1. Background
According to the DSM-V diagnostic criteria, borderline personality disorder (BPD) is classified as a personality disorder while it can be considered a complex and controversial issue (1).
Psychiatric literature indicates that BPD often presents highly comorbidity with affective disorders (2). Also, many of the diagnostic criteria for BPD (e.g., interpersonal sensitivity, impulsivity, and affective instability) are observed in affective disorders that resulted some researchers attempt to link BPD with mood disorders (3-6). In this respect, Akiskal first suggested that BPD is actually part of a bipolar spectrum (3). Heprimarily suggested the concept of spectrum to describe mania and hypomania but also mild, subclinical and atypical bipolar disorders. Moreover, Bipolar spectrum definition includes personality traits and temperaments such as cyclothymic or hyperthymic. By definition, spectrum is a range of linked conditions, symptoms and traits that are supposed to be caused by the same underlying mechanism and may represent a range of severity from relatively severe disorders to relatively mild and nonclinical deficits (7). Based on this suggestion, characteristics such as affective instability and impulsive behaviors in BPD and bipolar disorder derive from the same underlying mechanism. Reich and colleagues (8) suggested that bipolar disorders and BPD may share a cyclothymic temperament involving affective reactivity and interpersonal sensitivity.
An important corollary of the attachment theory is that emotion regulation and interpersonal relationships are related to the attachment style. Therefore, based on this theory, insecure attachment may be the cause of many of pathological conditions including dysfunctional anger or disturbed relationships. Thus, to better understand the relationship between these 2 syndromes, it is useful to investigate the phenomenological similarities and differences between these 2 disorders. Reviewing the literature on this issue (9) suggests that BPD patients represent 2 prototypes of insecure attachments: disorganized/ disoriented form of attachment in childhood that is characterized by denial, confusion, or fearfulness about dependency and anxious/preoccupied form of attachment in adulthood, which is described by pleas for attention or help, clinging, and checking for proximity (10, 11). Patients with BPD are characterized by sever difficulties within intimate relationships (12-14). Most literature indicates that severe intimate relationship dysfunction in patients with BPD is characterized by frequent conflict (15), higher risk of abuse (16), lower rate of marriage (17), higher number of relationships (but not more time-spending in relationship), higher chronic stress in the romantic relationship, poorer relationship satisfaction (18), higher perceived parental criticism (19), and higher perceived parental conflicts as well as fewer appropriate responses (20).
As indicated by the DSM-V criteria for BPD (number 8 criteria, inappropriate intense anger or difficulty controlling anger) dysfunctional anger has been identified as the key problematic aspect of BPD. Accordingly, empirical researches indicated that patients with BPD typically show high levels of trait anger (21). Furthermore, Gardner et al. (22) found heightened levels of anger, hostility, and irritability among individuals with BPD compared to healthy controls.
According to the results of previous researches, we hypothesized that patients with BPD and bipolar type II (BD-II) disorder would have similar attachment styles, namely anxious or insecure attachment styles. Another purpose of this study was to compare the similarities and differences of quality of intimate relationship and anger experience in both groups. Our assumption is that regarding the same attachment style these 2 groups would have similar intimate relationships and anger experiences. Finally, in this study we implicitly investigate the validity of re-classification of BPD on the axis I disorders, especially bipolar spectrum disorders.
2. Objectives
Objectives of the current research were investigating of attachment style, quality of intimate relationship, and anger experience in patients with borderline personality and bipolar-II disorders and comparing these variables in these groups
3. Materials and Methods
3.1. Participants
Research method of the current study is comparative. A total of 37 borderline personality disorder patients (13 males and 24 females, mean age: 26.25 ± 5.6, range: 18 - 45 years old) and 41 stable bipolar-II (22 males and 19 females, mean age: 27.36 ± 8.3, range: 18 - 42 years old) that met the DSM-V criteria (according to psychiatrist diagnosis) participated in this study (Table 1). Subjects were gathered by Convenience Sampling method and the numbers were according to the research team ability, while these numbers are quite proper for comparing purposes. All participants were selected in similar circumstances that are hospital conditions. Gathering data in similar situations make it more possible to gather unbiased data. All patients were recruited from the psychiatric ward of Kamkar-Arabnia hospital, Qom, Iran. BPD and bipolar-II diagnostic evaluation consisted of psychiatrist diagnosis, Borderline Personality Inventory (BPI), and mood disorder questionnaire (MDQ). Inclusion criteria were a minimum age of 18, ability to read and answer questions, and patients’ agreement to participate in the study. Participants were excluded if they were currently psychotic, substance abusers, or suicidal, if they had a lifetime history of schizophrenia, schizoaffective, or other psychotic disorders, and serious organic conditions.
Table 1.Demographic Data for Patients with BPD and BD-IIa
| Variable | BPD (n = 37) | BD-II (n = 41) |
|---|---|---|
| Sex, female | 64.9 | 46.3 |
| Age, y | 26.25 ± 5.6 | 27.36 ± 8.3 |
| Education, y | 10 ± 3 | 12 ± 3 |
| Marital status, s, m, d | 59.5, 16.2, 24.3 | 53.7, 41.5, 4.9 |
| History of hospitalization, yes | 73.0 | 75.6 |
Abbreviations: BD-II, Bipolar II disorder; BPD, Borderline Personality Disorder; D, divorced; M, married; S, single.
aValues are expressed as mean ± S.D or %.
3.2. Measures
Borderline personality inventory (BPI) is a self-reported questionnaire that was designed by Leichsenring (23) on the basis of Kernberg’s (24) concept of borderline personality organization, DSM-V diagnostic criteria, and Gunderson’s (25) concept of borderline personality disorder. According to Leichsenring (23), the BPI combines dimensional and categorical models of BPD. Internal consistency and test-retest reliability for BPI have been satisfactory (Cronbach’s alpha = 0.68 - 0.91, rtt = 0.73 - 0.89). Results for sensitivity have been 0.85 to 0.89, and for specificity 0.78 to 0.89.
Mood disorder questionnaire (MDQ) is used in clinical and investigative practice for screening patients with bipolar I and II disorders. Results have shown that MDQ has relatively good sensitivity of 0.73 and specificity of 0.90 in an outpatient psychiatric sample (26).
Hazan and Shaver (27) created adult attachment styles questionnaire to assess attachment styles in adults. In this study, the alpha coefficient for it was .81 for the whole scale and test-retest reliability was 0.78.
Quality of relationship inventory (QRI) is a measure for assessment of support, conflict, and depth in particular relationships (i.e. familial, romantic, intimate and peer relationships) (28, 29). The quality of relationship scale resulted in good internal consistency (support scale = 0.80, conflict scale = 0.88, and depth scale = 0.81) (28).
Multidimensional Anger Inventory (MAI) is designed to assess the following dimensions of anger: frequency, duration, magnitude, range of anger-arousing stimuli, mode of expression, and hostile outlook (30). The MAI proved adequate test-retest reliability (r = 0.75) and moderate to high internal consistency alpha coefficients ranging from .84 to .89 for the 2 samples.
3.3. Procedure
All patients announced their consent before beginning. In order to confirm the psychiatric diagnosis, Borderline Personality Inventory (BPI) was administered to patients who received the diagnosis of BPD and mood disorder questionnaire (MDQ) to patients who received the diagnosis of BD-II. Then, patients were asked to complete the adult attachment styles questionnaire, (QRI) and (MAI). Current study was approved by the ethics committee of Al-Zahra University, Tehran, Iran.
3.4. Statistical Analysis
Independent sample t-test was used to compare the patients with borderline and bipolar-II. The t-test is used for parametric data like the data in our research. The 2 tailed significance level was set at .05. Statistical analyses were performed by SPSS version 20.
4. Results
The BPD patients were similar to the BD-II patients in age, sex ratio, education, history of hospitalization, and marital status (Table 1).
In general, 19 (51.2%) of BD-II patients had secure attachment style and 22 (53.7%) insecure attachment style. The t-test comparing BPD and BD-II patients showed no significant difference in Secure and Anxious attachment styles scores (P > 0.05). However, 2 groups had significant difference in Avoidant attachment styles subscale. [t (76) = 2.21, P < 0.05]. BPD patients showed higher levels of Avoidant attachment styles compared to the BD-II patients (Table 2).
Table 2.T-Test Results for the Borderline Personality Disorder (BPD) and Bipolar II Disorder (BD II) Patients’ Scores on the Attachment Style Questionnaire
| N | Mean ± S.D | T | Sig | df | ||
|---|---|---|---|---|---|---|
| Avoidant | BPD | 37 | 10.13 ± 4.11 | 2.21 | 0.030 | 76 |
| BD_II | 41 | 8.19 ± 3.64 | ||||
| Secure | BPD | 37 | 10.32 ± 4.12 | -1.88 | 0. 063 | 76 |
| BD_II | 41 | 12.04 3.95± | ||||
| Anxious | BPD | 37 | 10.91 ± 4.33 | 1.44 | 0.154 | 76 |
| BD_II | 41 | 9.48 ± 4.42 |
Analysis revealed no significant difference in 7 scales. However, difference in the subscales of «conflict with parents» and «conflict with friends» were significant. BPD patients showed higher levels of «conflict with parents» [t (27) = 1.23, P < 0.05] and higher levels of «conflict with friends» [t (76) = 1.36, P < 0.05] compared to the BD-II patients (Tables 3 and 4).
Table 3.T-Test Results for the Borderline Personality Disorder (BPD) and Bipolar II Disorder (BD II) Patients’ Scores on the QRI
| N | Mean ± SD | T | df | Sig | ||
|---|---|---|---|---|---|---|
| Perceived support from parents | BPD | 37 | 11.40 ± 5.48 | -1.75 | 76 | 0.72 |
| BD_II | 41 | 13.68 ± 5.90 | ||||
| Perceived support from partner | BPD | 10 | 11.30 ± 6.18 | -1.72 | 27 | 0.65 |
| BD_II | 19 | 14.05 ± 5.84 | ||||
| Perceived support from friends | BPD | 37 | 9.86 ± 4.66 | 0.67 | 76 | 0.43 |
| BD_II | 41 | 9.12 ± 5.00 | ||||
| Conflict with parents | BPD | 37 | 24.08 ± 6.41 | 1.93 | 76 | 0.24 |
| BD_II | 41 | 20.95 ± 7.39 | ||||
| Conflict with partner | BPD | 10 | 26.00 ± 3.52 | 1.23 | 27 | 0.02 |
| BD_II | 19 | 23.21 ± 7.46 | ||||
| Conflict with friends | BPD | 37 | 15.27 ± 7.14 | 1.36 | 76 | 0.04 |
| BD_II | 41 | 13.29 ± 5.66 | ||||
| Intimacy with parents | BPD | 37 | 15.43 ± 5.00 | -1.89 | 76 | 0.60 |
| BD_II | 41 | 17.60 | ||||
| Intimacy with partner | BPD | 10 | 15.00 ± 4.92 | 0.83 | 27 | 0.29 |
| BD_II | 19 | 17.15 ± 6.06 | ||||
| Intimacy with friends | BPD | 37 | 12.48 ± 6.66 | 0.75 | 76 | 0.14 |
| BD_II | 41 | 12.39 ± 4.55 |
Table 4.T-Test Results for the Borderline Personality Disorder (BPD) and Bipolar II Disorder (BD II) Patients’ Scores on the MAI
| N | Mean ± SD | T | df | Sig | ||
|---|---|---|---|---|---|---|
| Anger arousal | BPD | 37 | 42.86 ± 13.32 | 2.19 | 76 | 0.03 |
| BD_II | 41 | 37.00 ± 10.25 | ||||
| Anger-eliciting situations | BPD | 37 | 31.75 ± 4.90 | 0. 58 | 76 | 0.56 |
| BD_II | 41 | 33.34 ± 3.97 | ||||
| Hostile outlook | BPD | 37 | 34.70 ± 6.91 | 0.03 | 76 | 0.96 |
| BD_II | 41 | 34.75 ± 5.21 | ||||
| Anger-in | BPD | 37 | 20.32 ± 5.03 | 0.23 | 76 | 0.81 |
| BD_II | 41 | 12.34 ± 4.44 | ||||
| Anger-out | BPD | 37 | 12.05 ± 2.34 | -0.56 | 76 | 0.57 |
| BD_II | 41 | 12.34 ± 2.12 | ||||
| Total score | BPD | 37 | 1.24 ± 21.97 | 0.96 | 76 | 0.33 |
| BD_II | 41 | 1.20 ± 16.79 |
The t-test results on MAI indicated no significant difference in all scales between 2 patient groups except for the «anger arousal» that was statistically higher in BPD patients compared to the BD-II patients [t (76) = 2.19, P < 0.05].
5. Discussion
The present study aimed to examine the proposal that borderline personality disorder is understood as an axis I disorder within the bipolar spectrum.
The results showed that insecure attachment style are more frequent in the BPD as some prior studies had acquired this result (31). In addition, patients with PBD had higher levels of avoidant attachment style. This observation can be due to the core of BPD such as fear of rejection and having unstable feelings in relation to other important people. Conversely, bipolar patients usually do not have these fears and worries of BPD patients that may have resulted to this observation.
Anxious attachment style had the highest frequency in both groups as it can be concluded that patients with BPD and BD-II may have the same attachment experiences. Findings from longitudinal studies (32) on the development of BPD have revealed that traumatic events, especially early negative experiences in parent-child relationship, play a central role in the impairments to the underlying attachment organization and later BPD symptomatology. It appears that the occurrence of these traumatic events early in life leads to vulnerability and development of insecure attachment and later results in bipolar disorder symptoms. For example, results from Levitan and colleagues’ (33) study indicated that there is a strong relationship between childhood physical abuse and mania. Traumatic experiences can deeply disrupt the ability to trust and form secure attachments.
There was no significant difference between patients in 2 groups in regards to the quality of intimate relationship except 2 subscales. Prior researches broadly show that both patient groups possess the same impaired interpersonal processes. These impaired interpersonal processes lead to increased conflicts, intimate relationship impairment, and difficulties in perceived support in meaningful relationships. Alongside, Chen et al. (15) found that BPD symptoms are associated with sustained elevations in partner conflict. In this study, patients with BPD had higher scores on the “Conflict with partner” and “Conflict with friends” subscales of (QRI) compared with bipolar disorder group, although non-significant. As it is discussed in the prior section, BPD is determined with more disrupted and chaotic relational dynamics where bipolar can explain this issue.
Some authors suggested that affective instability and impulsivity might be the important causes of interpersonal dysfunction for patients with PBD (34, 35). In patients with bipolar disorders, as some (36) have proposed, irritability during both the manic and depressive episode can lead to the interpersonal problems and tendency to either despise or idealize others. Based on this study, BPD and bipolar-II patients experience similar intimate relationships. It seems that, these same experiences are the result of the same affective instability.
Findings from our study indicated that the same attachment style and high overlap of quality of intimate relationship and anger experience in the BPD and BD-II patients provide the empirical evidence for placing BPD in bipolar spectrum. Psychiatric researchers such as Akiskal (4, 37) as well as Bradford Reich et al. (8) suggested a common biological and temperamental basis for bipolar disorders and BPD. This biological and temperamental basis is responsible for phenomenological similarities and overlapping symptoms between the 2 disorders. Attachment patterns can be considered a common underlying factor in the development of a variety of disorders. Attachment insecurity in 2 groups generally results in the same experiences in intimate relationships and anger emotions.
We are suggesting that BPD and BD-II due to having the same attachment styles and similar phenomenology in their intimate relationships and anger experiences, maybe considered on a single continuum, namely bipolar spectrum. Reclassification of BPD in bipolar spectrum would result in new clinical and research implications. On the other hand, shifting BPD to bipolar spectrum would require further research on the nature, pathology, and treatment of it. Finally, we affirm and confirm that comparing design of this study is not quite fit to meet this proposal completely, although finding similarities and differences between 2 disorders could be precursors of reclassifying them in 1 category as this research tried to do.
