the experience of tension in patients with borderline

Journal of Social and Clinical Psychology, Vol. 27, No. 5, 2008, pp. 425–446 STIGLMAYR THE EXPERIENCE ET AL.OF TENSION

THE EXPERIENCE OF TENSION IN PATIENTS WITH BORDERLINE PERSONALITY DISORDER COMPARED TO OTHER PATIENT GROUPS AND HEALTHY CONTROLS CHRISTIAN E. STIGLMAYR Consortium for Scientific Psychotherapy Berlin (AWP–Berlin), Germany JEANNETTE BISCHKOPF, VICTORIA ALBRECHT, NANCY PORZIG, STEPHAN SCHEUER Freie Universität Berlin, Germany CLAAS–HINRICH LAMMERS Medical University Berlin, Germany ANNA AUCKENTHALER Freie Universität Berlin, Germany

People displaying a pattern of thinking, feeling, and behaving that is sufficiently severe to exceed a set of formal diagnostic criteria of borderline personality disorder (BPD) report states of aversive inner tension, which they attempt to alleviate by impulsive self–harming behavior. Further research is needed to understand the actual experience of inner tension and its meaning and origin. The present study involves a systematic examination of different clinical groups’ experience of inner tension. 117 participants (30 participants fulfilling the diagnostic criteria of borderline personality disorder, 30 participants fulfilling the diagnostic criteria of depression, and 27 participants fulfilling the diagnostic criteria of anxiety disorders as well as 30 participants without any mental diagnosis) were asked about their experiences of inner tension using an open questionnaire. Qualitative Content Analysis (Mayring, 2000) was applied for coding participants’ self reports. Among all groups inner tension is represented as an unpleasant state of arousal. However, in participants fulfilling diagnostic criteria the experience of tension is triggered by a sense of inner helplessness and inaction whereas healthy controls experience the stress of a pressure to perform well. The quantitative analyzes revealed significant disorder–spe-

Address correspondence to Christian Stiglmayr, Ph.D., AWP–Berlin, Bundesring 58, 12101 Berlin, GERMANY; E–Mail: [email protected].

425

426

STIGLMAYR ET AL.

cific differences relating to the experience of inner tension, e.g., that the states of tension in participants with BPD are related to a negative view of the self and participants with anxiety disorders show the highest frequency of health concerns. The term “tension” carries various meanings for different client populations. In the clinical setting, different approaches need to be applied to address different meanings and aspects of the experience of inner tension.

In the literature, the term tension is mainly used to describe an emotional state or with regard to the experience of stress. In general, emotions can be differentiated and described on three levels: emotional experience, emotional expression, and behavioral aspects of emotions (Izard, 1999; Schmidt–Atzert, 1996). Tension can be understood and described on all of these levels and various models have differentiated the experience of tension accordingly. For instance, in the two–dimensional model of Russell (1980) the term “tense” is placed between high arousal and low valence, whereas the term “relaxed” lies exactly opposite between low arousal and high valence. Watson and Tellegen (1985) in their study found two orthogonal factors, positive affect—negative affect, with the term “tense” being related to negative affect. According to Lang, Bradley, and Cuthert (1990) affective valence generally directs behavior. Low affective valence leads to protective and defensive behavior, whereas high affective valence leads to approaching behavior. In general, the term “tension” describes at the semantic level the subjective perception of extreme emotional arousal in the face of aversive attribution. The extent of emotional arousal regulates the resources available to bring this state to an end. People displaying a pattern of thinking, feeling, and behaving that is sufficiently severe to exceed a set of formal diagnostic criteria of borderline personality disorder (BPD) report states of aversive inner tension, which they often try to terminate by impulsive self–harming or other dysfunctional behavior (Coid, 1993; Favazza, 1989; Favazza & Rosenthal, 1993; Herpertz, 1995; Leibenluft, Gardner, & Cowdry, 1987; Russ et al., 1992; Shearer, Peters, Quaytman, & Wadman, 1988; Wilkins & Coid, 1991; Zanarini et al., 1998). Recent studies have shown that women who fulfil the diagnostic criteria of BPD experience more elevated rates of aversive inner tension compared to healthy controls. In addition, their states of aversive inner tension are more frequent, rise faster, and last longer (Stiglmayr et al., 2005).

THE EXPERIENCE OF TENSION

427

Compared to people fulfilling the diagnostic criteria of an anxiety disorder, depression or dysthymia people fulfilling the diagnostic criteria of BPD experience higher level states of aversive inner tension and their states of aversive inner tension rise faster (Stiglmayr, Mohse, Behm, Auckenthaler, & Bohus, 2004). Aversive inner tension in people with severe borderline symptoms is strongly related to dissociative features, both somatoform and psychological (Stiglmayr, Shapiro, Stieglitz, Limberger, & Bohus, 2001). This positive correlation between aversive inner tension and dissociation suggests that people fulfilling the diagnostic criteria of BPD feel much more impaired and distressed by states of inner tension and dissociation compared to other groups. Therefore, aversive inner tension has been recommended to be included in the diagnostic criteria for BPD (Bohus, 2002), and therapeutic interventions focus on teaching skills for tension regulation (Linehan, 1993a). Furthermore, acute states of tension in people with borderline features are the target for psychopharmacological treatment (Philipsen et al., 2004). The experience of tension is not only discussed with regard to features of BPD but is also a diagnostic criterion for a number of other mental disorders, i.e., generalized anxiety disorder, impulse control disorders, or sleep disorders (WHO, 1991). Moreover, the experience of tension is understood as part of the etiological concepts for somatoform disorder and thus part of its treatments (Hiller & Rief, 2000). In fact, for some disorders, inner tension is such a clear feature that it has been included in psychometric assessment. The Hamilton Anxiety Scale (Hamilton, 1996), for example, and the Montgomery Asberg Depression Rating Scale (MADRS; Gorman, Korotzer, & Su, 2002) both contain items relating to tension.

THE PRESENT STUDY Relatively little work has been done in order to understand the phenomenon of inner tension specifically and its origin and subjective significance seems quite unclear. Little is known about what causes the emergence or worsening of aversive inner tension. As a result, the concept of tension remains a relatively poorly defined category and warrants more attention than it has received to date given its clinical importance. In order to achieve a more content– and context–based under-

428

STIGLMAYR ET AL.

standing, we first focused on participants’ usage of the term “tension” in order to find a spectrum of all that can be labelled “inner tension,” as it remains unclear which inner state participants themselves label “aversive inner tension.” Building on the results of this pre–study we developed a questionnaire that was then administered in a control group study. We asked participants fulfilling different formal diagnostic syndromes about their usage of the term and their experience of tension. Thus, the overall question of the study was: What inner state is labelled “inner tension” and does this differ among groups with different diagnostic features and participants not fulfilling any formal diagnostic syndrome? The study was conducted at the Freie Universität Berlin, Department of Clinical Psychology and Psychotherapy.

METHODS PARTICIPANTS One hundred and seventeen people aged between 18 and 50 years participated in the study on the experience of tension and completed a questionnaire that was specifically developed for this study. The items of this questionnaire were based on the results and the experiences from the pre–study (see Materials). The questionnaire survey was conducted between March and September 2003 in 9 psychiatric clinics and 12 private practices in Germany and Switzerland.1 A control–group design was used with the following inclusion criteria: age between 18 and 50 years, sufficiently severe symptoms to exceed the formal criteria of a diagnosis of BPD, depression or anxiety disorder, and in–patient or out–patient treatment. The sample consisted of 30 women with a diagnosis of BPD, 30 participants with a diagnosis of depression (22 women, 8 men), 27 participants with a diagnosis of anxiety disorder (19 women, 8 men) and 30 participants not fulfilling any mental diagnosis (18 women, 12 men; this group will be refered to as “controls” further on). Participants in treatment were diagnosed by their therapists using DSM–IV–criteria. For all participants the following exclusion criteria were defined: a life–time diagnosis of schizophrenia or bipolar–I–disorder, current alcohol or 1. The survey was conducted by Victoria Albrecht and Nancy Porzig.


429

drug abuse, comorbid BPD in participants fulfilling the criteria of an anxiety disorder or depression. Exclusion criteria for the controls were as follows: diagnosis of an axis–I–disorder in a clinical interview (SCID–I for DSM–IV; First, Spitzer, Gibbon, & Williams, 1997), diagnosis of an axis–II–disorder in a clinical interview (SCID–II for DSM–IV; First, Gibbon, Spitzer, Williams, & Benjamin, 1997), as well as current psychotherapy or a first grade relative with a mental disorder. Two participants (6.25%) were excluded from the study because of an axis–I–disorder. The controls were recruited through the snow ball system. Among the participants with anxiety disorder 13 had a diagnosis of panic disorder, 6 a diagnosis of phobia, 4 a combined diagnosis of phobia and panic disorder, one a diagnosis of generalized anxiety disorder, and 3 an anxiety diagnosis not further specified. In case of comorbid disorders the anxiety disorder had to be the primary diagnosis. Among the participants with a diagnosis of depression 10 had a diagnosis of a depressive episode, 11 recurrent depression, 6 dysthymia, and 3 depression not further specified. The depression had to be the primary diagnosis in case of co–morbidity. For further clinical parameters of the participants see Table 1. Participants differ significantly in regard to age (F = 7.79; p < .001) due to the higher age of participants with depression compared to participants with BPD and controls. Participants with BPD also differ from all other groups in regard to gender (BPD vs. anxiety: χ2 = 10.34; df = 1; p < 0.001; BPD vs. depression: χ2 = 9.23; df = 1; p < 0.01; BPD vs. controls: χ2 = 15.00; df = 1; p < 0.001). This difference can be explained by the fact that only women with severe borderline features participated. All other participants did not differ significantly in regard to gender. There are no group differences in regard to education (χ2 = 5.81; df = 3; p = .901). However, there are significant group differences in regard to treatment (inpatient/day care/outpatient/no treatment) at the time of sampling (χ2 = 13.96; p < .05), which is due to the higher number of outpatients among participants with anxiety disorders. Finally, no significant group differences were found regarding psychopharmacological medication (χ2 = 5.52; p = .063).

430

Medication Current Axis I diagnosis Eating Disorder Depression/Dysthymia Anxiety Disorder PTSD

Outpatient

Therapeutic setting Inpatient/ Daycare

High school

Education 10 yrs of school education

Age

female male female male

female male female male

female male

63.3 0 36.7 0 76.7 26.7 36.7 43.3 13.3

8 11 13 4

56.7 0 43.3 0

%

19 0 11 0 23

17 0 13 0

N

Borderline Personality Disorder (N = 30) M SD 29.30 6.78 29.30 6.78

1 5 27 1

6 1 13 7 13

9 4 10 4

3.7 18.52 100.0 3.7

22.2 3.7 48.1 25.9 48.1

33.3 14.8 37.0 14.8

Anxiety Disorders (N = 27) M SD 34.26 8.96 32.00 8.97 39.63 6.67 N %

TABLE 1. Sample Characteristics

1 30 6 0

15 3 7 5 21

9 3 13 5

3.3 100.0 20,0 0,0

50.0 10.0 23.3 16.7 70.0

30.0 10.0 43.3 16.7

Depression (N = 30) M SD 37.70 8.98 37.95 8.57 37.00 10.64 N % 15 2 3 10

Controls (N = 30) M 29.30 28.44 30.58 N

50.0 6.7 10.0 33.3

SD 7.27 7.50 7.04 %


431

MATERIALS Structured clinical interviews (SCID–I and SCID–II for DSM–IV) with the controls were conducted by two graduate psychology students under the first author’s supervision. Participants in treatment were diagnosed by their therapists according to DSM–IV–criteria. Data on sociodemographic variables were also collected. In addition, a questionnaire was developed on the basis of the results of a pre–study. In this pre–study semi–structured interviews (Witzel, 2000) were conducted with seven people from various backgrounds in order to explore their experience of inner tension and map out a spectrum of meanings “tension” can have.2 The following broad data generating question was used at the beginning of the interview: “Please tell me in as much detail as possible how you personally experience something that you would call inner tension or feeling tense.” Participants were asked to define exactly what they mean when they use the term “inner tension.” Subsequent clarifying questions like the following were used: “What do you do when you feel tense? How would others see you or describe you when you are in such a state? What do you think causes the tension?” Following Witzel’s (2000) suggestions, a memo was written immediately after the interview had been conducted on the global themes which had been addressed in the conversation. Each interview was tape recorded and transcribed. All cases were summarized after each case had been analyzed using Qualitative Content Analysis (Mayring, 2000). As the present study focused exclusively on the experience of tension all subcategories from the pre–study were included in the questionnaire and one item added asking the participants to describe as detailed as they may their experience of tension. All six questions of the questionnaire are given in Table 2. DESIGN The design of the study entails a two–step procedure of data analysis (Hanson, Clark, Petska, Creswell, & Creswell, 2005). In a first step open questions in the questionnaire were subjected to a Qualitative

2. All interviews were conducted and analyzed by Stephan Scheuer.

432

STIGLMAYR ET AL. TABLE 2. Items of the Questionnaire

1. Please tell me in as much detail as possible how you personally experience something that you would call inner tension or feeling tense. Please provide more information on the following aspects: 2. What thoughts do you notice when you feel tense? 3. What emotions do you notice when you feel tense? 4. What are you aware of inside your body when you feel tense? 5. What would you like to do when you feel tense, what impulse to act to do you feel regardless of the consequences which your action have? 6. What do you actually do when you feel tense?

Content Analysis (Mayring, 2000). Second, all data were dichotomized in order to investigate disorder specific differences statistically. PROCEDURE Following an introduction into the aims and procedures of the study the study participants gave their informed consent prior to data collection. Therapists (psychiatrists or psychologists) provided additional information on their patients by answering a short questionnaire concerning DSM–IV diagnosis, current co–morbidities, current psychotropic medication, and treatment setting (outpatient, day care, inpatient). For the participants´ questionnaire about their experience of tension please refer to Table 2. QUALITATIVE ANALYZES Following Qualitative Content Analysis (Mayring, 2000) open questionnaire data were subjected to open coding, categorizing, linking, and reduction in order to generate categories and modify existing categories. First, a list of domains was made using the literature on tension and the results of the first study as guidelines. This first list of categories was then modified if new domains emerged from the data or existing domains were collapsed into a higher order category. Regular consultations with the second authors as well as participation in a group familiar with the method helped the re–examination and re–coding of data. In this way, it was attempted to validate the results of each stage in the research process.


433

QUANTITATIVE ANALYZES In order to apply statistical analyzes the qualitative data were dichotomized into 0 (nonexistent) and 1 (existent) along the categories. The number of quotes in each category was summed up for each group of disorder and transformed into percentile scores for easier interpretation (the percentage of people in one group, whose answers were coded in the category in question). The level of significance was set at .05; as no direction of the hypothesis had been formulated, a two–sided test was performed. Correlations were calculated to control for confounding variables. Depending on the measurement scale either Pearson´s r, Spearman´s rho or point biserial correlations were used. For between group comparisons analyzes of variance were admistered; individual comparisons were calculated applying the Scheffé–Test. As a pre–requisite for the administration of parametrical tests homogeneity of variance was calculated. If the criterion of homogeneity of variance was not met, a nonparametrical test was used. For overall analyzes the Kruskal–Wallis–Test was chosen and for comparing two independent samples the Kolmogorov–Smirnov–Test was used. Only, if results differed in relation to the test used, the result of the test with the lower level of scale is given. For calculating group differences along individual categories the χ2–Test was used due to the nominal data. In case the expected frequency in one or more cells was less than five, the Fisher–Yates–Test was administered. An alpha correction according to Bonferroni was carried out in the case of multiple comparisons.

RESULTS QUALITATIVE RESULTS The qualitative analysis categorized the experience of tension into the following fields: cognitive, emotional, physical, and behavioral aspects of tension, action tendencies, and coping mechanisms. Cognitive aspects were described broadly, i.e., “It is as if things pile up in my head and then, when it is getting too much, I can really feel how the tension comes.” In some cases the actual thoughts that run through a person’s head in a state of inner tension were documented, i.e., “I cannot stand it any longer! I need to calm down! I will go crazy! Will that never stop? I have to pull myself together!” Feeling tense

434

STIGLMAYR ET AL.

can have many emotional nuances, for example: “Anxiety, anger, grief, despair, helplessness, hopelessness, guilt, pain, loneliness, nervousness.” If people were asked how they would alleviate the tension (whether or not that was possible), they most often reported trying to go away or numbing oneself: “Leave my body and fly away, cut my body off, cut my hair off, burn something or take drugs.” Tension is often felt in the body, for instance as tense muscles, especially of the neck and shoulders. It can also be experienced as an uneasiness and restlessness or a tightening inside the body: “I feel unsteady and nervous. I can feel this in all my body, this uneasiness.” “I feel a pressure in my stomach that wants to get out but can’t, like something that wants to explode.” People often use the metaphor of a steam cooker when referring to inner tension. Tension slowly builds up and leaves the individual in a hyperactive state: “When I feel tense I run around like mad, I don‘t know, for 12 hours, and that’s nothing then.” Tension may be experienced as uncontrollable: “Most of the time I lose my temper. Well, I do not hit anyone, but I shout.” Ways of coping with tension relate to its control and management or the socially acceptable expression of tension. If people feel overwhelmed or are afraid of losing control, they may want to isolate themselves for a while: “I need to be alone then, because I would not know, if someone started talking to me, how I would react. I am constantly worried, that I would react in an uncontrollable aggressive way.” QUANTITATIVE RESULTS Confounding Variables In order to assess the potential influence of variables on the results, only sex is included in the analyzes as a confounding variable. The participants’ age and their treatment settings are not being controlled as it can be assumed that their distribution reflect general treatment conditions. For the sub–categories “Anger/Hatred” (rpbis = .20, p < .05), “Grief/Sadness” (rpbis = .21, p < .05) as well as “Autoaggression” (rpbis = .24, p < .05) a significant correlation with the variable “sex” was found. For these subcategories the results are calculated taking a potential influence of the patient’s sex into account.


435

Results of Disorder Specific Comparisons Results of disorder specific comparisons are presented along the main categories that were found in the qualitative analyzes: cognitive aspects, emotional aspects, physical aspects, action tendency, behavioral aspects and coping. Due to the number of individual comparisons only those sub–categories in each main category are presented which yielded significant differences among groups. Cognitive Aspects (Figure 1) Within the main category “cognitive aspects” participants with a diagnosis differ significantly regarding the subcategory “Negative view of self” (BPD [6]3 vs. controls [0]: Fisher–Yates: p < .05), “Excessive demand” (BPD [9] vs. controls [1]: Fisher–Yates: p < .05), “Pressure to perform” (controls [7] vs. BPD [0]: Fisher–Yates: p < .05), “Thinking about self–harming behavior/suicide/death” (BPD [12] vs. controls [0]: Fisher–Yates: p < .001; BPD [12] vs. depression [1]: Fisher–Yates: p = .001; BPD vs. anxiety [1]: Fisher–Yates: p = .001), and “Health concerns” (anxiety [4] vs. controls [0]: Fisher–Yates: p < .05; anxiety [4] vs. BPD [0]: Fisher–Yates: p < .05). Emotional Aspects (Figure 2) In order to make interpretation easier results in the main category “mood/feelings/emotions” were divided into “distinct emotions” and “nondistinct emotions.” Regarding “distinct emotions” groups differ significantly with respect to “Anxiety/Panic” (anxiety [15] vs. controls [4]: Fisher–Yates: p < .01; depression [12] vs. controls [4]: Fisher–Yates: p < .05), “Fear of failure” (controls [7] vs. BPD [0]: Fisher–Yates: p < .05), “Anger” (BPD [18] vs. controls [5]: χ2 = 11.92, df = 1, p = .001; BPD [18] vs. depression [6]: χ2 = 10.00, df = 1, p < .01; BPD [18] vs. anxiety [5]: χ2 = 10.16, df = 1, p = .001), “Hatred/Aggression” (BPD [8] vs. controls [1]: Fisher–Yates: p < .05), “Despair” (BPD [6] vs. controls [0]: Fisher–Yates: p < .05), and “Loss of control” (BPD [6] vs. depression [0]: Fisher–Yates: p < .05). Regarding “nondistinct emotions” the same pattern was found for “Confusion/Chaotic feelings” (BPD [7] vs. controls [0]: Fisher–Yates: p < .05; BPD [7] vs. anxiety [0]: Fisher–Yates: p < .05), “Numbness” (BPD [7] vs. controls [0]: Fisher–Yates: p < .05), “Feeling 3. Number of quotes.

436

STIGLMAYR ET AL.

45 40 negative view of self *

percent %

35 30

excessive demand *

25

pressure to perform *

20

thinking about selfharming/suicide/death ***

15

health concerns *

10 5 0 controls (N=30)

depression (N=30)

anxiety (N=27)

BPD (N=30)

* p < .05 ** p < .01 *** p < .001

FIGURE 1. Disorder specific comparisons within the main category “cognitive aspects”

under stress/Inability to cope or manage” (controls [8] vs. BPD [0]: Fisher–Yates: p < .01; depression [8] vs. BPD [0]: Fisher–Yates: p < .01), and “Insecurity/Worthlessness” (depression [10] vs. anxiety [2]: Fisher–Yates: p < .05). Physical Aspects (Figure 3) Significant group differences were found in the category “physical aspects” with regard to “Dizziness/Faint” (anxiety [11] vs. controls [0]: Fisher–Yates: p < .001; anxiety [11] vs. depression [0]: Fisher–Yates: p < .001; anxiety [11] vs. BPD [2]: Fisher–Yates: p < .01), “Breathing” (anxiety [9] vs. controls [1]: Fisher–Yates: p < .01; BPD [8] vs. controls [1]: Fisher–Yates: p < .05), “Restlessness” (depression [15] vs. anxiety [3]: Fisher–Yates: p < .01; depression [15] vs. controls [5]: χ2 = 7.50, df = 1, p < .01), “Physical numbness” (BPD [6] vs. controls [0]: Fisher–Yates: p < .05), “Depersonalisation” (BPD [7] vs. controls [0]: Fisher–Yates: p < .05; BPD [7] vs. anxiety [0]: χ2 = 7.18, df = 1, p < .01), “Explosion” (BPD [10] vs. controls [0]: Fisher–Yates: p = .001; BPD [10] vs. anxiety [0]: Fisher–Yates: p = .001; BPD [10] vs. depression [1]: Fisher–Yates: p < .01), as well as “Weakness” (BPD [7] vs. depression [0]: Fisher–Yates: p < .05).


437

Distinct Emotions

70 60

percent %

50

anxiety/panic ** fear of failure *

40

anger ***

30

hatred/aggression *

20

loss of control *

despair *

10 0 controls (N=30)

depression (N=30)

anxiety (N=27)

BPD (N=30)

* p < .05 ** p < .01 *** p < .001

Non-Distinct Emotions

35 30

confusion/chaotic feelings **

percent %

25 numbness **

20 feeling under stress *

15

insecurity/worthlessn ess *

10 5 0 controls (N=30)

depression (N=30)

anxiety (N=27)

BPD (N=30)

* p < .05 ** p < .01 *** p < .001

FIGURE 2. Disorder specific comparisons within the main category “emotional aspects”

Action Tendency (Figure 4) Significant group differences were found in the category “action tendency” with regard to “Flight/Avoidance” (anxiety [15] vs. BPD [5]: χ2 = 9.44, df = 1, p < .01), “Self–harming behavior” (BPD [21] vs. controls [0]: Fisher–Yates: p < .001; BPD [21] vs. anxiety [1]: Fisher–Yates: p < .001; BPD [21] vs. depression [2]: Fisher–Yates: p < .001), and “Dis-

438

STIGLMAYR ET AL. Physical Aspects/Vegetative 45 40 35

percent %

30 25

dizziness/faint *** breathing *

20 15 10 5 0 controls (N=30)

depression (N=30)

anxiety (N=27)

BPD (N=30)

* p < .05 ** p < .01 *** p < .001

Physical Aspects/Other

60

percent %

50 40

restlessness ** physical numbness *

30

depersonalisation ** explosion ***

20

w eakness *


depression (N=30)

anxiety (N=27)

BPD (N=30)

* p < .05 ** p < .01 *** p < .001

FIGURE 3. Disorder specific comparisons within the main category “physical aspects”

rupted eating/Drinking (e.g., refusal to eat)” (BPD [6] vs. controls [0]: Fisher–Yates: p < .05; BPD [6] vs. depression [1]: Fisher–Yates: p < .05). Behavioral Aspects (Figure 5) Group differences were found in the category “behavioral aspects” with regard to “Trembling” (BPD [9] vs. controls [2]: Fisher–Yates: p


439

80 70 flight/avoidance *

percent %

60 50 40

self-harming ***

30 20

disrupted eating/drinking ***


depression (N=30)

anxiety (N=27)

BPD (N=30)

* p < .05 ** p < .01 *** p < .001

FIGURE 4. Disorder specific comparisons within the main category “action tendency”

< .05; BPD [9] vs. anxiety [2]: Fisher–Yates: p < .05), and “Uncontrollable movements” (BPD [12] vs. anxiety [2]: χ2 = Fisher–Yates: p < .01). Coping Mechanisms (Figure 6) Significant group differences were found in the category “coping” with regard to “Using skills” (BPD [10] vs. controls [0]: Fisher–Yates: p < .01; BPD [10] vs. depression [0]: Fisher–Yates: p < .01; BPD [10] vs. anxiety [0]: Fisher–Yates: p < .01), and “Self–harming behavior” (BPD [11] vs. controls [0]: Fisher–Yates: p < .001; BPD [11] vs. depression [1]: Fisher–Yates: p < .01; BPD [11] vs. anxiety [2]: χ2 = Fisher–Yates: p < .05).

DISCUSSION This study partly confirms the findings of earlier studies (Stiglmayr et al., 2001, 2004, 2005) but also reveals disorder specific aspects of tension, which have not previously been considered. Participants with a diagnosis of BPD, for instance, reported a negative view of the self compared to controls. Thoughts about coping with tension are more often related to self harm, suicide, and death for participants with a diagnosis of BPD compared to all other groups. Participants with a diagnosis of anxiety disorders show the highest frequency of

440

STIGLMAYR ET AL.

45 40

percent %

35 30

trembling *

25 uncontrollable movements **

20 15 10 5 0 controls (N=30)

depression (N=30)

anxiety (N=27)

BPD (N=30)

* p < .05 ** p < .01 *** p < .001

FIGURE 5. Disorder specific comparisons within the main category “behavioral aspects”

health concerns compared to controls and participants with a diagnosis of BPD. As emotional aspects of tension, participants with a diagnosis of BPD report feelings of anger, hatred/aggression, and despair in relation to tension more often than controls. Regarding anger, participants with a diagnosis of BPD also differ significantly from participants with a diagnosis of depression and participants with a diagnosis of anxiety disorders. And, participants with a diagnosis of BPD experience loss of control more often than participants with depression. As expected, participants with a diagnosis of anxiety disorders report more feelings of fear and panic than controls. However, the same pattern was found for participants with a diagnosis of depression. Feelings of fear of failure and the pressure to perform well were more frequent among controls compared to participants with a diagnosis of BPD. Participants with a diagnosis of BPD experience more confusion/chaotic feelings compared to controls and participants with a diagnosis of anxiety disorders, and numbness was more frequent in participants with a diagnosis of BPD compared to controls. With regard to physical aspects for participants with a diagnosis of BPD tension is associated with breathing, physical numbness, depersonalisation, feeling torn, and ready to explode due to pent–up tension. Participants with a diagnosis of BPD differ especially to partici-


441

40

percent %

35 30

using skills ***

25

self-harming ***

20 15 10 5 0 controls (N=30)

depression (N=30)

anxiety (N=27)

BPD (N=30)

* p < .05 ** p < .01 *** p < .001

FIGURE 6. Disorder specific comparisons within the main category “coping mechanisms”

pants with a diagnosis of depression and a diagnosis of anxiety disorders in regard to these physically experienced feelings of not being under control (feeling torn, ready to explode). Weakness is reported more often by participants with a diagnosis of BPD compared to participants with a diagnosis of depression. Participants with a diagnosis of anxiety disorders report symptoms as dizziness/faint and altered breathing patterns whereas participants with a diagnosis of depression report restlessness in relation to tension. With regard to action tendencies related to tension, participants with a diagnosis of BPD report impulses to harm themselves more often than all other groups. Moreover, they report disrupted eating and drinking behaviors more often than controls and participants with a diagnosis of depression. Participants with a diagnosis of anxiety disorders report a tendency to flight or avoid the situation more often than participants with a diagnosis of BPD. With regard to behavioral aspects, participants with a diagnosis of BPD show more trembling related to tension compared to controls and participants with a diagnosis of anxiety disorders—and only compared to participants with a diagnosis of anxiety disorders more uncontrollable movements. Participants describe a wide range of coping strategies. However, the only difference between the groups was found related to specific

442

STIGLMAYR ET AL.

skills, which were used by participants with a diagnosis of BPD most often compared to all other groups. At the same time, however, participants with a diagnosis of BPD considered self–harm a coping strategy. The high frequency of skills can be explained by the fact that most of the participants with a diagnosis of BPD in this study had participated or were currently in a skills–group following the manual of DBT (Linehan, 1993b; N = 25; 83.3%). The higher frequency of self–harming behavior in participants with a diagnosis of BPD, however, cannot be explained by those participants receiving no treatment, as 37.5% of the participants who had received or were currently in DBT still used self–harming behavior as a coping strategy. In summary, based on our results, tension could be best described as a state of negative emotional arousal and differentiated in emotional experience, emotional expression, and behavioral aspects for each of the groups studied as well as people with no mental diagnosis. This observation seems to be true for each group, that was included in the current study, that means for people experiencing features fulfilling the diagnostic criteria of BPD or depression or anxiety disorders or even people not fulfilling any mental diagnosis. In participants with a diagnosis the experience of tension is triggered by a sense of inner helplessness and inaction whereas controls experience the stress of a pressure to perform well. Thus, patients’ experience of tension is more strongly related to the self. Other research has shown, for instance, that the semantic organization of the autobiographic memories of people with BPD is more self–related than other–related. In an autobiographic memory test, people with BPD retrieve more self–related memories than other related memories and more negative than positive ones compared to controls (Renneberg, Theobald, Nobs, & Weisbrod, 2005; Renneberg & Fusekova, 2005). The results of our study are in line with the study of Wolff, Stiglmayr, Bretz, Lammers, and Auckenthaler (2007) in which people with BPD reported more unpleasant feelings which they experienced more intensely compared to controls. Moreover, in their study people with BPD had greater difficulty identifying their emotions compared to controls. This can be related to our finding of a higher frequency of confusion/chaotic feelings as well as numbness in the borderline group. According to Ebner–Priemer et al. (2007) people with BPD do not have difficulties differentiating emotions as a general lack of skill. Their difficulty differentiating emotions depends on


443

the level of tension they experience (Wolff et al., 2007). We thus suggest to understand tension as “white noise” that makes clear differentiation of emotions impossible if it reaches a critical level. The information the emotion carries can then not be used. People with BPD feel helpless and overwhelmed and may react inadequately to terminate this state (Stiglmayr et al., 2004). A number of strengths and limitations of the study have to be addressed. Understanding peoples’ self reports as representations of their feelings (Barrett, 2004), using self reports can be considered a valid means of investigating peoples’ experience of tension. The combination of qualitative and quantitative perspectives proved to strengthen our understanding of inner tension in different patient groups. For a more detailed discussion of the mixed–method design see Bischkopf et al. (2006). As expected, a great number of participants with a diagnosis of BPD also fulfill the criteria of a formal diagnosis of depression or anxiety disorder which calls disorder specific comparisons into question. Furthermore, the diagnostic groups of depression and anxiety disorder in the current study are heterogeneous with regard to diagnostic categories. For instance, people suffering from panic attacks may use different coping mechanisms compared to people with social phobia. Although we found a pattern of disorder–specific differences, we cannot be sure that they are a function of the specific features that define the respective disorder. In our view, future research should include larger sample sizes to address the problem of heterogeneity of samples. In addition, in order to better understand psychopathology differences between diagnosed and undiagnosed groups should be systematically examined. The concept of tension seems to be informative in this respect as it is relevant for various disorders. Due to the dichotomization answers could not be weighed which may have resulted in a loss of information from the rich qualitative material. Future studies should include a measure of intensity as patients do not only differ in regard to emotion recognition but also regarding the intensity in which they experience emotions. Most participants with a diagnosis of BPD had been exposed to DBT prior to investigation and the term “tension” is part of the psychoeducation and skills training in this therapy. However, as tension is part of borderline symptoms it may also be addressed in other therapeutic contexts. For instance, one participant with a diagnosis

444

STIGLMAYR ET AL.

of BPD who had no DBT experience reported using skills as a coping strategy. Finally, we did not validate patients’ diagnoses and relied on therapist’s assessment and only assessed controls ourselves. There are a number of implications of our results for assessment and psychotherapy. Assessing tension more systematically could be worthwhile given its high prevalence among people with BPD. A more differentiated view of tension related to different symptoms and diagnoses may proof useful for therapy planning, for instance, for working on tension producing mechanisms that may differ among patient groups. As tension is related to chaotic feelings and confusion, helping clients clearly feel and express their emotions and helping them access underlying meaning may be one way of working with tension (Greenberg, 2002). However, for some people it may be as important to help them use healthier ways of coping when experiencing tension. Thus, understanding tension and its underlying mechanisms more fully could be one more step towards developing effective treatments—not only—for people with BPD.

REFERENCES Barrett, L. F. (2004). Feelings or words? Understanding the content in self–report ratings of experienced emotion. Journal of Personality and Social Psychology, 87, 266–281. Bischkopf, J., Stiglmayr, C., Scheuer, S., Albrecht, V., Porzig, N., & Auckenthaler, A. (2006). Combining qualitative and quantitative research strategies for analysing the experience of aversive tension in different patient groups. In P. Mayring, G. L. Huber, L. Gürtler, & M. Kiegelmann (Eds.), Mixed Methodology in Psychological Research. Rotterdam: Sense. Bohus, M. (2002). Borderline–Störung (Borderline Personality Disorder). In. D. Schulte, K. Grawe, K. Hahlweg & D. Vaitl (Eds.), Fortschritte der Psychotherapie (Band 14). Göttingen: Hogrefe. Coid, J. W. (1993). An affective syndrome in psychopaths with borderline personality disorder? British Journal of Psychiatry, 162, 641–650. Ebner–Priemer, U., Welch, S., Grossman, P., Reisch, T., Linehan, M. M., & Bohus, M. (2007). Psychophysiological ambulatory assessment of affective dysregulation in borderline personality disorder. Psychiatry Research, 15, 265–275. Favazza, A. R. (1989). Why patients mutilate themselves. Hospital and Community Psychiatry, 40, 137–145. Favazza, A. R., & Rosenthal, R. J. (1993). Diagnostic issues in self–mutilation. Hospital and Community Psychiatry, 44, 134–140. First, M. B., Gibbon, M., Spitzer, R. L., Williams, J. B. W., & Benjamin, L. (1997).


445

User´s Guide for the Structured Clinical Interview for DSM–IV Axis II Personality Disorders (SCID–II). Washington, DC: American Psychiatric Press. First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1997). User´s Guide for the Structured Clinical Interview for DSM–IV Axis I Disorders (SCID–I)—Clinician Version. Washington, DC: American Psychiatric Press. Gorman, J. M., Korotzer, A., & Su, G. (2002). Efficacy comparison of escitalopram and citalopram in the treatment of major depressive disorder: Pooled analysis of placebo–controlled trials. CNS Spectrums, 7, 40–44. Greenberg, L. (2002). Emotion–focused therapy: Coaching clients to work through their feelings. Washington, DC: American Psychological Association Press. Hamilton, M. (1996). Hamilton Angst Skala [Hamilton Anxiety Scale]. In Colleqium Internationale Psychiatrie Scalarum (Eds.), Internationale Skalen für Psychiatrie (pp. 19–21). Weinheim: Beltz. Hanson, W.E., Clark, V.L.P., Petska, K.S., Creswell, J.W. & Creswell, J.D. (2005). Mixed Methods Research Designs in Counseling Psychology. Journal of Counseling Psychology, 52, 224–235. Herpertz, S. (1995). Self–injurious behavior: Psychopathological and nosological characteristics in subtypes of self–injurers. Acta Psychiatrica Scandinavica, 9, 57–68. Hiller, W., & Rief, W. (2000). Somatoforme störungen (Somatoform Disorders). In M. Hautzinger (Ed.), Kognitive verhaltenstherapie bei psychischen störungen [cognitive behavior therapy for mental disorders] (pp. 416–458).Weinheim: Beltz. Izard, C. E. (1999). Die emotionen des menschen (The emotions of human beings). Weinheim: Beltz. Lang, P. J., Bradley, M. M., & Cuthert, B. N. (1990). Emotions, attentions, and the startle reflex. Psychological Review, 97, 377–398. Leibenluft, E., Gardner, D.L., & Cowdry, R.W. (1987). The inner experience of the borderline self–mutilator. Journal of Personality Disorder, 1, 317–324. Linehan, M. M. (1993a). Cognitive–behavioral treatment of borderline personality disorder. New York: The Guilford Press. Linehan, M. M. (1993b). Skills training manual for treating borderline personality disorder. New York: The Guilford Press. Mayring, P. (2000). Qualitative inhaltsanalyze: Grundlagen und techniken (Qualitative content analysis: Basics and techniques). Weinheim: Beltz. Philipsen, A., Richter, H., Schmahl, C., Peters, J., Rusch, N., Bohus, M., & Lieb, K. (2004). Clonidine in acute aversive inner tension and self–injurious behavior in female patients with borderline personality disorder. Journal of Clinical Psychiatry, 65, 1414–1419. Renneberg, B., & Fusekova, J. (2005). Contents of autobiographical memories in borderline personality disorder and depression. Paper presented at the Autobiographical Memory Meeting, Oxford, U.K. Renneberg, B., Theobald, E., Nobs, M., & Weisbrod, M. (2005). Autobiographical memory in borderline personality disorder and depression. Cognitive Therapy and Research, 29, 343–358.

446

STIGLMAYR ET AL.

Russ, M. J., Roth, S. D., Lerman, A., Kakuma, T., Harrison, K., Shindledecker, et al. (1992). Pain perception in self–injurious patients with borderline personality disorder. Biological Psychiatry, 32, 501–511. Russell, J.A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39, 1161–1178. Shearer, S.L., Peters, C.P., Quaytman, M.S., & Wadman, B.E. (1988). Intent and lethality of suicide attempts among female borderline inpatients. American Journal of Psychiatry, 145, 1424–1427. Schmidt–Atzert, L. (1996). Lehrbuch der Emotionspsychologie (Textbook of psychology of the emotions). Stuttgart: Kohlhammer. Stiglmayr, C. E., Grathwol, T., Linehan, M. M., Ihorst, G., Fahrenberg, J., & Bohus, M. (2005). Aversive tension in patients with borderline personality disorder: A computer–based controlled field study. Acta Psychiatrica Scandinavica, 111, 372–379. Stiglmayr, C. E., Mohse, M., Behm, R., Auckenthaler, A. & Bohus, M. (2004). Spannungszustände und dissoziative zustände bei patientinnen mit einer borderline–persönlichkeitsstörung, einer angststörung oder einer depressiven störung—eine computerbasierte kontrollierte Feldstudie (Experience of aversive tension and dissociation in female patients with borderline personality disorder, anxiety disorder or depressive disorder—a computer–based controlled field–study). In R. Merod (Ed.), Persönlichkeitsstörungen (Personality disorders) (pp. 549–572). Tübingen: DGVT. Stiglmayr, C. E., Shapiro, D. A., Stieglitz, R.–D., Limberger, M. F., & Bohus, M. (2001). Experience of aversive tension and dissociation in female patients with borderline personality disorder—a controlled study. Journal of Psychiatric Research, 35, 111–118. Watson, D., & Tellegen, A. (1985). Toward a consensual structure of mood. Psychological Bulletin, 98, 219–235. World Health Organisation (WHO) (1991). international classification of mental disorders: ICD–10, Chapter V (F). Bern: Huber. Wilkins, J., & Coid, J. (1991). Self–mutilation in female remanded prisoners: I. An indicator of severe psychopathology. Criminal Behavior and Mental Health, 1, 247–267. Witzel, A. (2000). Das Problemzentrierte Interview. Forum Qualitative Sozialforschung [The Problem–Focused Interview. Forum Qualitative Social Research]. On–line Journal, 1, http//qualitative–research.net/fqs. Wolff, S., Stiglmayr, C. E., Bretz, H. J., Lammers C.–H., & Auckenthaler, A. (2007). Emotion identification and tension in patients with borderline personality disorder. British Journal of Clinical Psychology, 57, 347–360. Zanarini, M. C., Frankenburg, F. R., DeLuca, C. J., Hennen, J., Khera, G. S., & Gunderson, J. G. (1998). The pain of being borderline: Dysphoric states specific to borderline personality disorder. Harvard Review of Psychiatry, 6, 201–207.