age
of onset of AN was 18.9 years (SD = 6.1). The average length of stay
was 13.5 weeks (SD = 6.4) and the mean weight gain was 11.5 kg
(SD = 6.3). Eighty-three percent were single, 12% were married or
living in common-law relationships and 5% were separated or di-
vorced. Most participants were students (42.2%), 38.7% were em-
ployed, and 19.1% were unemployed. With regard to racial
background, 87% were Caucasian, 3% were Asian, 2% were Afri-
can-Canadian, and 8% were Jewish or European. 39% of participants
met DSM-IV criteria for the binge–purge subtype of AN (AN-BP),
while the remaining 61% had the restricting subtype of the illness
(AN-R).
2.2. Measures
The following measures were administered at pre- and post-
treatment:
2.2.1. Eating Disorder Examination-Questionnaire (EDE-Q; Fairburn &
Bèglin, 1994)
Eating disorder psychopathology was measured using the fifth
edition of the EDE-Q. The EDE-Q produces four subscale scores:
shape concern, weight concern, eating concern, and dietary re-
straint, which can be combined into one Global score. It has been
shown to have good internal consistency and test–retest reliability
(Luce & Crowther, 1999).
 J.C. Carter et al. / Personality and Individual Differences 53 (2012) 169–174 171

Table 1
Means, standard deviations, and Cronbach’s alphas for the subscales of the IIP for patients with AN and the Normative Community Sample Comparison Group.

Alpha AN Pre-treatment AN Post-treatment Community

M SD M SD M SD
Domineering–Controlling .71 3.90 3.35 2.78 2.90 4.50 4.10
Vindictive/Self-Centered .76 3.23 4.11 2.44 3.20 4.80b 4.90
Cold-Distant .85 6.58 4.23 5.35 4.30 5.10a 5.60
Socially Inhibited .84 9.40 4.51 7.79 4.75 6.40a 5.70
Nonassertive .76 10.52 4.10 9.97 4.24 8.00a 6.10
Overly Accommodating .80 10.70 3.90 10.01 3.75 8.60a 5.40
Self-Sacrificing .71 10.20 4.10 9.14 3.94 8.80a 5.50
Intrusive-Needy .68 4.51 3.57 4.14 3.38 5.40b 4.60
a
Significantly lower than AN sample pre-treatment.
b
Significantly higher than AN sample pre-treatment.

Table 2
Hierarchical logistic regression demonstrating the association between AN subtype and interpersonal problems with treatment outcome.

B SE Wald p OR 95% CI
Step 1
AN Subtype .93 .29 10.61 .001 .40 .23–.69
Step 2
Domineering/Controlling .25 .17 2.18 .14 .78 .56–1.10
Vindictive/Self-Centered 2.24 .17 1.76 .19 .80 .58–1.13
Cold/Distant .31 .19 2.70 .10 1.36 .94–1.97
Socially Inhibited .48 .21 5.40 .02 .62 .41–.93
Nonassertive .29 .27 1.16 .28 1.34 .79–2.30
Overly Accommodating .20 .30 .45 .50 .82 .50–1.50
Self-Sacrificing .22 .19 1.30 .25 .80 .55–1.17
Intrusive/Needy .27 .17 2.55 .11 1.3 .94–1.83
Step 3
AN  Domineering/Controlling .29 .36 .69 .41 .75 .37–1.50
AN  Vindictive/Self-centered .28 .36 .57 .45 1.32 .65–2.70
AN  Cold/Distant .41 .40 1.10 .30 .66 .30–1.44
AN  Socially Inhibited .052 .44 .01 .91 1.05 .44–2.51
AN  Nonassertive .20 .58 .12 .73 .82 .26–2.60
AN  Overly Accommodating .15 .64 .05 .82 1.16 .33–4.10
AN  Self-Sacrificing .33 .41 .65 .42 1.40 .62–3.10
AN  Intrusive/Needy .49 .36 1.81 .18 .61 .30–1.25

Note. N = 218. Cox and Snell R2 for the model = .14. AN = anorexia nervosa; OR = odds ratio; CI = confidence interval.

2.2.2. Inventory of Interpersonal Problems-32 (IIP-32; Horowitz et al.,
2000)
This is a 32-item self-report measure that was shortened from
the 64-item IIP (Horowitz et al., 2000). Respondents are asked to
rate the extent to which they have had difficulty with various
interpersonal problems on a 4-point Likert scale from 1 (not at
all) to 4 (extremely). The IIP-32 is composed of eight 4-item sub-
scales: Domineering/Controlling (problems giving up control of
interpersonal situations); Vindictive/Self-Centered (problems car-
ing for and being involved with others); Cold/Distant (problems
feeling close to others); Socially Inhibited (problems being soci-
able); Nonassertive (problems asserting oneself), Overly Accom-
modating (problems with not standing up for oneself); Self-
Sacrificing (problems with placing the needs of others before one’s
own), and Intrusive/Needy (problems being open with regards to
one’s personal life). Subscales scores range from 4 to 16, and a total
score is obtained by summing the subscale scores. Higher scores on
all subscales reflect greater levels of interpersonal difficulties. This
measure has demonstrated strong internal consistency for each
subscale (Horowitz et al., 2000). For the purposes of the current
study, each subscale was used as an individual predictor and dem-
onstrated acceptable reliability (see Table 1).
2.3. Statistical analyses
All statistical analyses were conducted using SPSS version 19
(SPSS, Chicago). First, a series of two-tailed t-tests were conducted
to examine mean baseline differences in IIP subscale scores among
patients with AN-R (n = 132) and AN-BP (n = 86). Second, IIP scores
for AN patients were compared to a normative community sample
of females (n = 400) as described in Horowitz et al. (2000). An alpha
level of .01 was used to control the Type 1 error rate. The normative
community sample consisted of 400 females aged 18–89 years. Eth-
nicity, education level, and region were stratified across groups and
100 females from each age range (18–24, 25–44, 45–64, and 65+)
were included so that the sample was representative of the United
States (Horowitz et al., 2000).
Third, zero-order correlations between each of the IIP subscales
and the EDE-Q Global score at baseline were calculated to ascertain
the association of interpersonal problems with the core psychopa-
thology of AN. Next, a repeated measures multivariate analysis of
variance (RM MANOVA) was used to examine changes in interper-
sonal problems over the course of treatment. AN subtype was
employed as the between-group variable (two levels, AN-R and
AN-BP), and time (two levels; pre-treatment and post-treatment)
as the within-group variable. Pairwise comparisons using the Bon-
ferroni correction were used to ascertain the point of the significant
interaction or main effect. An overall alpha level of p < .05 was used.
This analysis was performed on only those participants who suc-
cessfully completed the treatment program.
Finally, a hierarchical logistic regression was performed to
examine whether interpersonal problems predicted treatment
non-completion. The dependent variable in this analysis was
whether or not the patient successfully completed the treatment
172 J.C. Carter et al. / Personality and Individual Differences 53 (2012) 169–174
program. Successful treatment completion was defined as achiev-
ing a body mass index (BMI) of at least 20 at the end of treatment
and remittance of binge eating and purging behavior during the
last 4 weeks of treatment. Patients who did not meet these criteria
were defined as non-completers. AN subtype was dummy-coded
and entered into block one, the eight subscales of the IIP were en-
tered into block two, and finally the interactions between AN sub-
type and each interpersonal problem were entered as a block into
step three. An alpha level of .05 was used.
3. Results
3.1. Baseline comparisons between AN-R and AN-BP
Overall, individuals with AN-BP (M = 62.92, SD = 17.09) re-
ported significantly greater levels of total interpersonal problems
than individuals with AN-R (M = 56.25, SD = 17.63) [t(216) = 2.76,
p = .006]. Baseline comparisons between AN-R and AN-BP on each
of the IIP subscales revealed that patients with AN-BP reported sig-
nificantly higher scores on the Vindictive/Self-Centered subscale
only than those with AN-R [t(216) = 2.88, p = .004]. However, it is
of note that scores on this subscale were actually quite low for both
subtypes (AN-R, M = 2.60, SD = 3.34; AN-BP, M = 4.21, SD = 4.93).
No other significant differences on any of the other IIP subscales
emerged between the two AN subtypes. Because there were next
to no baseline differences on IIP subscale scores between the two
subtypes, we conducted subsequent analyses on the sample as a
whole.
3.2. Baseline comparisons with normative community sample
In comparison with the normative community sample, patients
with AN reported significantly lower scores on the Intrusive/Needy
subscale [t(616) = 2.51, p = .01] and the Vindictive/Self-Centered
subscale [t(616) = 4.03, p < .001)], and significantly higher scores
on the Overly Accommodating [t(616) = 4.97, p < .001)], Nonasser-
tive [t(616) = 5.50, p < .001)], Social Inhibition [t(616) = 6.71,
p < .001)], Cold/Distant [t(616) = 3.50, p < .001)], and Self-Sacrific-
ing [t(615) = 3.22, p = .001)] subscales (see Table 1).
3.3. Baseline correlations between IIP subscales and EDE-Q Global
score
Severity of eating disorder pathology, as measured by the EDE-Q
Global score, was positively and significantly correlated with the
Domineering/Controlling (r = .24, p < .001), Cold/Distant (r = .33,
p < .001), Socially Inhibited (r = .42, p < .001), Nonassertive (r = .34,
p < .001), Overly Accommodating (r = .41, p < .001), and Self-Sacri-
ficing (r = .31, p = .003) subscales of the IIP.
3.4. Impact of treatment on interpersonal problems
One hundred and twenty-four (57.1%) patients successfully
completed the program, 40 (18.4%) dropped out and 53 (24.4%)
were discharged prematurely by staff due to insufficient progress
or failure to adhere to program norms. Treatment completers re-
mained in treatment for significantly more weeks (M = 16.84,
SD = 5.30) than noncompleters (M = 9.03, SD = 4.92), t(216) =
11.11, p < .001). Among treatment completers, there was a signifi-
cant within-subjects main effect of time (Wilk’s K = .75, F (8,
81) = 3.38, p = .002, partial g2 = .25) indicating that the level of
interpersonal problems significantly decreased over time for both
subtypes. However, there was no significant effect of AN subtype
(p = .67) nor was there a subtype by time interaction (p = .32). In or-
der to further explore the significant main effect of time, univariate
tests were conducted. For both subtypes, scores on the Domineer-
ing/Controlling (p = .006), Cold/Distant (p = .005), Nonassertive
(p = .031), Overly Accommodating (p = .026), Self-Sacrificing
(p = .001), and Intrusive/Needy (p = .012) subscales significantly de-
creased from pre- to post-treatment.
3.5. Interpersonal problems as predictors of treatment outcome
Individuals with AN-R were significantly more likely to com-
plete the program than those with AN-BP (70.2% vs. 29.8%;
t(216) = 3.41, p < .001). In order to investigate whether certain
interpersonal problems predicted treatment success, a hierarchical
logistic regression analysis was performed with treatment out-
come status of the participants entered as a categorical dependent
variable. AN subtype was dummy-coded and entered into block
one, the eight subscales of the IIP were entered into block two,
and finally the interactions between AN subtype and each interper-
sonal problem were entered as a block into step 3 (see Table 2). AN
subtype significantly predicted treatment outcome, v2(1) = 10.84,
p = .001. Specifically, AN subtype correctly classified 52.1% of those
who did not complete the program and 69.9% of those who did, for
an overall success rate of 62.2%. After controlling for AN subtype,
the IIP subscales showed a trend approaching significance in pre-
dicting treatment outcome, v2(8) = 15.11, p = .057. Examination
of the individual subscales revealed that only Social Inhibition sig-
nificantly contributed to the model (see Table 2), indicating that as
levels of Social Inhibition increased the odds of successful treat-
ment completion decreased. The model correctly classified 51.1%
of those who did not complete treatment and 76.4% of those who
did, for an overall success rate of 65.4%. A test of the model on
the full dataset (including interaction terms between each IIP sub-
scale and AN subtype) was not statistically significant, v2(8) = 6.36,
p = .607 (see Table 2), indicating that there was no difference be-
tween the two subtypes in terms of how interpersonal problems
affected response to treatment.
4. Discussion
The primary goals of this study were to examine the nature of
interpersonal problems in AN and to determine whether interper-
sonal problems are related to AN psychopathology and treatment
outcome. On average, patients with the binge/purge subtype of
AN reported greater overall interpersonal problems than patients
with the restricting subtype, and interpersonal problems were
associated with AN psychopathology across both subtypes. In com-
parison to a normative community sample, patients with AN re-
ported lower than normal levels of intrusiveness/neediness,
suggesting a lack of closeness in their relationships, and lower than
normal levels of vindictiveness/self-centeredness, suggesting great
difficulty expressing feelings of anger. AN patients also reported
more problems with being unassertive, socially inhibited, cold
and distant, overly accommodating, and self-sacrificing in their
relationships. We found that those interpersonal problems in
which AN patients were high tended to decrease over the course
of treatment, with the exception of social inhibition, which not
only remained unchanged but also predicted treatment non-com-
pletion. Social inhibition therefore seems to be defining, stable,
and detrimental in individuals with AN.
Our findings are consistent with previous research indicating
that people with AN tend to struggle with socially inhibited and
submissive traits (e.g., Tasca et al., 2004; Troop, Allan, Treasure,
& Katzman, 2003; Westen & Harnden-Fischer, 2001). The findings
are compatible with Schmidt and Treasure (2006) cognitive-inter-
personal model of AN suggesting that people with AN tend to avoid
close relationships, possibly as a way of avoiding the experience
 J.C. Carter et al. / Personality and Individual Differences 53 (2012) 169–174
and expression of emotions. Our results also suggest that people
with AN often have difficulty asserting their needs, and it is possi-
ble that AN symptoms may function to communicate certain
needs—for example, the need for care and attention—without hav-
ing to communicate these needs directly.
The current results indicated that the severity of eating disorder
psychopathology was significantly associated with the level of
interpersonal problems. Eating disorder severity scores were sig-
nificantly positively correlated with six out of the eight IIP sub-
scales suggesting that greater interpersonal difficulties in both
the affiliation and dominance dimensions are associated with more
severe dietary restraint, eating concern, as well as more extreme
concerns about eating, shape and weight. Given the cross-sectional
nature of the study, it is not possible to establish the direction of
this association—it may be that more severe eating disorder symp-
toms produce greater interpersonal dysfunction and it is also pos-
sible that greater interpersonal dysfunction contributes to more
severe eating disorder symptoms or to the maintenance of eating
disorder psychopathology.
In terms of treatment outcome, there were significant improve-
ments in interpersonal problems among those who successfully
completed the treatment program, and there were no differences
between AN-BP and AN-R in terms of improvements on the IIP.
Specifically, there were improvements in difficulties with feeling
close to others, asserting oneself, and placing the needs of others
before one’s own. It is possible that the strong group therapy ele-
ment of the program contributed to improvements in these inter-
personal difficulties. Interestingly, we did not detect significant
improvements in terms of social inhibition. This may suggest that
social avoidance is the interpersonal difficulty that is most inter-
twined with AN symptoms given that it seems to be resistant to
change. It is also possible that difficulties with social inhibition
may improve with time. However, a limitation of this study is that
we did not collect follow-up data on the IIP. It is intriguing that
there were decreases on the Domineering/Controlling subscale
(difficulty giving up control of interpersonal situations) and on
the Intrusive/Needy subscale (sharing too much about one’s per-
sonal life) even though these scores were not elevated compared
to normal controls at baseline.
In terms of the association between IIP scores and treatment
outcome, higher levels of social inhibition predicted treatment non-
completion, but AN subtype did not moderate this relationship.
Engagement with treatment is a significant challenge in this popu-
lation. Research by our group and others has shown that attrition
from intensive treatment programs is high and tends to occur soon-
er and at a substantially higher rate among AN-BP patients than AN-
R patients (Kahn & Pike, 2001; Woodside, Carter, & Blackmore,
2004). Consistent with previous studies, patients with AN-BP in
the current study were significantly more likely to drop-out or be
prematurely discharged from the treatment program, but our find-
ings did not uncover any evidence to suggest that interpersonal
problems explain this difference in rates of noncompletion. For
individuals with an avoidant interpersonal style, it is possible that
the level of interpersonal intensity in the program was perceived
as progressively threatening as it became associated with the risk
of being exposed to conflict, criticism and negative emotions.
The current study had certain limitations. First, the sample was
recruited from a tertiary care clinic that treats severely ill patients
and the findings may not generalize to other AN samples. Second,
interpersonal problems were measured using a brief self-report
measure and it is possible that responses may have been biased
by the effects of starvation or the desire to deny or minimize the
level of distress. Third, our analyses rely on only two time points
making it impossible to tease apart the temporal relationship be-
tween changes in interpersonal problems and changes in AN
symptoms.
173
The current findings may have some possible clinical implica-
tions. First, patients with AN may benefit from interventions de-
signed to help them identify and express their feelings and needs
directly. Second, since social avoidance was associated with pre-
mature termination of treatment and did not improve with treat-
ment of the eating disorder, it may be important to specifically
target social inhibition within acute treatment programs for AN.
Since most programs are group therapy based, it may be useful
to include an interpersonal group therapy component that is de-
signed to specifically address problems with social inhibition.
Future research should examine the temporal relationship be-
tween improvements in eating disorders symptoms and improve-
ment in interpersonal difficulties during treatment. It would also
be important for future studies to investigate the mechanisms by
which social avoidance leads to early withdrawal from AN treat-
ment. One interesting possible mechanism worth investigating is
emotion avoidance, as Schmidt et al. (2011) have suggested that
the desire to avoid experiencing intense emotions may explain
the social inhibition seen in AN. Finally, future research should
examine the course of interpersonal difficulties following acute
treatment of AN to determine whether certain problems continue
to improve post-discharge.
Acknowledgement
This research was supported by the Canadian Institutes of
Health Research (CIHR).
References
American Psychiatric Association. (2000). Diagnostic and statistical manual of
mental disorders, (4th ed.), Text Revision (DSM-IV-TR). Washington, DC:
American Psychiatric Association Press.
Beaumont, P. J. (2002). Treatment of anorexia nervosa. Journal of Consulting
Psychology, 23, 241–2249.
Bruch, H. (1973). Eating disorders: Obesity, anorexia nervosa and the person within.
New York, NY: Basic Books.
Bulik, C. M., Baucom, D. H., Kirby, J. S., & Pisetsky, E. (2011). Uniting Couples (in the
treatment of) Anorexia Nervosa (UCAN). International Journal of Eating Disorders,
44, 19–28.
Díaz-Marsá, M., Carrasco, J. L., & Sáiz, J. (2000). A study of temperament and
personality in anorexia and bulimia nervosa. Journal of Personality Disorders, 14,
352–359.
Fairburn, C. G., & Cooper, Z. (1993). The eating disorder examination (12th ed.). In C.
G. Fairburn & G. T. Wilson (Eds.), Binge eating: Nature, assessment and
treatment (pp. 317–360). New York, NY: Guilford.
Fairburn, C. G., & Bèglin, S. J. (1994). Assessment of eating disorders: Interview or
self-report questionnaire? International Journal of Eating Disorders, 16,
363–370.
Godart, N. T., Flament, M. F., Lecrubier, Y., & Jeammet, P. (2000). Anxiety disorders in
anorexia nervosa and bulimia nervosa: Co-morbidity and chronology of
appearance. European Psychiatry: The Journal of the Association of European
Psychiatrists, 15, 38–45.
Hartmann, A., Zeeck, A., & Barrett, M. S. (2010). Interpersonal problems in eating
disorders. International Journal of Eating Disorders, 43(7), 619–627.
Horowitz, L. M., Alden, L. E., Wiggins, J. S., & Pincus, A. L. (2000). Inventory of
interpersonal problems. San Antonio, TX: The Psychological Corporation.
Horowitz, L. M., Rosenberg, S. E., & Bartholomew, K. (1993). Interpersonal problems,
attachment styles, and outcome in brief dynamic therapy. Journal of Consulting
and Clinical Psychology, 61, 549–560.
Kahn, C., & Pike, K. M. (2001). In search of predictors of dropout from inpatient
treatment for anorexia nervosa. International Journal of Eating Disorders, 30,
237–244.
Klerman, G., Weissman, M., Rounseville, B., & Chevron, E. (1984). Interpersonal
psychotherapy of depression. New York, NY: Basic Books.
Le Grange, D., Binford, R., & Loeb, K. L. (2005). Manualized family-based treatment
for anorexia nervosa: A case series. Journal of the American Academy of Child &
Adolescent Psychiatry, 44, 41–46.
Lock, J., Le Grange, D. l., Agras, W. S., & Dare, C. (2001). Treatment manual for anorexia
nervosa: A family-based approach. New York, NY: Guilford Press.
Luce, K. H., & Crowther, J. H. (1999). The reliability of the Eating Disorder
Examination-Self-Report Questionnaire Version (EDE-Q). The International
Journal of Eating Disorders, 25, 349–351.
McIntosh, V. V., Bulik, C. M., McKenzie, J. M., Luty, S. E., & Jordan, J. (2000).
Interpersonal psychotherapy for anorexia nervosa. International Journal of Eating
Disorders, 27, 125–139.
174 J.C. Carter et al. / Personality and Individual Differences 53 (2012) 169–174
Olmsted, M. P., McFarlane, T. L., Carter, J. C., Trottier, K., Woodside, D. B., &
Dimitropoulos, G. (2010). Inpatient and day hospital treatment for anorexia
nervosa. The treatment of eating disorders: A clinical handbook (pp. 198–211).
New York, NY: Guilford Press.
Schmidt, U., Oldershaw, A., Jichi, F., Landau, S., Sternheim, L., Startup, H., et al. (2011,
September). A Randomised Controlled Trial of the Maudsley Model of
Treatment for Adults with Anorexia Nervosa (MANTRA) compared to
specialist supportive clinical management in outpatients with anorexia
nervosa. Paper presented at the annual meeting of the Eating Disorders Research
Society, Edinburgh, Scotland.
Schmidt, U., Tiller, J., & Morgan, H. G. (1995). The social consequences of eating
disorders. Oxford, England: John Wiley & Sons.
Schmidt, U., & Treasure, J. (2006). Anorexia nervosa: Valued and visible. A cognitive-
interpersonal maintenance model and its implications for research and practice.
British Journal of Clinical Psychology, 45, 343–366.
Selvini-Palazzoli, M. P. (1974). Self Starvation. London, England: Chaucer Publishing
Co..
Skodol, A. E., Oldham, J. M., Hyler, S. E., & Kellman, H. D. (1993). Comorbidity of
DSM-III-R eating disorders and personality disorders. International Journal of
Eating Disorders, 14, 403–416.
Tasca, G. A., Taylor, D., Bissada, H., Ritchie, K., & Balfour, L. (2004). Attachment
predicts treatment completion in an eating disorders partial hospital program
among women with anorexia nervosa. Journal of Personality Assessment, 83,
201–212.
Troop, N. A., Allan, S., Treasure, J. L., & Katzman, M. (2003). Social comparison and
submissive behaviour in eating disorder patients. Psychology and Psychotherapy:
Theory, Research and Practice, 76, 237–249.
Westen, D., & Harnden-Fischer, J. (2001). Personality profiles in eating disorders:
Rethinking the distinction between axis I and axis II. The American Journal of
Psychiatry, 158, 547–562.
Woodside, D. B., Carter, J. C., & Blackmore, E. (2004). Predictors of premature
termination of inpatient treatment for anorexia nervosa. The American Journal of
Psychiatry, 161, 2277–2281.