J .p,j<krar. Ra.. Vol 28. No. 5. pp 413-423. 1994
Copyrght c 1994 Elsewer Saencr Ltd
Prmted I” Great Britam All rights reserved
0022~~3956,94 $7 OOf 00
0022- 3956( 94) 00022- O
TRAITS IN EATING
ERIN I. KLEIFIELD, SUZANNE SUNDAY, STEPHEN HURT and KATHERINE A. HALMI
Cornell University Medical College, Westchester Division, White Plains, New York, U.S.A
(Rrceirvd for puhlicalion 3 Max 1994)
Summary---The Tridimensional Personality Questionnaire (TPQ) was tested in four subgroups
eating-disorder patients: anorectic-restrictors (AN--R). anorectic-bulimics
bulimics (BN). and bulimics with a past history of anorexia
were matched for gender and age. All subjects completed
addition to the TPQ. AN-Rs scored lower on the Novelty Seeking scale than the bulimic groups
and controls. and the two normal weight bulimic groups had higher Novelty Seeking scores than
the controls. On the Harm Avoidance scale. all eating disorder
than the control group. In addition. the AN-Rs scored lower than the AN-Bs and B-ANs. The Harm
Avoidance scale and depression scores were positively correlated
and depression scores were negatively correlated.
Novelty Seeking and Persistence scales remained
out. These results are discussed in terms of the Tridrmensional
measure of traits with eating disorder subjects.
(AN-B), normal weight
(B--AN). Normal controls
the Beck Depression (BDI) in
groups scored significantly higher
while the Reward Dependence
on the Differences
as a stable
THE RELATIVE contribution
opment of eating disorders
disorders (e.g. Ploog, 1984). However, attempts
teristics within eating disorder
of the DSM systems (Garner et al., 1990; Schmidt & Telch, 1990; Welch et al., 1990; Yates
et al., 1989). This may be due to the limitations
personality disorders (e.g. DSM-III-R criteria).
produce more consistent results.
The Tridimensional Personality Questionnaire
and measure behaviors associated with three dimensions
(NS-tendency toward intense exhilaration
tendency toward intense avoidance of aversive stimuli), and “reward
tendency toward intense response to rewards,
behaviors associated with these dimensions are reported to be highly routinized
within individuals (Cloninger, 1987a,b). The basic psychometric
of genetic, personality,
has been formulated
to find consistencies
factors in the devel-
models of eating
in personality charac-
subgroups mixed results within the context
of a categorical
of personality may
(TPQ) was developed
particularly rewards). The
integrity of the TPQ has
Road, White Plains, NY 10605. U.S.A.
to: Erin I. Kleifield, New York Hospital-Cornell, University Medical College. 21 Bloomingdale
E. I. K LHFIELD et al
received empirical support in investigations
Svrakic et al., 1991). In limited application
patients (Pfohl et al., 1990) and in a prospective
onset alcohol abuse (Cloninger
useful in making predictions
The TPQ may have particular
may highlight differences
that have been obscured
DSM. Further, the descriptions
Patients with eating disorders
literature. These observations
Bulimic patients have been found to show a tendency
behaviors (e.g. lying, stealing),
prominent mood lability and poor self-control,
et al.. 1990; Norman & Herzog, 1984; Russell, 1979: Williamson
model on which the TPQ was founded
underlying tendency to seek out experiences
tendency reflected in ‘he NS scale of the TPQ.
Anorectic patients. on the other hand,
disposition toward behavioral
conventional values (Casper, 1990). Strober (1980) found anorcctic
an obsessional character structure marked by heightened
highly regimented behavior. and rigid adherence
standards. He also found anorectics
affect. These behaviors are also relevant to the dimensions
should be characterized by low scores on NS and high scores on the Persistence
of the RD scale.
The TPQ has additional relevance
eating disorder subgroups and normal controls
descriptions. Cloninger (1987b) originally
indicator of noradrenergic (RD). dopaminergic
Neurophysical studies of patients with eating disorders
neurotransmitter functioning (e.g. Kaye et al., 1991; McBride et al., 1991). If these differ-
ences can be linked to behavioral differences assessed by the TPQ, additional
to clarify the interrelationships of these neurotransmitter
Although the descriptive and empirical literature
personality characteristics of eating disorder
ducted using the TPQ with eating disorder subjects.
bulimic patients. Wailer et al. (1991) found elevated
these patients. In another report (Brewerton
with clinical samples of obsessiveecompulsive
et al., 1988), the dimensions
of personality styles derived
samples (see Cloninger, 1987a;
of the TPQ were found to be
from the descriptive. clinical
relevance to the eating disorders.
of eating disorder
of eating disorder
patients and normal controls
approaches such as those of the
found in the DSM parallel the
by more traditional
are commonly believed to demonstrate extremes of per-
clinical and such beliefs have been documented
also have received some support
in the descriptive,
in empirical investigations.
et al., 1985). The theoretical
are the result of an
exhilaration and excitement.
to engage in impulsive
and to have interpersonal
to have associated abuse problems.
suggests that these behaviors
that produce a
have been found
to display a temperamental
with a strong belief in
patients to demonstrate
industriousness and responsibility.
and excessive conformance
to be interpersonally insecure
of the TPQ and anorectic patients
and emotional combined
to rules and
and to show minimal
as an instrument for assessing differences
beyond its relationship
(NS) and scrotoncrgic
the TPQ as a possible behavioral
differences have demonstrated in
systems becomc possible on a
In one study examining
NS and HA. and low RD scores in
et al., 1993). the scores of bulimics
to the commonality
have been con-
the TPQ with
TPQ IN E AT ING DISOR DE R S 415
disorder patients scored significantly
While these studies suggest that the TPQ may be a useful instrument
patients, the small numbers of subjects in some of the diagnostic
to take into account other comorbid
depression) compromise the results. Present state factors should be accounted
conclusions about more enduring differences in personality
Several reports have indicated that state variables
nificantly affect patient’s responses on personality
This issue has special relevance for assessing personality
subjects hospitalized for treatment because of the high levels of depression
in these patients (e.g. Swift et al., 1986). Furthermore,
recent evidence (Brown et al., 1992) indicating
conditions of anxiety and depression.
The purpose of this paper was to evaluate the utility of the TPQ to distinguish
characteristics among four subtypes of eating disorder patients with anorexia
This was accomplished by controlling
comparing these subgroups with normal control subjects.
by past history of anorexia),
and patients with both anorexia
showed that all subtypes
to female controls. of eating
higher P scores.
with eating disorder
subgroups and the failure
higher on HA, than controls.
higher levels of NS and AN patients
states which have been shown to affect the TPQ (i.e.
can be drawn.
such as anxiety and depression
inventories (see Loranger
et al., 1991).
among eating disorder
with respect to the TPQ, there is
that the HA scale is affected by current state
for the effects of current levels of depression, and
The eating disorder
were divided into four subgroups
(according to DSM-III-R):
and no history of bulimia
nosis of anorexia
n = 27) and bulimics
patients, these eating disorder
population in the U.S.
Control subjects were recruited
and prior weight status, dietary habits, drug history, psychiatric
Subjects were included in the study if they did not have a current
eating disorder or obesity, substance
normal weight. Fifty-one subjects met the inclusion
TPQ and BDI questionnaires and were included
subjects were 97 females hospitalized
unit at the Cornell Medical
based on current and historical
(AN-R; n = 29) anorectic-bulimics,
and bulimia (AN-B; n = 21) bulimics with no history of anorexia
with a past history of anorexia
patients represent a small segment
for an eating disorder
eating disorder diagnosis
those with a present diagnosis
those with a present diag-
Center-Westchester The patients
(B-AN; n = 20). As hospitalized
of the eating disorder
from a local college and the community
subjects completed a screening
questionnaire which assessed current
history, and family history.
or past history
illness, and were currently
criteria. All 51 subjects completed
in the study.
abuse, or psychiatric
designed to assess, as originally
1987a) is a 100 item
The NS scale is comprised
gance (NS3), and Disorderliness
to be curious,
scale tend to be reflective, stoical, slow-tempered
The four HA subscales are: Worry!‘Pessimism
with Strangers (HA3), and Fatigability;‘Asthenia
tend to be apprehensive.
tend to be optimistic. carefree. outgoing
Cloninger and his colleagues
subscale has been removed and analyzed
investigations (Svrakic et al.. I99 I ; Kleitield et al.. 1993) revealed that the P subscale formed
a separate factor from the remaining RD subscales.
scale and RD is comprised of the remaining
(RDI ). Attachment (RD3). and Dependence
to be sentimental, socially sensitive, and tender-hearted.
tend to be insensitive, practical, tough-minded
The P scale measures persistence at non-socially
on this scale are determined and resolute; those scoring low give up easily and arc irresolute.
The Beck Depression Inventory (BDI) (Beck et al., I Y61). consists of 21 items rctlccting
symptoms and cognitions associated with depression.
disturbance. I7 20 borderline clinical depression,
over 30 indicate severe dcprcssion.
(RD). These higher-order
Novelty Seeking (NS). Harm Avoidance
of: Exploratory Excitability
(NS4). Thus. individuals
quick-tempered, and disorderly,
(HA), and Reward
scoring high on the NS scale tend
whereas those low on the NS
(HA I), Fear of Uncertainty
and fatigable, whereas those low on the HA scale
have recently revised the RD scale so that the Persistence
as a separate scale. Factor analyses in two recent
are comprised subscales.
high on the HA scale
In this paper, P is treated as a separate
three subscales. including:
(RD4). Individuals high on the RD scale tend
whereas those low on the RD scale
mediated tasks. Individuals
A score of I I I6 indicates mild mood
and 2 I 30 moderate
the weight gain protocol.
bingeing and purging and to establish
For the anorcctic
Thus. for the majority
the TPQ and BDI questionnaires
subjects. the weight gain protocol
For the bulimic subjects.
determination of their non-clinical
within 2 weeks of hospital
began after I week 01‘
subjects, testing preceded cntrancc
the program is designed
Control subjects completed
the TPQ scales. Then. because
depression. and previous
TPQ scales (Svrakic
each of the TPQ scales with BDI scores as the covariate.
simple. pairwise contrasts
While such a procedure as the Bonfcrroni
evaluating mean differences
by the TPQ, analyses of variance
of the significant
et al.. IYY I ). covariance
of eating disorder patients dilrered on personality
A priori. single degree of freedom,
were conducted following the ANOVAs
correction is a widely accepted
when multiple post-hoc comparisons
on each of
on levels of
and the ANCOVAs.
arc carried out, it is an
TPQ IN E AT ING DISOR DE R S
with an overall F test. Because our hypotheses
with a priori comparisons,
to determine the relationships
with Systat version 5. I.
to preplanned, single degree of freedom comparisons
were a priori and could be
we report these data. Pearson correlations
between the TPQ scales and the BDI. All analyses
Table I shows the mean age and BDI score for the eating disorder
control subjects. For the eating disorder
weight at current weight) is also shown. There were no significant
the five groups. As expected, there were significant
disorder subgroups [F(3,93) = 57.49. p < .OOl], with both of the anorectic
and AN-B) significantly lower in percent of ideal weight than the BN and B-AN groups
[F(1,93) = 156.06. p < .OOl]. The AN Rs were also significantly
AN-Bs [F( 1,93)
= 6.88, p < ,011. The BN and B-AN groups did not differ in current body
Depression scores (BDI) differed significantly
= 35.61, p < .OOl]. Patients with bulimia and either current
cantly more depressed than the average patient
(AN-B group: [F(l,142) = 16.44, p < .OOl]: B&AN group:
All eating disorder groups were significantly
142) = 130.83, p < .OO
I]. The means ( f SD) for the TPQ scales appear in Table 2. The
ANOVAs revealed significant differences between diagnostic
On the NS scale. the AN-R group had the lowest mean score which was significantly
lower than the mean score for the control
significantly lower than the mean score
I. 144) = 17.89. p < .OO I].
The two normal weight bulimic groups (BN and B-AN) scored
significantly higher than the control group [F(
On the HA scale, control subjects had the lowest mean score and were significantly
than the patient groups [F(1,144) = 38.19. p < .OOl]. Among
of ideal subgroups, weight (shown
differences in age between
differences in weight between the eating
more emaciated than the
or past anorexia
OY bulimia with either anorexia alone
[F(1,142) = 27.96, p < .OOl].
more depressed than the control group
groups on all TPQ scales.
[F(1,144) = 6.64. p < ,011 and was
the remaining three patient groups
144) = 5.8, p < ,051.
the the patient groups,
(II = 29)
(n = 21)
Mean percent of
ideal weight at
(II = 28)
(II = 20)
(II = 51)
418 E. 1. KL~IFIELL> et al
(I/ = 2’))
(n = ?I)
(U = 27)
(II = 20)
(PI = 51)
= 5.3Y.p < .()()I.
= 2.37. ,I < .05.
ANY Rs showed the lowest mean score which was significantly
combined diagnostic groups [F( 1,144) = 6.34, p < ,011.
On the RD scale, the bulimic
were significantly lower than the control
score for the AN-R group was higher than that of the bulimic groups and below that of
the control group.
On the P scale. the mean scores of patients
(AN--R). or a past history of anorexia with current
the patient groups. The AN-Rs had a mean score that was significantly
of the AN -B, BN and control groups [F( 1.144) = 8.69, p < .Ol].
The results of the ANCOVAs varied from those of the ANOVAs
RD scales. Scores on these two scales were significantly
F( 1.141) = 26.74.~ < ,001 and r = -.33, F( 1.141) = 6.49,/l < .01. respectively].
ses of between groups differences after adjustment
for the HA and RD scales. Because the scores on the BDI and the NS and P scales were
not significantly correlated. covariate adjustment
ANOVAs reported above.
lower than that of the two
groups (AN B, BN. and B-AN)
scored the lowest and
[F(1,144) = 10.X2. p < .OOl]. The mean
with a current
(B- AN), were the highest of
higher than those
of anorexia only
only for the HA and
with BDI scores [F = .5X. correlated
for the BDI scores were nonsignificant
made no difference in the results of the
This study was designed
with eating disorder
sample on TPQ personality
major TPQ scales.
Overall, the most consistent
between the AN--Rs and bulimic
controls. The dual diagnosis
would differ from one another,
dimensions, were supported
personality traits from a dimensional
hypotheses of this study. that subgroups
and differ from the normative
by the ANOVAs
conducted on the
in this study (AN-Bs
TPQ IN EATING DISORDERS
like those who are only bulimic (BNs) than those who are purely anorectic
same finding has been documented in a number
& Halmi, 1992) and lends support for distinguishing
more, the pattern of TPQ personality
those observed in the normal population.
The relationships observed here between the NS dimension
NS and bulimia nervosa validate, and perhaps
accounts of these disorders presented
their tendency to demonstrate an obsessional
industriousness and responsibility, highly regimented
excessive conformance to rules and standards
depict bulimics as externalizers who are impulsive
1990). In line with these accounts, the AN-Rs
orderly and obsessional (low NS scores) than the bulimic groups and the controls.
over, while all bulimic groups had elevated
normal weight bulimic groups (BN and B-AN) also had higher NS scores than the controls.
The BN patients in the studies by Waller et al. (1991) and Brewerton
exhibited high levels of NS.
Cloninger (1987b) hypothesized that NS reflects behavioral
principally modulated by the neuromonoamine
taneous exploratory behavior by mammals in a novel environment
of mesolimbic dopaminergic projections and that dopamine
activation. As mentioned in the Introduction,
impulsive and antisocial behaviors which could well represent
seek out experiences that produce exhilaration
NS scale of the TPQ.
Dopaminergic systems are necessary for self-administration
major link in the role of food as a reinforcer.
dopamine turnover during feeding (Heffner et al., 1986) which suggests that central dopa-
mine mechanisms mediate rewarding effects of food as they mediate
intracranial self-stimulation and self-administration
impulsive and anti-social behaviors and binge eating behavior may represent a dysregulation
of the dopaminergic neurotransmitter system. The bulimic eating disorder subgroups
significantly higher on the NS dimension than the AN-Rs
disposition toward behavioral and emotional
Results of the ANOVAS on the HA scale revealed that all eating disorder groups scored
significantly higher than the controls, and that the AN-Rs
combined diagnosis groups (AN-Bs and BPANs). Waller et al. (199 1) and Brewerton
(1993) also found increased HA scores among their BN, and AN, AN-B and BN patients.
respectively. Cloninger (1987b) hypothesized
to serotonergic activity which also facilitates
responses (Halmi & Sunday, 1991). and show evidence
(McBride et al., 1991). Anorectic patients who exhibit emotional
obsessional character structure may have increased serotonergic
of other studies (for review, see DaCosta
these groups of anorectics.
traits among eating disorder
patients, differs from
and anorexia nervosa, and
suggest a mechanism
in the clinical literature.
for, the descriptive
Anorectics are noted for
behavior, and rigid adherence
1980). In contrast,
in this study were significantly
(Strober, clinical reports
the two NS scores relative to the AN-Rs,
et al. (1993) also
depends on the integrity
have a tendency
and excitement, a tendency
which in turn is
noted that spon- dopamine.
reflected in the
behaviors and could be a
hypothalamic There is evidence of increased
drugs. The bulimics’
who display a temperamental
scored lower than the two
that HA, or behavioral
have a defect in satiety
activity (Kaye et al., 1991).
and have an
group. have the sane increased prevalence of anxiety disorders as do the anorectic
when cotnparcd to a control population
two behaviors in bulimia that are influenced
syndrome behaviors. Therefore. the role of serotonin
On the RD scale, the three bulimic groups scored significantly
sample. On Ihc P scale (formerly a subscale of RD). the AN Rs scored signiticantly
than the AN -Bs, BNs and controls. Cloninger
variation in behavioral maintenance, or rcsistancc
behavior. Consistent with this formulation.
adherence to their beliefs and have highly regimented
Cloningcr further suggests. on the basis of animal
modulated by the noradrenergic neurotr~tnstnitter
worthy that norepinephrine has a role in regulating
ventricular nucleus (PVN). the lateral hypothalamus,
Abnormalities in noradrcnct-gic activity
patients (Kaye et al., IWO. 1984). The persistent
term. so-called weight restored anorcctics.
continued aberrant eating behavior and dieting?
activity afl‘ect and sustain abnormal eating beha\ ior?
The three neurotransliiitters (dopamine.
the personality ditnensions of NS. HA. and RD arc all involved
beha\.ior. The studies of‘ neurotransmitter
above arc preliminary investigations in this arca. These initial promising
further and tnore extensive investigations
for the three dimensions of pcrxonnlity defined by C’loninget-.
Further questions are raised by the significant
dimensions. In the present study, the current
on the HA and RD scales. but not on the NS or P scales. The results here corroborate
those from a longitudinal assessment of the TPQ conducted
three months of treatment (Brown et al.. IW?). Thcsc investigators
measured by the HA scale were influenced by states of anxiety and depression.
measured by the NS scale remained relatively
These findings raise interesting questions
substrates mediating the HA and RD dimensions
hypothesis of depression postulates that some fortns of depression
relative deficiency of norepinephrinc ;tt central
serotonin hypothesis of depression (e.g. Brown et al.. 1991). T~LIS, it may be that
dysregulation of the scrotonergic system corresponds
HA. while dysregulation in the noradrenergic
depression and RD. One approach to testing this notion
in depression over Cmc correlate with changes on TPQ scales.
in u sitnple picture of anorectics
is that bulimia
having activity nervosa patients.
(Braun et al.. 1994). This means there are at Lust
by serotonin: eating bchnviors
function in bulimia
lower than the normative
anoreclics are typically obsessional
and in the perifornical
have been found in both anorcctic
low CSF norepinephrinc
needs an explanation.
If so, how does decreased
that RD represents
of previously remarded
with :I rigid
1987b). It is note-
levels in long-
Does this merely reflect
Ihat RD is
serotonin, and norepincphrine) associated with
function in eating disorder mentioned
in eating disorder patients of the biological basis
impact of depression on some TPQ
level of depression exerted a significant ctfect
\vith depressed patients o\et
also found that traits
about uhcther there arc common
are attributable to ;I
Others have proposed an anal-
to cle\.ations in both depression
be to obser\,c how changes
system corresponds both in
TPQ IN EATING DISORDERS
A recent study (Kleifield et al., 1994) provides suggestive evidence to support this hypoth-
esis. In this study, the relationship between depression,
and changes on the TPQ scales over time were measured.
observed between mood and HA; any group differences and treatment
were traceable to levels of, and changes in levels of depression.
the same way, suggesting that these scales are sensitive to changes in depression
The NS and P scales were unaffected by depression
scales seem especially sensitive to the effects of current mood states, prospective
assess changes in these scales as mood improves
Before discussing the implications of TPQ responses
eating disorder patients, two caveats should be issued. First, because the findings
study are correlational, it remains to be determined
traits are operative in the formation of bulimic behaviors,
Likewise. in the case of anorexia, it remains
AN-Rs to persist at activities with a higher threshold
the anorectic condition. Prospective studies would be useful in this regard.
validity of the TPQ in predicting response to treatment
be determined. This is necessary before statements
strategies for particular patient groups.
Nevertheless, the current findings bear on our understanding
and treatment of patients with eating disorders.
exhibit dimensional traits most distinct from the other groups, including
are both anorectic and bulimic, suggests that a specially tailored
these patients is needed. Regarding the style and tone of treatment.
style of anorectic patients and their penchant
they would be likely to respond well to treatment
specified goals. These goals should be carefully articulated
successfully complete one goal before proceeding
goal of weight restoration, specific psychosocial
these patients not persist so tenaciously and obsessively
decrease rigid, black and white thinking and open options and explore alternatives.
these data underscore the importance of considering
before drawing conclusions about steadfast personality
because many personality variants are exaggerated
treatment of depression must be a priority when treating the eating disordered
changes in depression
effects on this scale
Mood affected RD in much
or treatment. Because the HA and RD
under treatment would be required.
for the clinical management of
whether the presence of novelty seeking
or merely an outcome of bulimia.
to be determined whether
for fatigue precedes or results from
the tendency for
has yet to
and treatment outcome
can be made about appropriate
of the personality structure
anorectics First, the fact that restricting
those patients who
given the personality
and orderly behavior,
to the next. In addition
treatment goals should
for highly regimented
that is well structured
to the obvious
traits of eating disordered
by current state of depression.
such as depression
Reprint requests should be sent to Dr Erin I. Kleifield. New York Hospital
21 Bloomingdale Road, White Plains. New York. 10605.
authors wish to thank Drs C Robert Cloninger and Thomas R. Przyheck for their
to thts study.
Cornell University Medical College.
Beck. A. T., Ward. C. H.. Mendelson.
‘I l~:~/?iw.\ O/‘Gw~w/I P~r~~hicrrr~. 4, 5S63.
M.. Mock. J.. & Erbaugh. J. (1961). An inventory for measuring depression.
E. I. KI.EIFIELD et al.
P.~~d~~lo,yic~ul Mrdicine, In press.
Brewerton, T. D., Hand. L. D., & Bishop, E. M. Jr. (1993). The tridimensional
disorder patients. lntrrnationnl Journcrl of Eating Disord~w. 14, 213 ~218.
Brown, S. L.. Bleich, A.. & van Praag,
serotonin. In S. L. Brown & H. M. van Praag (eds.) 7% role of rwotor~in in p.v~~c~hicrt,-ic~
Brown, S. L.. Svrakic. D. M., Przybeck, T. R.. & Cloninger.
and anxiety states: A dImensiona approach.
Caspcr, R. C. (1990). Personality features of women
52, I56 170.
Clomnger. C. R. (lY87a). A systematic method
ilwl7iw.s o/ Grrwrrl Ps,~chicrr~~~, 44, 573-588.
Cloningcr. C. R. (lY87b). Neurogcnetic adaptive
Cloninger. C. R.. Sigvardsson. S.. & Bohman. M. (1988). Childhood
adults. AIddi.nx Clinicd crnci Espc~r.irmwtol Re.xwtd~. 12, 494~ 505.
DaCosta. M.. & Hahni. K. A. (1992). Classifications
Jowwrl o/‘&/&/ ni.sort/w.v. I I) 305 3 13.
Garner. D. M.. Ohnsted, M. P.. Davis. R.. Rockert.
hetwjeen bulimic symptoms and reported psychopathology.
Halmi, K. A.. & Sunday. S. R. (1091). Temporal
anorexia and bulimia. .4/~/wfilc. 16, 219 237.
Heffner. T. G. Hartman. J. A., & Seiden. L. S. (1986). Dopamine
Kaye, W. H.. Guirtsman. H. E.. & George. D. T. (1991). Altered
long-term weight restoration. ilwhiw.\ of (;c/~~LI/ P.\,u hirrr~~~~.
Kaye. W. H.. Ehert. M. H.. & Gwirtsman. H. E. (1990). CSF monoamine
ewdence for abnormal noradrenergic activity.
Kaye, W. H., Ebert, M. H.. & Raleigh. M. (1984). Abnormalities
nervosa. Awhirc,s of tiwwal
P.c,~dficr/r~~. 41, 350 -3 55.
Kleifield, E. 1.. Sunday. S.. Hurt, S.. Br Halmi.
personality questionnaire: Application to subgroups
Kleifield, E. 1.. Sunday. S.. Hurt, S. Bi Hahni. K. A. (lYY3). The effects ofdeprealon
Biolqqic~crl P.\~h;u~,‘~~. 36, 6X 70.
Loranger. A. W.. Lenzenweger. M. F.. Gartner.
Abrams. R.. & Young. R. C. (I99 I). Trait-state
Gcnrrcrl P.y~c/lic//~~~, 48, 720~-728.
McBride, P. A.. Anderson. G M.. Khait, V.. Sunday.
P.~~~hophrr~i?f~~~~~/~~~~, Rullcrin. 27, 365- 372.
Norman. D. K.. & Herzog. D. B. (1984). Persistent
Auwrium Joutxtrl of I’.\ drirrrr,v,
141, 444 446.
Pfohl, B., Black. D.. Noyes. R. Jr., Kellcy, M.. & Blum, N. (1990). A test of the tridimenaional
theory: .4ssociation with diagnosis and platelet imipramlne
P.\Jdlirl/,‘~. 28, 4 I --46.
Ploog. D. (1984). The importance of physiologic.
anorexia nervosa. In K. M. Pirke & D. Ploog (Ed?.). Tl,c /,\~,[,hohio/~!r/~. of wow\-w
D.. Sunday, S. R., & Halml. K. A. (lYY4). Psychiatric comorbidity in patients with eating disorders.
personality questionnaire in eating
H. M. (1991). The monoamine hypothesis of depression:
o’i.w,rkw.s. New York:
The role of
C. R. (1992). The relationship
Journcrl o/‘P.s~c~hirrtric R~~.wcrrd~. 26, 192-21 I.
with good outcome
of personality to mood
for clinical description and classitication of personality variants.
mechanisms in alcoholism. S&rw, 236, 410-416.
predicts alcohol abuse in young personality
of anorexia nervosa; questlon of subtypes. Ir~rwrwtiorrtrI
W.. Goldhloom. D., & Eagle. M. (1990). The association
Inrcwclrio/tal Jowwl of Ea/ir?g L)i.vo~clc~~.s.
of hunger and fullness ratings and related cognitionh
9, I 15.
in feeding. S~~icwc. 208, I 16X-l 170.
actiwty 117 anorexia nervosa after
Ic\cls in normal weight bulimia and
Anw~%zn Jow?tr/ of Pvd~iarr~~. 147, 225 23Y.
in CNS monoamine metabolism 111 anorexia
K. A. ( 1903). Psychometric
of eating disorders.
C’o/,r/“r’/~l,n.~;~.~, P.c ~d?/tr/r~,. 34, 249
of the tridimenslonal
and treatmsnt on the TPQ.
A. I-‘.. Susman,
V. L.. Herzig. J.. Zammlt.
and the diagnosis
<;. K.. Gartner.
of personality .4w/riw.\ of’
S.R.. & Halmi. K. A. (IYY I). Serotonerglc responsivitl in
social maladjustments 111 bulimia: A one-year Ihllowup.
Bi~~/qqia/ hlnding in obsessiw compulsive disorder.
and endocrine studies for the undcrstandmg
(pp. I 3). Ne\r
eaters, and normal controls.
Strober. M. (1980). Personality
Jounrcrl o/ Ps,~c~ho.son~crtic~
Svrakic. D. M., Prrvbeck.
the unified biosocial
Swift. W. J., Andrew,
disorders: A rcvicw of the literature.
Wailer, D. A.. Petty, F., Hardy.
N. B.. & T&h. M. J. (1990). Prevalence
Jwr/rtr/ (!f P.\?,c,/lo/‘f/‘l’/~/o~~, & R&~r~iow/ .4s.vcssnwuf. 12, I69 185.
and symptomatological features In young, nonchronic
R~wwc~/~, 24, 353-359.
T. R., & Clomnger. C. R. (I90 I). Further
model of personality; U.S. and Yugosla\
D.. & Barklagc. N. E. (1986). The relationship
AIW~IW~I Jourrwl o/‘P.syd~irr/r~~. 143, 290-299.
B. W.. Gullion, C. M.. Murdock,
and serotonin in bulimia nervow.
of personality disorders amon g bulimia.
anorexia nervosa patients.
contribution to the conceptual validity of
data. (‘~,nl/,~c,/lerr.sir.~, P.s~~hicrt~~~: 32, lY5 2Oi.
betbeen aKe&ve disorder and eating
M. V., & Rush. A. J. (1991). Tridimensional
Bio/oyic~c~/ P.\dlicrtrl~. 29, 55A.
TPQ IN EATING DISORDERS Download full-text
Welch, G., Hall, A., & Renner, R. (1990). Patient subgroupings
Williamson, D. A.. Kelley, M. L., Davis, C. J., Ruggiero.
disorders: Controlled comparison
Yates, W. R., Sieleni, B., & Bowers, W. A. (1989). Clinical correlates
in anorexia nervosa using psychologically-based
Journal of Eating Disorder.?, 3, 31 l-322.
L., & Blouin, D. C. (1985). Psychopathology
obese, and normal subjects.
Journal of Consulting
53, I6 I- 166.
of personality disorder in bulimia nervosa.
Journal gf Eating Disorders. 8,413471.