Journal of Affective Disorders 108 (2008) 207 216 www.elsevier.

com/locate/jad

Research report

Dissociative experiences differentiate bipolar-II from unipolar depressed patients: The mediating role of cyclothymia and the Type A behaviour speed and impatience subscale
Ketil J. Oedegaard a,b,c,, Dag Neckelmann b , Franco Benazzi d , Vigdis E.G. Syrstad b , Hagop S. Akiskal c , Ole Bernt Fasmer a,b
a

Department of Clinical Medicine, Section for Psychiatry, Faculty of Medicine, University of, Bergen, Bergen, Norway b Division of Psychiatry, Haukeland University Hospital, Bergen, Norway c International Mood Center, University of California at La Jolla, San Diego, USA d Hecker Psychiatry Research Center at Forli (IT), and Department of Psychiatry, National Health Service, Forli, Italy Received 19 October 2007; accepted 24 October 2007 Available online 20 February 2008

Abstract Background: Dissociative symptoms are often seen in patients with mood disorders, but there is little information on possible association with subgroups and temperamental features of these disorders. Methods: The Dissociative Experience Scale was administered to 85 patients with a DSM-IV Major Depressive Disorder (MDD) or Bipolar-II Disorder (BP-II). Both broad-spectrum dissociation (DES total score) and clearly pathological forms of dissociation (DES-Taxon) were assessed. Temperament was assessed using Akiskal and Mallya`s criteria of Affective Temperaments and the Jenkins Activity Survey (JAS) for Type A Behaviour. Results: Sixty-five patients gave valid answers to DES. The mean DES and DES-T scores were higher in BP-II (16.8 and 12.7 respectively) compared to MDD (9.0 and 5.7); DES odds ratio (OR) = 1.58 (95% CI 1.152.18) and DES-T OR = 1.60 (95% CI 1.142.25) using univariate logistic regression analyses. There was no significant difference in DES score in patients with (n = 30) and without an affective temperament (n = 35): mean (95% CI), 13.5 vs. 10.5 ( 7.8 to 1.9), p = 0.224. However the subgroup with a cyclothymic temperament (n = 18) had higher DES scores (mean (95% CI): 17.8 vs. 9.7 (2.913.3), p = 0.003), compared to patients without such a temperament. There was no significant difference in DES scores for patients with (n = 35) or without (n = 28) a Type A behaviour pattern (JAS N 0): mean (95% CI) 12. 7 vs. 10.9 ( 6.8 to 3.3), p = 0.491), but a positive JAS factor S score (speed and impatience subscale) was associated with significantly higher DES scores than a negative S-score: mean (95% CI) 14.9 vs. 9.0 (1.110.7), p = 0.017), and this was still significant (p = 0.005) using multiple linear regression of DES scores vs. the JAS subscale scores. DES-T scores were significantly higher in patients with OCD (n = 9) (mean (95% CI) 18.4 vs. 6.6 (6.017.7), p b 0.001); eating disorder (n = 13) (14.0 vs. 6.8 (1.812.6), p = 0.009), psychotic symptoms during depressions (n = 9) (16.6 vs. 6.9 (3.715.8), p = 0.002), and in those with a history of suicide attempt (n = 28) (11.9 vs. 5.4 (2.210.8), p = 0.003), but only OCD was an independent predictor after multiple linear regression of DES-T scores vs. all co-morbid disorders (p = 0.043). Limitations: The major limitation of the present study is a non-blind evaluation of affective diagnosis and temperaments, and assessment in a non-remission clinical status.

Corresponding author. International Mood Center, University of California at La Jolla, San Diego, USA. E-mail addresses: koedegaa@ucsd.edu, keti@haukeland.no (K.J. Oedegaard). 0165-0327/$ - see front matter 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.jad.2007.10.018

208

K.J. Oedegaard et al. / Journal of Affective Disorders 108 (2008) 207216

Conclusions: Dissociative symptoms measured with the Dissociative Experience Scale are associated with bipolar features, using formal DSM-IV criteria, cyclothymic temperament and the speed and impatience subscale of the JAS. 2007 Elsevier B.V. All rights reserved.
Keywords: Dissociation; Bipolar disorder; Affective temperament; Type A behaviour

1. Introduction Dissociative symptoms are seen in many psychiatric disorders, also in mood disorders. Dissociation is defined as a disruption in the usually integrated functions of consciousness, memory, identity, or perception of environment (American Psychiatric Association, 1994). The Dissociative Experience Scale (DES) is a 28-item, self-administered inventory to measure the frequency of dissociative experiences (Bernstein and Putnam, 1986; Carlson and Putnam, 1993). The subjects circle the percentage of time (given in 10% increments ranging from 0 to 100 on the response scale line) they have the experience described. The total DES score is the mean of all item scores. The scale was conceptualized as a trait measure and it inquires about the frequency of dissociative experiences in the daily life of the subjects. It is developed for use with adults (N 18 years), and the items are worded to be comprehensible to the widest possible range of individuals and seeks to avoid any social undesirability of the experiences. The DES was designed to be useful in determining the contribution of dissociation to various psychiatric disorders and as a screening instrument for dissociative disorders (Draijer and Boon, 1993). It was not intended as a diagnostic instrument. It has been used to measure dissociation in non-clinical populations, and these subjects typically score in a narrow range at the low end of the scale (mean values from 4 to 8 in different studies). The DES has been administered to patients with mood disorders (Carlson and Putnam, 1993) and bipolar disorder ( Nijenhuis et al., 1999), but to the best of our knowledge there has not previously been presented data on dissociative symptoms in subgroups of patients with MDD and BP-II disorder. These disorders are heterogeneous and it would be interesting both from a theoretical and a practical point of view to investigate possible differences in the occurrence of such symptoms. The present study is part of a larger study on the phenomenology and associated characteristics of patients with major affective disorders (Fasmer, 2001; Fasmer and Oedegaard, 2001; Oedegaard and Fasmer, 2005). In

addition to subgrouping the patients according to DSMIV criteria, we have systematically evaluated two other dimensions, affective temperaments (Akiskal and Mallya, 1987; Akiskal and Akiskal, 1992) and type A behaviour. Type A behavioural pattern (TABP) is a construct that emerged from research trying to identify behavioural factors associated with coronary heart disease, and the most critical aspects of TABP were considered to be excesses of aggression, hurry, restlessness and competitiveness (Friedman and Rosenman, 1974). We have used the Jenkins Activity Survey (JAS), a self report questionnaire, to evaluate symptoms and behaviour indicative of TABP (Jenkins et al., 1979). Several patterns in TABP are similar to traits often associated with bipolar disorders (Barrick, 1999). We have in a previous report shown an association between the speed and impatience factor of JAS and bipolar II disorder (Oedegaard et al., 2006). The purpose of the present study has been to see if there is a specific connection between bipolar disorder or bipolar temperamental features and dissociative symptoms. 2. Methods 2.1. Characterizing the sample The study group (n = 85) comprised 47 psychiatric inpatients at the University Hospital in Bergen and 38 outpatients under specialist treatment for affective disorder. The inpatients were consecutively admitted patients to a 12 bed open psychiatric ward. The remaining patients were from the day care unit or from psychiatric specialist praxis. Patients were included if they met DSM-IV criteria for major depressive disorder (MDD) or a bipolar-II disorder (BP-II), which was not secondary to an organic or substance abuse disorder, were between 18 and 65 years old, and gave informed consent to participate. Patients that did not speak Norwegian with sufficient quality to be interviewed without interpreter were excluded. The patients that were psychotic at intake were not psychotic when interviewed. The local ethics committee has approved the study protocol. We used a semi-structured interview based on DSM-IV criteria (American Psychiatric Association, 1994; Fasmer, 2001) for affective disorders (major depressive disorder,

K.J. Oedegaard et al. / Journal of Affective Disorders 108 (2008) 207216

209

bipolar I and II disorders), anxiety disorders (panic disorder, agoraphobia, specific phobia, social phobia, generalized anxiety disorder, obsessive compulsive disorder), eating disorders (anorexia nervosa and bulimia nervosa) and abuse of alcohol and drugs. In addition the following information was recorded: the number of depressive episodes, the presence of psychotic symptoms during the present or previous depressive episodes, and a history of suicide attempt during the current affective episode or previously. The criteria of the Headache Classification Committee of the International Headache Society (1988) were used to establish the diagnosis of migraine. In addition to migraine with and without aura the occurrence of migraine auras without headache was specifically recorded. All patients were interviewed by the first author (KJO), but 10 patients were interviewed in the company of the last author (OBF) in order to establish agreement concerning the diagnostic approach. 2.2. Instruments 2.2.1. Affective temperaments Affective temperaments were diagnosed according to the criteria of Akiskal and Mallya (1987) and Akiskal and Akiskal (1992), which are psychometrically validated (Placidi et al., 1998; Akiskal et al., 1998). Criteria for the cyclothymic temperament requires at least three of five attributes of each of the following two sets, with an indeterminate early onset (b 21 years). First group: 1. Hypersomnia versus decreased need for sleep. 2. Introverted self-absorption versus uninhibited peopleseeking. 3. Taciturn versus talkative. 4. Unexplained tearfulness versus buoyant jocularity. 5. Psychomotor inertia versus restless pursuit of activities. Second group: 1. Lethargy and somatic discomfort versus eutonia. 2. Dulling of senses versus keen perceptions. 3. Slowwitted versus sharpened thinking. 4. Shaky self-esteem alternating between low self-confidence and overconfidence. 5. Pessimistic brooding versus optimism and carefree attitudes. The hyperthymic temperament requires at least five of the following characteristics, with an indeterminate early onset (b 21 years): 1. Irritable, cheerful, overoptimistic, or exuberant. 2. Naive, overconfident, selfassured, boastful, bombastic, or grandiose. 3. Vigorous, full of plans, improvident, and rushing off with restless impulse. 4. Overtalkative. 5. Warm, people-seeking, or extroverted. 6. Overinvolved and meddlesome. 7. Uninhibited, stimulus-seeking, or promiscuous.

The irritable temperament requires at least five of the following characteristics, with an indeterminate early onset (b 21 years): 1. Habitually moody, irritable and choleric, with infrequent euthymia. 2. Tendency to brood. 3. Hypercritical and complaining. 4. Illhumoured joking. 5. Obtrusiveness. 6. Dysphoric restlessness. 7. Impulsive. The depressive temperament requires at least five of the following characteristics, with an indeterminate early onset (b 21 years): 1. Gloomy, pessimistic, humourless, or incapable of fun. 2. Quiet, passive, and indecisive. 3. Sceptical, hypercritical, or complaining. 4. Brooding and given to worry. 5. Conscientious or selfdisciplining. 6. Self-critical, self-reproaching, and selfderogatory. 7. Preoccupied with inadequacy, failure, and negative events to the point of morbid enjoyment of ones failures. 2.2.2. Dissociative Experience Scale (DES) The DES has been translated into Norwegian in 1992 by Be, Haslerud and Knudsen. The Norwegian version has then been translated back into English and has been approved by Eve B. Carlson (personal communication Helge Knudsen 2006). The Norwegian version has been administered to patients with different psychiatric disorders, with results comparable to that found in other studies (personal communication, Helge Knudsen 2006). For each patient we have also calculated the mean of the 8 items comprising the DES taxon (Waller et al., 1996; Waller and Ross, 1997), to obtain a measure of pathological dissociation, and we have calculated the mean across all patients for each individual item. In the original paper by Waller and Ross (1997) pathological dissociation was defined by a special scoring program, but we have, in accordance with another report ( Simeon et al., 1998), simply used the mean value. The 8 items of the DES taxon are: Item 3: Some people have the experience of finding themselves in a place and having no idea how they got there. Item 5: Some people have the experience of finding new things among their belongings that they do not remember buying. Item 7: Some people sometimes have the experience of feeling as though they are standing next to themselves or watching themselves do something and they actually see themselves as if they were looking at another person. Item 8: Some people are told that they sometimes do not recognize friends or family members. Item 12: Some people have the experience of feeling that other people, objects, and the world around them are not real. Item 13: Some people have the experience of feeling that their body does not seem to belong to them. Item 22: Some people find that in one situation they may act so

210

K.J. Oedegaard et al. / Journal of Affective Disorders 108 (2008) 207216

differently compared with another situation that they feel almost as if they were two different people. Item 27: Some people sometimes find that they hear voices inside their head that tell them to do things or comment on things that they are doing. 2.2.3. Jenkins Activity Survey (JAS, Form C) TABP was assessed by the Jenkins Activity survey (JAS, Form C). This is a self-report multiple-choice questionnaire of 52-items designed to measure TABP, defined by Jenkins as an overt behaviour syndrome or style of living characterised by extreme competitiveness, striving for achievement, aggressiveness, impatience, haste, restlessness and feelings of being challenged by responsibility and being under the pressure of time (Jenkins et al., 1967; Jenkins et al., 1979). Type B behaviour, are those who exhibit the opposite of type A behaviour; a relaxed, unhurried, mellow, satisfied style of living. In addition, the JAS provides separate factor scores for three components of TABP: 1. Speed and Impatience (factor S). This factor deals with the time urgency revealed in the style of behaviour of the type A person. Those scoring high on this factor tend to eat rapidly, become impatient with the conversation of others, hurry other people along, have strong tempers, and become irritated easily. 2. Job Involvement (factor J). This factor expresses degree of dedication to occupational activity. Typically, persons scoring high on this factor report having a challenging, high-pressure job. They work overtime and confront important deadlines. They prefer promotion to a pay raise, but usually have received both in the last few years. 3. Hard-Driving and Competitive (factor H). This factor involves perceptions of oneself as being harddriving, conscientious, responsible, serious, competitive, and putting forth more effort than other people. This series of traits suggests highly socialized but intense drives. At the end of the interview the JAS was given to the patients with the appropriate instructions concerning the responding process. Patients were asked to return the test to the nursing staff as soon as possible. All forms were scored by hand scoring, using the method described in the JAS Manual. A Norwegian translation of the JAS was used (Espenes and Opdahl, 1999). 2.3. Data analyses The means and frequencies of general characteristics and dissociative symptoms (DES scores) were estimated

for patients with MDD and BP-II disorder. Using univariate logistic regression analyses we calculated the odds ratios (ORs) with 95% confidence intervals (CIs) for the difference between MDD and BP-II in all items. Testing if the association between DES and BP-II was confounded by cyclothymic temperament (associated with DES), was done using multivariable logistic regression controlled for cyclothymic temperament and age. The means for DES and DES-T were calculated and compared (by the t-test) for all patients with and without different affective temperaments, and for patients with and without a cyclothymic temperament. This was done in the MDD and BP-II groups separately. We also estimated the mean DES and DES-T scores for different co-morbid disorders and JAS factors. T-test was used to
Table 1 The study sample: Compared clinical characteristics of MDD and BPII patients Variables: mean (SD), MDD % (n = 41) Gender (male) Age Married or co-habiting Working or studying Inpatients Affective temperament (any) Cyclothymic temp. Hyperthymic temp. Irritable temp. Depressive temp. Number of depressive episodes Psychotic symptoms during depressions Suicide attempt (lifetime) Number of anxiety disorders Panic disorder Drug abuse ADHD OCD Eating disorder Migraine Type A (JAS score) Speed and impatience (JAS-S score) Hard driving and Competitive (JAS-H score Job Involvement (JAS-J score) 26.8 36.2 (9.9) 56.0 24.3 43.9 34.1 BP-II (n = 24) 29.2 30.8 (8.3) 37.5 12.5 50.0 66.7 OR 1.12 0.94 0.47 0.44 1.27 3.86 7.20 0.85 3.63 0.25 1.05 95% CI 0.373.44 0.880.99 0.171.32 0.111.80 0.473.51 1.3311.20 2.1024.66 0.079.88 0.3142.40 0.032.25 0.941.17

12.2 50.0 4.9 4.2 2.4 8.3 14.6 4.2 3.2 (4.7) 4.2 (4.4) 14.6 26.8 1.6 (1.3) 34.1 29.3 9.8 12.2 12.2 61.0 2.3 (7.4) 2.5 (8.9) 20.8 70.8

1.41 0.434.66 6.60 2.1620.28

2.8 (1.2) 2.02 1.293.15 58.3 41.7 16.6 33.3 33.3 75.0 3.3 (7.3) 3.0 (9.1) 2.70 1.61 1.61 2.56 3.60 1.92 1.11 1.07 0.967.61 0.604.33 0.436.04 0.847.82 1.0212.73 0.635.87 1.031.21 1.011.14

6.7 (6.6) 5.5 (6.7) 1.03 0.951.11 5.4 (7.1) 6.5 (7.2) 0.97 0.911.05

OR = odds ratio, 95% CI = 95% Confidence Intervals. = p b 0.05. = p b 0.01.

K.J. Oedegaard et al. / Journal of Affective Disorders 108 (2008) 207216 Table 2 DES and DES-T scores in patients with MDD and BP-II Variables: Mean (SD) DES (Total score all items) DES-T (Total score taxon items) DES-T item 3 DES-T item 5 DES-T item 7 DES-T item 8 DES-T item 12 DES-T item 13 DES-T item 22 DES-T item 23 MDD (n = 41) 9.0 5.7 5.4 3.7 6.8 1.0 11.8 6.0 9.8 1.5 (8.4) (7.3) (13.1) (7.7) (13.9) (3.0) (17.5) (17.2) (16.2) (5.3) BPII (n = 24) 16.8 (10.1) 12.7 (10.2) 8.8 (12.3) 9.2 (16.9) 10.9 (14.4) 2.6 (6.9) 19.6 (26.6) 9.6 (13.7) 33.0 (30.4) 8.7 (20.7) OR 1.58 1.60 1.02 1.04 1.02 1.08 1.02 1.01 1.05 1.05 95% CI 1.152.18 1.142.25 0.981.06 0.991.09 0.981.06 0.951.21 0.991.04 0.981.05 1.021.08 0.981.12 OR# 1.43 1.51

211

95% CI# 1.012.03 1.082.12

OR = odds ratio, 95% CI = 95% Confidence Intervals. OR#/CI# = odds ratio and Confidence Intervals controlled for cyclothymic temperament and age. = p b 0.05. = p b 0.01.

calculate differences between groups and multiple linear regression analyses were used to determine independent predictors. A significance level of alpha 0.05 was chosen and all p values were two-sided. SPSS version 14.0 and STATA 9.2 were used for the statistical analyses. 3. Results The characteristics of the 65 patients with valid DES scores are shown in Table 1, comparing the MDD (n = 41) and the BP-II (n = 24) sample. The proportion of
Table 3 DES and DES -T scores in different affective temperaments Variables: Mean (SD) Cyclothymic temperament (n = 17) No cyclothymic temperament (n = 48) Hyperthymic temperament (n = 3) No hyperthymic temperament (n = 62) Irritable temperament (n = 3) No irritable temperament (n = 62) Depressive temperament (n = 7)) No depressive temperament (n = 58) Any affective temperament (n = 30) No affective temperament (n = 35) DES 17.8 (11.1) 9.7 (8.4) 3.7 (2.3) 12.3 (9.8) 10.9 (9.7) 11.9 (9.4) 8.1 (6.3) 12.3 (10.7) 0.287 13.5 (10.6) 10.5 (9.0) 0.224 0.140 0.003 P DES-T 12.2 (9.3) 6.9 (8.7) 0.3 (0.6) 8.6 (9.1) 0.123 0.038 P

DES non-responders (n = 20) was comparable in MDD (n = 12) and BP-II (n = 8) patients (data not shown). Among the responders no significant difference was found regarding the socio-demographic variables, although the BP-II patients were (mean 30.8 years) younger than the MDD patients (mean, 36.2 years). The BP-II patients had a significantly higher frequency of affective temperaments (particularly cyclothymic temperament), suicide attempts, anxiety disorders, eating disorders and Type A behaviour pattern (particularly factor S) than patients with MDD. In Table 2 are displayed the ORs for the comparison of DES, DES-T and the 8 DES-T items in BP-II and MDD patients. Both DES (OR = 1.58, 95% CI 1.15 2.18, p b 0.01) and DES-T scores (OR = 1.60, 95% CI 1.142.25, p b 0.01) were significantly higher in BP-II patients with only minor reductions in ORs when correcting for age and cyclothymic temperament. Only the DES-T item 22 was significantly raised in BP-II versus MDD patients.

0.869

8.6 (13.0) 8.2 (9.0) 0.965 5.0 (4.0) 8.6 (9.4) 8.9 (9.0) 7.7 (9.2) 0.602 0.322

Table 4 DES and DES-T scores in MDD and BP-II with and without cyclothymic temperament Variables: Mean (SD) DES P DES-T P

MDD Cyclothymic temp. 16.4 (16.0) 0.033 7.6 (9.4) 0.529 (n = 5) No cyclothymic 7.9 (6.6) 5.4 (7.0) temp. (n = 35) BP-II Cyclothymic temp. 18.4 (9.2) 0.441 14.0 (8.0) 0.522 (n = 12) No Cyclothymic 15.1 (11.1) 11.3 (11.6) temp. (n = 12) P = P values for the comparison of DES and DES-T scores in MDD and BP-II patients with and without cyclothymic temperament (t-test).

P = P values for the comparison of DES and DES-T scores in all patients with and without the temperament type (t-test).

212

K.J. Oedegaard et al. / Journal of Affective Disorders 108 (2008) 207216

Table 5 DES and DES-T scores in different co-morbid disorders Variables: Mean (SD) Eating disorder (n = 13) No eating disorder (n = 52) Panic disorder (n = 28) No panic disorder (n = 37) OCD (n = 9) No OCD (n = 56) Psychotic depression (n = 9) Not psychotic depression (n = 56) Suicide attempt (n = 28) No suicide attempt (n = 37) Drug abuse (n = 20) No drug abuse (n = 45) DES 17.3 (12.5) 10.5 (8.6) 11.9 (7.3) 11.8 (11.4) 22.7 (10.7) 10.1 (8.5) 18.3 (13.7) 10.8 (8.7) 15.3 (11.1) 9.3 (7.8) 14.7 (9.8) 10.6 (9.6) P 0.023 0.976 b 0.001 0.033 0.012 0.117 P# 0.279 0.649 0.084 0.518 0.138 0.306 DES-T 14.0 (9.9) 6.8 (8.3) 8.0 (7.5) 8.3 (10.2) 18.4 (10.6) 6.6 (7.7) 16.6 (14.1) 6.9 (7.3) 11.9 (10.7) 5.4 (6.4) 11.5 (10.3) 6.8 (8.1) P 0.009 0.905 b 0.001 0.002 0.003 0.052 P# 0.215 0.570 0.043 0.106 0.072 0.201

P = P values for the comparison of DES and DES-T scores in patients with and without the co-morbid disorder (t-test). P# = P values for the multiple linear regression of DES and DES-T scores vs. co-morbid disorders.

Table 3 compares the DES and DES-Tscores in patients with and without different affective temperaments. Cyclothymic temperament was significantly associated with both high DES and DES-T scores, whereas the other affective temperaments demonstrated no such tendency. In

Table 6 DES and DES-T scores in the JAS subscales Variables: Mean (SD) JAS Type A score N 0 (n = 35) JAS Type A score b 0 (n = 28) JAS-factor S score N 0 (n = 31) JAS-factor S score b 0 (n = 32) JAS-factor H score N 0 (n = 14) JAS-factor H score b 0 (n = 49) JAS-factor J score N 0 (n = 12) JAS-factor J score b 0 (n = 51) DES 12.7 (9.5) P P# DES-T 9.2 (9.3) P P#

10.9 (10.5) 0.491 0.350

7.1 (9.0)

0.374 0.952

14.9 (11.8)

10.6 (10.3)

9.0 (6.6)

0.017 0.005

6.0 (7.3)

0.044 0.089

11.1 (9.4)

7.6 (10.6)

Table 4 the DES and DES-T scores for patients with and without a cyclothymic temperament is presented for the MDD and BP-II groups separately, showing that MDD patients with a cyclothymic temperament have a significantly higher DES score than MDD's without such a temperament. It is also noteworthy that these MDD patients have DES scores equal to the BP-II patients. Table 5 shows the DES and DES-T scores in relation to different co-morbid disorders and symptoms. Although DES and DES-T scores were significantly increased in patients with eating disorders, OCD, psychotic depressions and suicidal behaviour, only OCD in DES-T (p = 0.043) remained a significant predictor after multiple linear regression of DES and DES-T scores vs. all co-morbid disorders/symptoms. DES and DES-T scores in relation to the Jenkins Activity Survey (JAS) are displayed in Table 6. Significantly increased DES and DES-T scores were found in patients with a positive JAS factor S score (the speed and impatience subscale of the type A construct), and this relation was still significant for JAS factor-S in DES (p = 0.044) after multiple linear regression of DES and DES-T scores vs. all the JAS subscale scores. 4. Discussion 4.1. Substantive findings

12.1 (10.1) 0.744 0.841

8.4 (8.8)

0.755 0.757

9.7 (9.6)

6.9 (11.4)

12.4 (9.9)

0.389 0.532

8.6 (8.6)

0.572 0.846

P = P values for the comparison of DES and DES-T scores in patients with and without positive JAS subscale scores (t-test). P# = P values for the multiple linear regression of DES and DES-T scores vs. the JAS subscale scores.

The main finding from this study is a higher prevalence of dissociative symptoms in patients with BP-II disorder compared to patients with MDD. Furthermore, we have found that the level of dissociative symptoms is related to temperamental features; as demonstrated for patients with a cyclothymic temperament and for patients with a positive JAS speed and impatience subscale score.

K.J. Oedegaard et al. / Journal of Affective Disorders 108 (2008) 207216

213

Interestingly, MDD patients with a cyclothymic temperament demonstrated significantly higher DES scores than MDD's without such a temperament, and their scores were equivalent to those found in BP-II patients. This is in line with substantial evidence supporting the division of patients with cyclothymic temperament from the unipolar depressive, and into the bipolar group (Akiskal and Akiskal, 1992; Akiskal, 1996). We have previously reported (Oedegaard et al., 2006) that TABP is associated with having a cyclothymic temperament, whereas depressive temperaments were found in persons with type B behaviour. The difference in JAS score between unipolar and bipolar patients was caused by a divergence in the factor S scale. This factor also explained the positive association between cyclothymic temperament and high JAS scores. This is the factor in the JAS that is thought to detect the most stable temperamental trait (strongest genetic factor), and most unlikely to be influenced by environmental factors or a change of life style (Jenkins et al., 1971). Likewise, cyclothymia is well known to be a genetic trait (Evans et al., 2005). All these findings thus point to an association between dissociative symptoms and bipolarity, assessed both by strict clinical criteria (DSM-IV) and by temperamental features. 4.2. Clinical implications Overall the use of DES-T scores (Waller et al., 1996; Modestin and Erni, 2004) did not give additional information compared to DES, but makes it clear that the symptoms in our sample are not only normal dissociative phenomena, but indeed represent pathological symptoms. BP-II disorder and related bipolar spectrum disorders are characterized by a number of both psychiatric and somatic co-morbid conditions (Vieta et al., 2000). Silberman et al. (1985) described a number of transient symptoms, including derealization and amnesia, that occurred more often in patients with affective disorders (mainly bipolar) and complex partial epilepsy compared to controls; interestingly, these psychosensory symptoms were higher among lithium responders! We have previously found that migraine is associated with bipolar II disorder and we have also argued that migraine can be used as a marker for bipolarity in patients with DSM-IV defined unipolar depressive disorders. In the present study we did not find any specific connection between migraine and dissociative symptoms. Similarly we have not found any association between migraine and type A behaviour (Oedegaard et al., 2006). It thus seems that migraine on the one hand and dissociative symptoms and type A behaviour on the other may be associated with different aspects of bipolarity.

Scores on the DES in the present study apparently are in the same range as that found in other studies on affective disorders. Carlson and Putnam (1993) reported on two studies of patients with affective disorders with mean values of DES 12.7 and 6.0. Knudsen and co-workers (personal communication Helge Knudsen 2006) found in a study using the Norwegian version of DES a mean score of 11 among 27 patients with affective disorders. In a study of psychiatric inpatients mean DES was 13.7 ( Modestin et al., 1996). Nijenhuis et al. (1999) found that in a group of 23 bipolar patients the mean DES score was 11.1, but did not specify type of bipolar disorder. When comparing scores on the 8 individual symptoms comprising the DES taxon, BP-II patients score higher than MDD patients on all items, but the most marked difference is on item 22, that describes a feeling of being two different people in different situations, and this was the only independent predicting item. Since BP-II patients experience more marked mood shifts that span a wider range of behavioural alterations than patients with unipolar depressions, it is not surprising that it is just this item that best separates these groups. This finding may also lend support to the hypothesis that serious dissociative disorders (dissociative identity disorder) may lie on a continuum with bipolar disorders (Savitz et al, 2004). The difference we have found between BP-II and MDD patients support the findings of Akiskal et al. (1995) that depersonalization is one of the factors that characterize those unipolar patients at baseline who convert to a BP II pattern during prospective follow-up. We are not aware of other studies that have looked specifically at bipolar patients in contrast to patients with unipolar depressions, and we have not seen other studies on affective temperaments and DES. However, Grabe et al. (1999) found that in a sample of inpatients with different diagnoses and some non-clinical subjects that none of the temperament dimensions from Cloningers Temperament and Character Inventory predicted scores on the DES. They therefore concluded that genetic factors probably play a minor role in dissociative phenomena. On the other hand Maremmani et al. (2005) have reported that, in a non-clinical sample, there is a relation between scores on Cloningers Tridimensional Personality Questionnaire and affective temperaments. Novelty seeking was related to hyperthymic and cyclothymic temperament and there was an association between cyclothymic temperament and harm avoidance. We would therefore have expected an association between DES scores and some of Cloningers temperamental dimensions. This has indeed been demonstrated for patients with depersonalization disorder (Simeon et al., 1998, 2002). DES scores were significantly correlated with harm avoidance. This is

214

K.J. Oedegaard et al. / Journal of Affective Disorders 108 (2008) 207216

thus in line with our finding of an association between higher DES scores and cyclothymic temperament. Bipolar disorders have a strong genetic component and affective temperaments are possibly intermediate (endophenotypes) between the genetic factors and overt clinical disorders ( Evans et al., 2005) so our studies point to a possible genetic basis for dissociative symptoms. This is supported by a twin study (Jang et al., 1998) where it was found that genetic influences accounted for 4855% of the variance in scores on the DES. However, it is also conceivable that behaviour associated with the affective temperaments and mood disorders predispose these persons to interpersonal conflicts and crises that in turn give rise to traumatic events associated with the development of dissociative symptoms (Carlson and Putnam, 1993; Kaplan et al., 1998). We have not made any systematic collection of information concerning previous traumatic events and it is therefore not possible to say if our findings are related to such events. We did not find any gender difference in our sample. This is in agreement with the findings of Ross et al. (1990) in a general population sample and also with Bernstein and Putnam (1986) and Spitzer et al. (2003) in different clinical non-clinical samples. However, Karadag et al. (2005) found in a study of in-patients with drug or alcohol dependency that females had higher scores than males (36.7 vs. 24.3). In a general population sample there is a low, negative correlation between DES score and age (Ross et al., 1990; Bernstein and Putnam, 1986). We did not find any correlation with age, but clearly our sample is different. In a variety of studies dissociative symptoms have been investigated in patients with anxiety disorders and OCD. In a study of adolescents, using the Adolescent Dissociative Experience Scale (Muris et al., 2003), higher scores were related both to PTSD symptoms and to other anxiety disorder symptoms (GAD, OCD and panic disorder). Cassano et al. (1989) found that derealization and depersonalization were common during panic attacks and were related to a more severe form of panic disorder. However, Ball et al. (1997), using the DES, found low scores, well within the nonpathological range (mean in the range 45), in patients with panic disorder. Higher DES scores were associated with co-morbid depressive symptoms. This is in line with our results, that dissociative symptoms are not specifically associated with panic disorder. In our study the highest DES scores were in patients with OCD. Ross and Anderson (1988) have suggested that there may be a phenomenological overlap between multiple personality disorder and some cases of OCD and that OCD patients with dissociative features

may be a distinct subgroup. Goff et al. (1992) found slightly higher dissociation scores in patients with OCD than in normal controls and comparable to patients with other anxiety disorders. Merckelbach and Wessel (2000) found a mean DES score of 24.5 in 19 OCD patients, compared to 12.6 in 16 nonpatient controls. Higher scores on DES were related to reduced confidence in correct reality monitoring decisions. Watson et al. (2004) found an association between checking and dissociative symptoms in a non-clinical sample and Rufer et al. (2006) reported, in a patient sample, that only the checking dimension of OCD showed a correlation with dissociation and that only higher scores on the DES subscale amnestic dissociation were associated with higher scores for checking compulsions. There is a strong association between OCD and both bipolar disorder ( Angst et al., 2005) and cyclothymic temperament (Hantouche et al., 2003) and it is therefore not surprising that all of these are associated with dissociative symptoms. Higher scores on DES in patients with a history of suicide attempt is consistent with suicidal behaviour being connected with scores on the DES ( Kaplan et al., 1998) and that dissociative disorders are associated with suicide attempt ( Karadag et al., 2005). Similarly in a general population study Maaranen et al. (2005) found that pathological dissociation, defined by DES-T scores over 20, was associated with depression and suicidality. We found higher DES scores in patients with psychotic symptoms during depressions. Dissociative symptoms have also previously been described in patients with psychotic mood disorders (Giese et al., 1997). Similarly eating disorders were associated with higher DES scores, also in line with previous research (Carlson and Putnam, 1993). The major limitation of this study is that we have recruited patients that represent a comparatively ill group, in a non-remission state. It is therefore difficult to know if these results can be generalized to less ill patients with mood disorders, or whether the differences observed regarding dissociative symptoms and temperamental features between MDD and BP-II patients would still be present in a state of euthymia. However, even if higher DES scores in BP-II versus MDD should be a state phenomenon, the DES might still be a useful supplement to the existing inventory for the delineation of unipolar from bipolar features in patients with mood disorders. Another limitation is of course that the diagnosis of mood disorders and temperaments have been made non-blind, however the use of self-report questionnaires for the assessment

K.J. Oedegaard et al. / Journal of Affective Disorders 108 (2008) 207216

215

of dissociative symptoms and type A behaviour assures that these data have not been influenced by interviewer bias. 5. Conclusion The finding that dissociative symptoms as measured with the Dissociative Experience Scale are associated with bipolar features, using formal DSM-IV criteria, cyclothymic temperament or the JAS, suggest that DES could be used in the routine clinical evaluation of patients with mood disorders. Calculating DES-Tscores gives little extra information, but of course using only these 8 symptoms would make both administration and scoring easier. Thus, DES can give valuable information apart from its use in the screening for narrowly defined dissociative disorders.
Role of funding source This research has been supported financially by an unrestricted grant from the legacy of Gerda Meyer Nyquist Gulbrandson & Gerdt Meyer Nyquist. Conflict of interest No conflict declared.

Acknowledgement We thank Helge Knudsen for the valuable assistance. He has provided us with the Norwegian version of DES and has given important advice concerning the study of dissociative symptoms. References
Akiskal, H.S., 1996. The prevalent clinical spectrum of bipolar disorders: beyond DSM-IV. J. Clin. Psychopharmacol. 16 (Suppl 1), 4S14S. Akiskal, H.S., Mallya, G., 1987. Criteria for the soft bipolar spectrum: treatment implications. Psychopharmacol. Bull. 23, 6873. Akiskal, H.S., Akiskal, K., 1992. Cyclothymic, hyperthymic and depressive temperaments as subaffective variants of mood disorders. In: Tasman, A., Riba, M.B. (Eds.), Annual Review of Psychiatry, Vol. 11. American Psychiatric Press, Washington DC, pp. 4362. Akiskal, H.S., Maser, J.D., Zeller, P.J., Endicott, J., Coryell, W., Keller, M., Warshaw, M., Clayton, P., Goodwin, F., 1995. Switching from 'unipolar' to bipolar II. An 11-year prospective study of clinical and temperamental predictors in 559 patients. Arch. Gen. Psychiatry 52, 114123. Akiskal, H.S., Placidi, G.F., Signoretta, S., Liguori, A., Gervasi, R., Maremmani, I., Mallya, G., Puzantian, V.R., 1998. TEMPS-I: delineating the most discriminant traits of cyclothymic, depressive, irritable and hyperthymic temperaments in a nonpatient population. J. Affect Disord. 51, 719. American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. American Psychiatric Press, Washington DC (DSM IV).

Angst, J., Gamma, A., Endrass, J., Hantouche, E., Goodwin, R., Ajdacic, V., Eich, D., Rossler, W., 2005. Obsessivecompulsive syndromes and disorders: significance of comorbidity with bipolar and anxiety syndromes. Eur. Arch. Psychiatry Clin. Neurosci. 255, 6571. Ball, S., Robinson, A., Shekhar, A., Walsh, K., 1997. Dissociative symptoms in panic disorder. J. Nerv. Ment. Dis. 185, 755760. Barrick, C.B., 1999. Sad, glad, or mad hearts? Epidemiological evidence for a causal relationship between mood disorders and coronary artery disease. J. Affect. Disord. 53, 193201. Bernstein, E.M., Putnam, F.W., 1986. Development, reliability, and validity of a dissociation scale. J. Nerv. Ment. Dis. 174, 727735. Carlson, E.B., Putnam, F.W., 1993. An update on the Dissociative Experience Scale. Dissociation 6, 1627. Cassano, G.B., Petracca, A., Perugi, G., Toni, C., Tundo, A., Roth, M., 1989. Derealization and panic attacks: a clinical evaluation on 150 patients with panic disorder/agoraphobia. Compr. Psychiatry 30, 512. Draijer, N., Boon, S., 1993. The validation of the Dissociative Experiences Scale against the criterion of the SCID-D, using receiver operating characteristics (ROC) analysis. Dissociation 6, 2837. Espenes, G.A., Opdahl, A., 1999. Associations among behavior, personality, and traditional risk factors for coronary heart disease: a study at a primary health care center in mid-Norway. Psychol. Rep. 85, 505517. Evans, L., Akiskal, H.S., Keck, P.E., McElroy, S.L., Dessa Sadovnick, A., Remick, R.A., Kelsoe, J.R., 2005. Familiality of temperament in bipolar disorder: support for a genetic spectrum. J. Affect Disord. 85, 153168. Fasmer, O.B., 2001. The prevalence of migraine in patients with bipolar and unipolar affective disorders. Cephalalgia 21, 894899. Fasmer, O.B., Oedegaard, K.J., 2001. Clinical characteristics of patients with major affective disorders and comorbid migraine. World J. Biol. Psychiatry 2, 149155. Friedman M., Rosenman R.H. 1974. Type A Behavior and Your Heart, Knopf, New York. 163(4):6239. Giese, A.A., Thomas, M.R., Dubovsky, S.L., 1997. Dissociative symptoms in psychotic mood disorders: an example of symptom nonspecificity. Psychiatry 60, 6066. Goff, D.C., Olin, J.A., Jenike, M.A., Baer, L., Buttolph, M.L., 1992. Dissociative symptoms in patients with obsessivecompulsive disorder. J. Nerv. Ment. Dis. 180, 332337. Grabe, H.-J., Spitzer, C., Freyberger, H.J., 1999. Relationship of dissociation to temperament and character in men and women. Am. J. Psychiatry 156, 18111813. Hantouche, E.G., Angst, J., Demonfaucon, C., Perugi, G., Lancrenon, S., Akiskal, H.S., 2003. Cyclothymic OCD: a distinct form? J. Affect. Disord. 75, 110. Headache Classification Committee of the International Headache Society, 1988. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia 8 (Suppl.7), 196. Jang, K.L., Paris, J., Zweig-Frank, H., Livesley, W.J., 1998. Twin study of dissociative experience. J. Nerv. Ment. Dis. 186, 345351. Jenkins, C.D., Rosenman, R.H., Friedman, M., 1967. Development of an objective psychological test for determination of the coronaryprone behaviour pattern in employed men. J. Chronic. Dis. 20, 371379. Jenkins, C.D., Zyzanski, S.J., Rosenman, R.H., Cleveland, G.L., 1971. Association of coronary-prone behaviour pattern scores with recurrence of coronary heart disease. J. Chronic. Dis. 24, 601611.

216

K.J. Oedegaard et al. / Journal of Affective Disorders 108 (2008) 207216 Ross, C.A., Anderson, G., 1988. Phenomenological overlap of multiple personality disorder and obsessivecompulsive disorder. J. Nerv. Ment. Dis. 176, 295299. Ross, C.A., Joshi, S., Currie, R., 1990. Dissociative experiences in the general population. Am. J. Psychiatry 147, 15471552. Rufer, M., Fricke, S., Held, D., Cremer, J., Hand, I., 2006. Dissociation and symptom dimensions of obsessivecompulsive disorder: a replication study. Eur. Arch. Psychiatry Clin. Neurosci. 256, 146150. Savitz, J., Solms, M., Pietersen, E., Ramesar, R., Flor-Henry, P., 2004. Dissociative identity disorder associated with mania and change in handedness. Cogn. Behav. Neurol. 17, 233237. Silberman, E.K., Post, R.M., Nurnberger, J., Theodore, W., Boulenger, J.P., 1985. Transient sensory, cognitive and affective phenomena in affective illness. A comparison with complex partial epilepsy. Br. J. Psychiatry 146, 8189. Simeon, D., Guralnik, O., Gross, S., Stein, D.J., Schmeidler, J., Hollander, E., 1998. The detection and measurement of depersonalization disorder. J. Nerv. Ment. Dis. 186, 536542. Simeon, D., Guralnik, O., Knutelska, M., Schmeidler, J., 2002. Personality factors associated with dissociation: temperament, defenses, and cognitive schemata. Am. J. Psychiatry 159, 489491. Spitzer, C., Klauer, T., Grabe, H.J., Lucht, M., Stieglitz, R.D., Schneider, W., Freyberger, H.J., 2003. Gender differences in dissociation. Gender differences in dissociation. A dimensional approach. Psychopathology 36, 6570. Vieta, E., Colom, F., Martinez-Aran, A., Benabarre, A., Reinares, M., Gasto, C., 2000. Bipolar II disorder and comorbidity. Compr. Psychiatry 41, 339343. Waller, N.G., Ross, C.A., 1997. The prevalence and biometric structure of pathological dissociation in the general population: taxometric and behavior genetic findings. J. Abnorm. Psychology 106, 499510. Waller, N., Putnam, F.W., Carlson, E.B., 1996. Types of dissociation and dissociative types: a taxometric analysis of dissociative experiences. Psychol. Methods 1, 300321. Watson, D., Wu, K.D., Cutshall, C., 2004. Symptom subtypes of obsessivecompulsive disorder and their relation to dissociation. J. Anxiety Disord. 18, 435458.

Jenkins, C.D., Zyzanski, S.J., Rosenman, R.H., 1979. Jenkins Activity Survey. Manual. The Psychological Corporation, Cleveland. Kaplan, M.L., Erensaft, M., Sanderson, W.C., Wetzler, S., Foote, B., Asnis, G.M., 1998. Dissociative symptomatology and aggressive behavior. Compr. Psychiatry 39, 271276. Karadag, F., Sar, V., Tamar-Gurol, D., Evren, C., Karagoz, M., Erkiran, M., 2005. Dissociative disorders among inpatients with drug or alcohol dependency. J. Clin. Psychiatry 66, 12471253. Maaranen, P., Tanskanen, A., Honkalampi, K., Haatainen, K., Hintikka, J., Viinamaki, H., 2005. Factors associated with pathological dissociation in the general population. Aust. N. Z. J. Psychiatry 39, 387394. Maremmani, I., Akiskal, H.S., Signoretta, S., Liguori, A., Perugi, G., Cloninger, R., 2005. The relationship of Kraepelian affective temperaments (as measured by TEMPS-I) to the tridimensional personality questionnaire (TPQ). J. Affect. Disord. 85, 1727. Merckelbach, H., Wessel, I., 2000. Memory for actions and dissociation in obsessivecompulsive disorder. J. Nerv. Ment. Dis. 188, 846848. Modestin, J., Erni, T., 2004. Testing the dissociative taxon. Psychiatry Res. 15, 7782. Modestin, J., Ebner, G., Junghan, M., Erni, T., 1996. Dissociative experiences and dissociative disorders in acute psychiatric inpatients. Compr. Psychiatry 37, 355361. Muris, P., Merckelbach, H., Peeters, E., 2003. The links between the Adolescent Dissociative Experiences Scale (A-DES), fantasy proneness, and anxiety symptoms. J. Nerv. Ment. Dis. 191, 1824. Nijenhuis, E.R., van Dyck, R., Spinhoven, P., van der Hart, O., Chatrou, M., Vanderlinden, J., Moene, F., 1999. Somatoform dissociation discriminates among diagnostic categories over and above general psychopathology. Aust. N. Z. J. Psychiatry 33, 511520. Oedegaard, K.J., Fasmer, O.B., 2005. Is migraine in unipolar depressed patients a bipolar spectrum trait? J. Affect. Disord. 84, 233242. Oedegaard, K.J., Neckelmann, D., Fasmer, O.B., 2006. Type A behaviour differentiates bipolar II from unipolar depressed patients. J. Affect Disord. 90, 713. Placidi, G.F., Signoretta, S., Liguori, A., Gervasi, R., Maremmani, I., Akiskal, H.S., 1998. The Semi-Structured Affective Temperament Interview (TEMPS-I): reliability and psychometric properties in 1010 1426 year students. J. Affect Disord. 47, 110.