Harm avoidance and self-directedness as essential features of
panic disorder patients
Cláudia Wachleski, Giovanni Abrahão Salum, Carolina Blaya, Letícia Kipper,
Angela Paludo, Ana Paula Salgado, Gisele Gus Manfro⁎
Anxiety Disorder Program, Hospital de Clínicas de Porto Alegre and Graduate Program in Medical Sciences,
Psychiatry, Universidade Federal do Rio Grande do Sul, Brazil
Objective: The purpose of the study is to compare the personality traits assessed with the Temperament and Character Inventory (TCI)
between patients with panic disorder (PD) and a control group in a Brazilian sample.
Methods: One hundred thirty-five patients with PD paired according to sex and age with 135 controls without any psychiatric disorders were
assessed with the Mini International Neuropsychiatric Interview (MINI), a structured interview. Temperament and character were assessed
with the TCI.
Results: Consistently, patients with PD presented higher scores on the harm avoidance (HA) temperament scale (23.20 ± 5.41 vs 15.21 ±
4.92; P b.001) and lower scores on the self-directedness (SD) (27.81 ± 7.25 vs 35.16 ± 5.47; P b.001) if compared to the control group and
has been associated independently from other TCI scales and confounders with PD. The multivariate logistic model containing HA and SD
explains 38.6% to 51.4% of the differences between PD and controls.
Conclusions: Harm avoidance could be a good candidate to be heritable because it appears to be a consistent finding across current literature
in anxious and depressed patients independent of their cultural context. Also, SD seems to be a key character characteristic of PD patients.
The dimensional assessment is an interesting alternative for understanding the relationship between the psychobiologic bases of temperament
and character and is highly related to the development of psychiatric syndromes.
© 2008 Elsevier Inc. All rights reserved.
The high prevalence of personality disorders, mainly
those included in the anxious cluster (cluster C), in patients
with panic disorder (PD) [1-4] suggests a common basis
for Axis I and II diagnoses that underlie these 2 constructs
in the current Diagnostic and Statistical Manual of Mental
Disorders (DSM) diagnosis system . Longitudinal
studies show that personality traits are generally presented
in a subclinical way before the onset of the anxiety disorder
and prospectively became personality disorders during
the symptomatic phase . Therefore, the dimensional
assessment may be an alternative for understanding the
relationship between the psychobiologic basis of tempera-
ment, character and development of Axis I and Axis II
These aspects raise a discussion about the need of
finding new definitions for psychiatric nosological entities
that go beyond the descriptive-phenomenological approach
of DSM-IV , which often does not include the
overlapping of diagnoses with similar etiologic factors .
According to this paradigm, new instruments that can assess
temperament characteristics have been investigated [8,9].
After this theoretical direction, dimensional constructs of
personality have been suggested with the purpose of finding
a common basis and a continuum that is able to join these
The Temperament and Character Inventory (TCI),
described by Cloninger , defines 4 dimensions of
temperament as follows: novelty seeking (NS), harm
Available online at www.sciencedirect.com
Comprehensive Psychiatry 49 (2008) 476–481
This research was supported by Fundaçäo de Incentivo à Pesquisa
(FIPE-HCPA), CAPES (CW and CB), and CNPq (GGM and GAS).
⁎Corresponding author.Ramiro Barcelos, 2350/400N.PortoAlegre-RS
90035-903, Brazil. Tel.: +55 51 21018294; fax: +55 51 21018294.
E-mail address: email@example.com (G.G. Manfro).
0010-440X/$ – see front matter © 2008 Elsevier Inc. All rights reserved.
avoidance (HA), reward dependence (RD), and persistence
(P) and 3 dimensions of character as follows: self-
directedness (SD), cooperativeness (C), and self-transcen-
dence (ST). According to Cloninger's theory, the tempera-
ment was hypothesized to involve neurobiological heritable
predisposition to primary emotions as fear, anger, and
empathy, arising automatic reactions in response to specific
environmental stimuli [8,9]. Character, in the other way,
includes individual differences in highly cognitive processes
as secondary motivations of a nonlinear interaction between
temperament, familiar environment, and individual life
The HA temperament has been systematically associated
with the anxiety disorders, and it presents a hereditary trend
of inhibited behaviors in relation to new stimuli, punishment,
and lack of reward [8,9,11]. Several studies that investigated
the association between the temperament and patients with
PD have confirmed higher levels of HA in these patients if
compared to the control group [12-17]. In spite of that, to our
knowledge, there were no studies investigating the Clonin-
ger's TCI in PD patients in South America; consequently, the
higher HA scores and other differences in TCI scales
between PD patients and controls were never evaluated in
South America countries.
In addition to the nosographic/nosological aspects invol-
ving the diagnostic issue, the TCI has been investigated in
studies that seek to define a genetic origin for the anxiety
disorders [18,19]. Furthermore, the construct represented by
the TCI has also been assessed as a predictive factor of
response to the pharmacological and psychotherapeutic
Because of the importance of investigating the tempera-
ment, the objective of this study is to compare the
temperament and character traits of patients with a diagnosis
of PD according to the DSM-IV and a control group in a
Brazilian sample. If replicated, an association between harm
avoidance (HA) and PD patients in another cultural context
addresses to TCI construct another argument to their
The patients included in the study were selected among
the individuals taking part in the Anxiety Disorders Program
(PROTAN) of the Psychiatry Service of Hospital de Clínicas
de Porto Alegre (HCPA), Porto Alegre, Brazil. These
patients had already participated in clinical trials and other
research projects of the same program. One hundred thirty-
five patients with PD and 135 controls without psychiatric
disorders were included.
Eighteen-year-old or older patients who met the diag-
nostic criteria for panic disorder according to the DSM-IV
criteria were included in this study. Patients with the
following characteristics were excluded: mental retardation,
dementia or other organic brain syndrome, drug abuse,
psychotic disorders, and disabling chronic disease. Patients
with comorbidities common to PD, such as generalized
anxiety disorder, major depression (MD), dysthymia, social
anxiety disorder, and obsessive-compulsive disorder, were
included in the study provided that the symptoms were not
clinically more relevant than the PD symptoms. The control
group comprised HCPA employees who did not meet the
criteria for Axis I psychiatric disorder.
Controls were selected from June 2006 to April 2007. The
control group was a convenience sample comprising HCPA
employees without psychiatric disorders. Among those
employees approaching 20 were excluded because they
met the criteria for a psychiatric disorder according to the
Mini International Neuropsychiatric Interview (MINI), and
5 were excluded because they were undergoing psychiatric
treatment. Approximately 30 employees (18%) refused to
take part in the study. The most common reasons for refusing
to participate were the length of the scales and the lack of
time to fill in the questionnaires adequately.
All participants signed the written consent (05-204/05-
539) that had been previously approved by the ethics
research committee of Hospital de Clínicas de Porto
2.2.1. Clinical Instruments
The diagnostic assessment of patients and controls was
MINI (Brazilian version 5.0.0-DSM IV)—a structured
interview that assesses in a standardized manner the main
Axis I psychiatric disorders according to the DSM-IV
criteria. The MINI was translated and adapted to Portuguese
The severity and intensity of the patients' symptoms were
assessed with the Clinical Global Impression (CGI), the
Panic Inventory (PI), and the Hamilton Anxiety Scale
(HAM-A). The CGI defines the severity of the symptoms
according to the frequency and intensity of panic attacks,
level of anticipatory anxiety, phobic avoidance, and impair-
ment of the quality of life and functioning . The Panic
Inventory classifies the panic attacks as spontaneous,
situational, complete, or incomplete and defines their
intensity, duration, frequency, as well as the intensity of
agoraphobia and anticipatory anxiety. The HAM-A provides
information on the level and severity of anxiety symptoms
(14 items) through values on the scale that range from
absence (zero) to maximum intensity of symptoms (4).
2.3. Cloninger's TCI
Personality was assessed with the Cloninger's TCI .
This instrument is a 240-item, true-false, self-report ques-
tionnaire that assesses 4 temperament dimensions: harm
avoidance (HA), novelty seeking (NS), reward dependence
(RD), and persistence (P). NS was defined as a pattern of
frequent exploratory activity and intense excitement in
477 C. Wachleski et al. / Comprehensive Psychiatry 49 (2008) 476–481
response to new stimuli. HA is described as a heritable
tendency to inhibit or stop behaviors in response to signals of
aversive stimuli to avoid punishment. RD is described as a
tendency to intensively respond to signals of reward to be
rewarded, and P is a tendency to maintain a specific response
despite intermittent reward.
Cloninger  added 3 new factors that represent the
character dimension of personality to the original model
. The character factors assess the individual differences
regarding self-concepts and perception of individual goals
and values. These factors are the following: self-directed-
ness (SD), cooperativeness (C), and self-transcendence
(ST). SD is defined as individual self-control, stable and
adapted behavior, and intentional strength of the individual
to assert and achieve his/her goals and intentions. Low SD
scores are associated with doubtful, reactive, and depen-
dent people, presenting low self-esteem, feelings of low
self-worth, and immaturity. C includes the identification
and the acceptance of individual differences regarding
other people. Cooperative individuals are described as
socially tolerant, empathic, and supportive, whereas
noncooperative individuals are described as socially
intolerant and inconsiderate of other people. ST is
described as acceptance, identification, or spiritual union
with nature, so that individuals conceive themselves as
integral parts of the universe. Fuentes et al  translated
and validated the TCI to Portuguese to study these 7
2.4. Statistical analysis
Data were presented as absolute frequency (%), mean ±
SD, and mean differences (95% confidence interval). The
normal distribution and the variance homogeneity were
assessed before any data analysis and was carried out with
Kolmogorov-Smirnov and Levene tests, respectively. The
comparison between the means of the TCI scales between
patients with PD and controls was performed with Student
t test for paired samples. The comparison between the
means of the TCI scores and the PD, PD + MD groups,
and controls was performed with 1-way analysis of
variance, using Bonferroni's correction in the post hoc
test. In addition, the TCI scales that achieve a P b .1 were
included in a multivariable logistic regression to identify
independent variables related to PD. All differences (with
P b .1) in sociodemographic data between cases and
controls that were also related with a P value of less than
.1 with TCI scales were included in the model to control
for confounding factors.
The statistical analysis was performed using the SPSS
(version 16). An α value of .05 and a 95% confidence
interval were used. Crossing the data of the 7 TCI scales with
cases and controls (7 comparisons) increases the chance of
finding random association. With the use of Bonferroni's
correction for the 7 comparisons, the P values of less than
.007 are safer regarding false findings if we are interpreting
bivariate analysis individually.
3.1. Sample characteristics
One hundred women (74.1%) and 35 men (25.9%) with
PD who were paired according to sex and age with 135
controls were included in the study. Mean age of the sample
was 38.1 ± 9.09.
The sample of patients has higher percentage of
individuals with incomplete high school studies if compared
to the controls (33.8% vs 5.9%; P b .001) and a higher
percentage of individuals who are married or living with a
partner (49.6% vs 29.7%; P b.001). Both variables were also
associated with TCI scales with P value less than 0.1 (data
The subjects of the patients group were outpatients of
HCPA, and their PD severity was moderate to severe (CGI =
4.7 ± 0.79), presenting high scores of anxiety symptoms
(HAM-A = 27.8 ± 8.29), agoraphobia (agoraphobia in the
PI = 7.7 ± 2.57), and anticipatory anxiety (anticipatory
anxiety in PI = 8.6 ± 1.73). Most patients also met the
diagnostic criteria for agoraphobia according to the MINI,
118 patients (87.4%). The most common comorbidities
(N10%) in the patients group were as follows: 50 (37.0%)
patients with generalized anxiety disorder; 38 (28.1%) with
major depression; 18 (13.3%) with dysthymia; and 23
(17.0%) with social anxiety disorder.
3.2. Temperament and Character Inventory between
patients and controls
Patients presented higher scores in the HA dimension,
which is an element of the temperament in the TCI, and also
higher scores in the ST dimension of character. In the other 2
character dimensions (SD and C), the patients had lower
scores than the controls. The results of the comparison
between the means of both groups are shown in Table 1.
Despite the moderate to strong correlation between 2 of
the prediction variables (HA and SD; r = −0.569), they could
be included in the same model of multivariable logistic
regression because there is a low probability of multi-
colinearity (an assumption for this statistical test) between
than (tolerance values are higher than 1 − r2). So the TCI
variables HA, P, SD, C, and ST entered in the multivariable
logistic model and were controlled by social status and
educational level to avoid confounding. Only HA and SD
remain independently associated with PD within TCI scales
and confounders (Table 2). HA and SD taken together in a
multivariable logistic regression model explain 38.6% of the
variance regarding the differences between controls and PD
patients according to Cox and Snell R Square and 51.4%
according to Nagelkerke R Square.
3.3. Influence of the comorbidity with MD in the TCI
The comorbidity with major depression (n = 38; 28.1%),
which is very frequent in PD, increases even more the
difference between patients and controls in the HA, SD, and
478C. Wachleski et al. / Comprehensive Psychiatry 49 (2008) 476–481
C dimensions of the TCI considering absolute values. The
means are also significantly different in the HA and SD
dimensions between patients with and without this comor-
bidity. In the other dimensions, there were no statistically
significant differences between the groups of patients with
and without the comorbid condition, and the comorbidity did
not increase the difference of controls. The results of this
analysis are shown in Table 3.
Cloninger, through his temperament model, suggested
that individuals with anxiety disorder might have higher
scores of HA and lower scores of NS and RD . Our study
is in agreement with most studies that assessed these
dimensional personality traits in patients with anxiety
disorders [12-17] and replicated the data on increased scores
of HA, without showing significant difference between the
NS and RD scores of patients and control groups.
In our study, in addition to the higher scores in HA, lower
scores in the character dimensions SD and C were found in
patients compared to the control group. Also, we found that
the ST dimension of character had higher scores in our
sample of patients if compared to the control group. The
character dimension includes individual differences in highly
cognitive processes with a crucial role in the integration and
functioning of the personality structure [9,25,26], so it is
expected that people with mental disorders show lower
scores in these scales. On the other hand, the ST dimension
describes a trait that characterizes imaginative, intuitive, and
unconventional people, providing a quantitative measure for
creativity or psychosis . In our sample, high STsuggests
magical ideation because SD was also low. This magical
ideation can be interpreted in PD as the catastrophic
cognitive thoughts of panic attacks, for example, fear of
death, fear of losing control, and others.
In spite of mean differences between patients and controls
in C and ST scales, these variables were not independently
associated with PD and, therefore, could be confounded by
demographic data as instruction level or civil status or
explained by the presence of the 2 independently associated
constructs—HA and SD. The multivariate logistic model
including HA and SD explains 38.6% to 51.4% of the
differences between PD and controls, so these 2 dimensions
will be more extensively discussed.
HA represents a heritable tendency to respond intensely
to aversive stimuli, with inhibition of behaviors to avoid
punishment, novelty, and frustrating nonreward, which
would hypothetically be mediated by the polygenic influence
on the serotoninergic function [8,11]. Also, the mechanism
of HA appears to be that it accounts for 30% of variance in
functional connectivity of the amygdala and the perigenual
cingulate, which are involved in perceptual processing of
fearful stimuli . People with high HA scores showed
pessimistic concerns about future problems, avoidant
behavior, fear of uncertainty, shyness with strangers, and
high fatigability  and could develop characteristics of
“anxious personality,” anxiety disorders, and depressive
disorders , as well as the comorbidity between anxiety and
mood disorders .
Individuals with lower SD are often reactive, dependent,
resourceless, and immature , and this scale is highly
associated to personality disorders . Personality dis-
orders and dysfunctional personality traits are very common
among PD patients [20,21,29-31]. In addition, HA has
already been associated with personality disorders .
Regarding this, maybe PD and personality disorders share
HA as their underlying temperament characteristic, interfer-
ing in the development of character (SD) and leading to the
structural psychiatric classification of these nosological
entities. The comorbidity between PD and MD can be
Multivariate logistic analysis investigating independent associations of
temperament in PD
VariablesB WaldP Odds ratio
(95% confidence interval)
1.25 (1.16 to 1.36)
0.91 (0.84 to 0.98)
1.02 (0.93 to 1.11)
1.06 (0.99 to 1.12)
Civil status and instruction level were also included in the model to control
for confounders because of selection of controls (P = .002 and P b.001).
Differences in temperament and character scales of TCI between patients and controls
(n = 135)
(n = 135)
Mean difference (95% confidence
interval of the difference)
19.04 ± 4.88
15.21 ± 4.92
15.37 ± 3.62
4.84 ± 1.77
19.80 ± 5.38
23.20 ± 5.41
15.41 ± 3.78
4.44 ± 1.71
−0.76 (−1.966 to 0.455)
−7.99 (−9.307 to −6.664)
−0.04 (−0.920 to 0.846)
0.39 (0.000 to 0.786)
35.16 ± 5.47
34.26 ± 4.16
12.61 ± 6.26
27.81 ± 7.25
31.68 ± 5.19
14.61 ± 6.20
7.35 (5.843 to 8.854)
2.58 (1.478 to 3.677)
−1.99 (−3.522 to −0.463)
Values are presented as mean ± SD. Statistics used was paired Student t test.
479C. Wachleski et al. / Comprehensive Psychiatry 49 (2008) 476–481
interpreted in the same way. Several studies found high
scores of HA in anxiety disorders [12-17] as well as in mood
disorders [15,16,32,33]. Therefore, the HA dimension of
temperament and SD dimension of character could favor the
development of these comorbidities  that usually increase
the clinical severity and worsen the prognosis of mental
disorders [2,3,34-36]. Moreover, Farmer et al  have
shown that HA, RD, NS, and SD have trait such as
characteristics that are related to the familiality of MD.
Marchesi et al  have demonstrated that PD patients
who remit after treatment had higher HA. On the other hand,
patients with lower scores in temperament and character
scales did not remit. After controlling for confounders, only
SD was independently associated to worse treatment
response. Considering this, Cloninger et al  suggest that
combined treatment (pharmacotherapy and psychotherapy)
should be indicated for patients with some characteristics to
achieve higher remission rates.
In our study, it is possible to assume that MD and HA
presented important colinearity between themselves because
they lost statistical significance when individually con-
trolled. This finding suggests that, maybe, a more avoidant
temperament is the real factor of worse prognosis, regardless
of the specific symptoms of mood disorder. However, the
possibility that HA also represents a measure of severity of
symptoms should be considered for future studies.
The present study has some limitations. The design is not
ideal since the selected controls are not representative of the
same population as are the patients. Additionally, the
controls have a common characteristic, that is, most of
them are health professionals; consequently, we cannot
exclude specific temperament and character characteristics
linked to the group they belong to. Also, our study is limited
to individuals of southern Brazil and, therefore, do not
represent the Brazilian population itself.
The replicated finding of higher rates of HA in PD
patients as compared to a control group in a Brazilian sample
addresses to the TCI construct a transcultural validation and,
therefore, contribute to the hypothesis that HA could be a
good candidate to be heritable because it remains increased
in PD patients independently of their cultural context, that is,
the environment. Also, SD seems to be a key character
characteristic of PD patients. In conclusion, the dimensional
assessment is an interesting alternative for understanding the
relationship between the psychobiologic basis of tempera-
ment, character, and development of psychiatric syndromes.
 Latas M, Starcevic V, Trajkovic G, Bogojevic G. Predictors of
comorbid personality disorders in patients with panic disorder with
agoraphobia. Compr Psychiatry 2000;41(1):28-34.
 Marchesi C, Cantoni A, Fonto S, Gianelli MR, Maggini C. The effect
of pharmacotherapy on personality disorders in panic disorder: a one
year naturalistic study. J Affect Disord 2005;89:189-94.
 Marchesi C, De Panfilis C, Cantoni A, Fonto S, Gianelli MR, Maggini
C. Personality disorders and response to medication treatment in panic
disorder: a 1-year naturalistic study. Prog Neuropsychopharmacol
 Albert U, Maina G, Bergesio C, Bogetto F. Axis I and II comorbidities
in subjects with and without nocturnal panic. Depress Anxiety 2006;23
 Diagnostic and Statistical Manual of Mental Disorders (DSM IV)4th
Ed. . Washington (DC): American Psychiatric Association; 1994.
 Brandes M, Bienvenu OJ. Personality and anxiety disorders. Curr
Psychiatry Rep 2006;8(4):263-9.
 Battaglia M, Przybeck TR, Bellodi L, Cloninger CR. Temperament
dimensions explain the comorbidity of psychiatric disorders. Compr
 Cloninger CR. A systematic method for clinical description and
classification of personality variants. A proposal. Arch Gen Psychiatry
 Cloninger CR, Svrakic DM, Przybeck TR. A psychobiological model
of temperament and character. Arch Gen Psychiatry 1993;50:975-90.
of personality development: fundamentals of a self-organizing
psychobiological complex. Dev Psychopathol 1996;8:247-72.
 Cloninger CR. A unified biosocial theory of personality and its role in
the development of anxiety states. Psychiatr Dev 1986;4(3):167-226.
 Saviotti FM, Grandi S, Savron G, Ermentini R, Bartolucci G, et al.
Characterological traits of recovered patients with panic disorder and
agoraphobia. J Affect Disord 1991;23(3):113-7.
 Ampollini P, Marchesi C, Signifredi R, Maggini C. Temperament and
personality features in panic disorder with or without comorbid mood
disorders. Acta Psychiatr Scand 1997;95(5):420-3.
 Battaglia M, Bertella S, Bajo S, Politi E, Bellodi L. An investigation of
the co-occurrence of panic and somatization disorders through
temperamental variables. Psychosom Med 1998;60(6):726-9.
Differences in TCI scales between patients with PD without MD, patients with PD comorbid with MD, and controls
(n = 135)
Patients 1-way analysis of variance
Without MD (n = 97)With MD (n = 38)Fdf(2;267)
19.04 ± 4.88
15.21 ± 4.92
15.37 ± 3.62
4.84 ± 1.77
35.16 ± 5.47
34.26 ± 4.16
12.61 ± 6.26
20.06 ± 5.30
22.35 ± 5.23a
15.44 ± 3.80
4.49 ± 1.73
29.07 ± 6.34a
32.06 ± 4.91a
14.48 ± 6.52
19.13 ± 5.60
25.37 ± 5.33b
15.32 ± 3.78
4.32 ± 1.69
24.58 ± 8.43b
30.71 ± 5.80a
14.92 ± 5.36
Values are presented as mean ± SD. Statistics used were 1-way analysis of variance (with Bonferroni's correction in post hoc test). Regarding ST, Bonferroni's
post hoc test did not found between which groups the related differences occur.
aDifferences related to controls.
bDifferences related to PD patients without MD.
480C. Wachleski et al. / Comprehensive Psychiatry 49 (2008) 476–481
 Ampollini P, Marchesi C, Signifredi R, Ghinaglia E, Scardovi F, et al.
Temperament and personality features in patients with major
depression, panic disorder and mixed conditions. J Affect Disord
 Kennedy BL, Schwab JJ, Hyde JA. Defense styles and personality
dimensions of research subjects with anxiety and depressive disorders.
Psychiatr Q 2001;72(3):251-62.
 Wiborg IM, Falkum E, Dahl AA, Gullberg C. Is harm avoidance an
essential feature of patients with panic disorder. Compr Psychiatry
 Hajduk A. A search for psychobiological determinants of anxiety
disorders. Ann Acad Med Stetin 2004;50(2):65-75.
 Draganic-Rajic S, Lecic-Tosevski D, Paunovic VR, Cvejic V, Svrakic
D. Panic disorder-psychobiological aspects of personality dimensions.
Srp Arh Celok Lek 2005;133(3-4):129-33.
temperament and character on response to selective serotonin reuptake
inhibitors in panic disorder. Acta Psychiatr Scand 2006;114(3):203-10.
 Prasko J, Houbová P, Novák T, Záleský R, Espa-Cervená K, Pasková
B, Vyskocilová J. Influence of personality disorder on the treatment of
panic disorder—comparison study. Neuro Endocrinol Lett 2005;26(6):
 Amorim P. Mini International Neuropsychiatric Interview (MINI):
validação de entrevista breve para diagnóstico de transtornos mentais.
Rev Bras Psiquiatr 2000;22(3):106-15.
 Guy W. CGI—Clinical Global Impressions. Assessment manual for
psychopharmacology. Rockville, MD: National Institute of Mental
Health; 1976. p. 217-22.
 Fuentes D, Tavares H, Camargo CHP, Gorenstein C. Inventário de
Temperamento e de Caráter de Cloninger—validação da versão em
Português. Escalas de Avaliação Clínica em Psiquiatria e Psicofarma-
cologia; 2000. p. 363-9.
 Svrakic DM, Whitehead C, Przybeck TR, Cloninger CR. Differential
diagnosis of personality disorders by the seven-factor model of
temperament and character. Arch Gen Psychiatry 1993;50:991-9.
 Svrakic DM, Draganic S, Hill K, Bayon C, Przybeck TR, Cloninger
CR. Temperament, character, and personality disorders: etiologic,
diagnostic, treatment issues. Acta Psychiatr Scand 2002;106(3):
 Bayon C, Hill K, Svrakic DM, Przybeck TR, Cloninger CR.
Dimensional assessment of personality in an out-patient sample:
relations of the systems of Millon and Cloninger. J Psychiatr Res 1996;
 PezawasL,Meyer-LindenbergA,Drabant EM,VerchinskiBA, Munoz
KE, Kolachana BS, et al. 5-HTTLPR polymorphism impacts human
cingulate-amygdala interactions: a genetic susceptibility mechanism
for depression. Nat Neurosci 2005;8(6):828-34.
 Maser JD, Cloninger CR. Comorbidity of mood and anxiety disorders.
Washington (DC): American Psychiatric Press; 1990.
 Ozkan M, Altindag A. Comorbid personality disorders in subjects with
panic disorder: do personality disorders increase clinical severity.
Compr Psychiatry 2005;46(1):20-6.
 Wachleski C, Blaya C, Salum GA, Vargas V, Leistner-Segal S, Manfro
GG. Lack of association between the serotonin transporter promoter
polymorphism (5-HTTLPR) and personality traits in asymptomatic
patients with panic disorder. Neurosci Lett 2008;431(2):173-8.
 Hansenne M, Reggers J, Pinto E, Kjiri K, Ajamier A, Ansseau M.
Temperament and character inventory (TCI) and depression. J
Psychiatr Res 1999;33(1):31-6.
 Ongur D, Farabaugh A, Iosifescu DV, Perlis R, Fava M. Tridimen-
sional personality questionnaire factors in major depressive disorder:
relationship to anxiety disorder comorbidity and age of onset.
Psychother Psychosom 2005;74(3):173-8.
 Prasko J, Houbová P, Novák T, Záleský R, Espa-Cervená K, Pasková
B, et al. Influence of personality disorder on the treatment of panic
disorder—comparison study. Neuro Endocrinol Lett 2005;26(6):
 Rosenbaum JF, Pollack MH, Pollock RA. Clinical issues in the long-
term treatment of panic disorder. J Clin Psychiatry 1996;57(10):44-50.
 Lecrubier Y. The impact of comorbidity on the treatment of panic
disorder. J Clin Psychiatry 1998;5(8):11-6.
 Farmer A, Mahmood A, Redman K, Harris T, Sadler S, McGuffin P. A
sib-pair study of the Temperament and Character Inventory scales in
major depression. Arch Gen Psychiatry 2003;60(5):490-6.
481C. Wachleski et al. / Comprehensive Psychiatry 49 (2008) 476–481