Bipolar disorder (BD) is a mental illness characterized by recurrent manic, hypomanic and depressive episodes separated by intervals of variable duration with relatively few symptoms (1). Most common subtypes are Bipolar-I disorder (BD-I; depressive and manic episodes) and Bipolar-II disorder (BD-II; depressive and hypomanic episodes only). Overall, depressive symptoms and episodes are more common than (hypo)mania (2–4). A majority of BD patients experience a comorbid condition (5) such as substance abuse (6, 7), anxiety disorder (8), or attention-deficit hyperactivity disorder (9). Co-occurrence of multiple disorders can lead to higher rates of suicidality and less response to pharmacological treatment (10). There is also evidence that the presence of comorbid personality disorder (PD) has an unfavorable impact on the course of BD. Post et al. (11) reported a close relationship between PD as measured by the self-rated 99-item Personality Disorder Questionnaire (PDQ-4) and poor prognosis factors of the course of BD such as history of child abuse, early age of onset, anxiety disorder comorbidity, rapid cycling course, and a high number of previous episodes. A longitudinal prospective study among respondents with a bipolar spectrum disorder (12) found that PD symptom severity predicted shorter time to hypomania and major depression (both particular in case of PD cluster B), and conversion to BD-I disorder (particular in case of PD cluster C) (13). Prevalence of PD in BD is estimated between 30 and 40%, which is significantly higher than in the general population (14). This concerns particularly PD from the B and C cluster, and especially borderline personality disorder (BPD) (15, 16). A systematic review and meta-analysis included 42 papers considering BPD in BD as well as BD in BPD concluded that up to 21.6% of BD samples have comorbid BPD with higher rates in BD-II than in BD-I (17). Although a comprehensive literature review (18) concluded that BD and BPD can be distinguished by comparing their phenomenology, etiology, family history, biological studies, outcome and response to medication in clinical practice the differential diagnosis between BD and BPD can be difficult given shared core characteristics of mood instability, emotional dysregulation and impulsivity (19). A study based on self-report measure showed great impairment in emotional regulation in both BD and BPD, but maladaptive strategies were greatest in comorbid than BPD and BD groups (20). A study by our group confirmed that affective instability, impulsivity, and self-mutilation/suicidality were symptoms associated with both rapid cycling BD and BPD (21). However, BD and BPD require different treatment approaches. Pharmacotherapy is a cornerstone for acute and maintenance treatment of BD (22–26), whereas BPD is primarily treated with psychological interventions. Evidence for the efficacy of psychotherapy in BD is less consistent. A recent review included 35 reports of 28 randomized controlled trials between 1995 and 2013 investigating individual or group psychosocial interventions for adults with BD (27), and found that psychotherapies, in combination with medication, show advantages over medication alone on measures of symptom burden and risk of relapse. Psychotherapeutic interventions that have demonstrated some prophylactic efficacy as an adjunct to medication in BD are individual (28) and group psychoeducation (29), cognitive behavioral therapy (CBT) (30, 31), Family-focused therapy (FFT) (32), and Interpersonal and Social Rhythm Therapy (IPSRT) (33). A meta-analysis including 55 trials with 6010 participants found that there is some evidence that psychological interventions are effective for people with BD, but further research is needed (34). Despite a high prevalence of comorbidity between BD and PD, and evidence that PD traits are associated with an adverse course of BD (11, 15, 35–37), there are no specific treatment recommendations for patients with comorbid PD. Given the overlapping phenomena in BD and BDP, we hypothesized that a treatment that is designed to improve emotion regulation in BPD, such as Systems Training for Emotional Predictability and Problem Solving (STEPPS) (38) could also be effective in BD and improve its overall course.

STEPPS was developed for improving emotion regulation in adult patients with borderline personality disorders (38), but could also be effective for other patient groups. A recent published pilot study (56) explored the effectiveness of STEPSS in adolescents with emotion regulation problems. Given the overlapping characteristics of BD and BPD including the significant impairment in emotion regulation strategies in comorbid BD and BPD (20), and the emotional instability of BD as such, we hypothesized that STEPPS would be suitable for patients with BD and some features of BPD. According to the STEPPS training manual the underlying assumption is that BPD is a disorder of emotion and behavior regulation and STEPPS learns participants specific emotion and behavior management skills. Since the BPF anger and impulsiveness as such are not included in the diagnostic criteria for BD, we considered that BD patients having these features could benefit most from STEPPS. Conversely, mood instability was excluded from our obligatory set of at least three BPF since this is a core symptom of BD. It was further hypothesized that STEPPS would not only improve its primary target of emotion regulation, but thereby would also improve the overall course of BD during and after the treatment. Half of the BD patients in our diagnostic study screened positive on BPD according to PDQ-4+, and half of these had a diagnosis of BPD after semi-structured interviewing. Reported prevalence of BPD in BD varies from 0.5 to 30% for BD-I and 12–23% for BD-II (21). Different studies (17, 57) conclude that ~20% of patients diagnosed with either BD or BPD also met criteria for the other diagnosis. Our study shows that the prevalence of BPD in BD patients depends highly on the instrument used, and that using a screener such as the PDQ4+ will result in many false positives. This confirms other studies that report low agreement between PDQ-4+ and diagnostic interviews (58). The diagnostic part of our study shows that BPF are widespread among patients with BD, and that anger outbursts and impulsivity are common. Since the STEPPS training focuses especially on these traits this supports our assumption that BD patients can befit from the treatment.

Our study was originally designed to evaluate the efficacy STEPPS in BD with BPF in a RCT (39). Due to much lower inclusion than expected the study design was subsequently redesigned as an open explorative study. Although that design and the limited number of participants prevent firm conclusions about the efficacy of the intervention, our findings give an indication that STEPPS may have a positive effect on illness symptoms as well as on patient empowerment and quality of life. This positive effect may be explained by the fact that the three most common BPF in our sample (affective instability, anger outbursts, and impulsivity) are especially targeted in the STEPPS training. Still, as seen often in psychotherapeutic interventions, the effect faded at 6-month follow-up.

Only nine patients completed the treatment part of the study. Several reasons can be hypothesized for this notable gap between the intended sample size and the low inclusion rate. We recruited patients from four different non-academic outpatient clinics. Although we expected a broad interest among clinicians as well as patients for this psychotherapeutic intervention, only 111 patients showed interest in the study and were willing to participate in the first diagnostic part of the study, completing the PDQ-4+, and if screened positive for BPD, receiving a semi-structured diagnostic interview. An important reason for not further participating in the subsequent treatment study, was that a treatment duration of 20 weeks and the additional follow-up of 18 months including continuous mood monitoring was considered too intensive and too long. Furthermore, some participants stated that the association of STEPPS with “borderline” had an aversive effect, which may have caused drop-out before the treatment stage. Moreover, the intervention was intendedly embedded in the regular STEPPS programs of the outpatient clinics, so not exclusively for the patients with BD in this study. Stigma is a well-known phenomenon in patients with BPD. In a study that measured self-stigma by self-report questionnaires, women with BPD showed higher rates of self-stigma than women with social phobia (59). Also perception of health professionals is laden with stigma toward diagnosis of BPD (60–62). One study showed that the description of a patient with diagnosis of BPD attracted more negative response from a sample of qualified mental health nursing staff compared with descriptions of patients with a diagnosis of depression or schizophrenia (63). Sympathy and optimism was rated less toward patients with a diagnosis of BPD, and personal experience with patients with BPD was rated as more negative than experience with patients with diagnosis of depression or schizophrenia (63). In line with these findings, negative stigma toward BPD could explain our disappointing inclusion in the treatment part of the study.

Our study raises the question whether a treatment primarily developed for one diagnostic group (such as personality disorder) can be applied to another patient population (such as mood disorder), even with a rationale as described. We assumed that embedding the study protocol in ongoing STEPPS programs would increase the feasibility of this treatment option in routine clinical practice, but this may have been counterproductive. In case of STEPPS, primarily designed for BPD patients, we therefore recommend to adapt the training manual and materials for participants before being applied to BD patients, and forming a group specifically for BD. Avoiding the term “borderline,” one could describe that BD includes problems with emotion regulation and that these are the main focus of STEPPS. In the context of research this would imply that inclusion should not be based on “borderline personality features” but instead focus on emotion regulation.

Our study has several limitations. First, the sample size of the treatment part was small for reasons explained above. Second, even if more patients had participated in the adapted study design, the methodological nature of this open non-controlled study prohibits conclusions of causal relations between the treatment and possible effects, since only a RCT can show the effectiveness of STEPPS in BD.

Despite these limitations, we conclude that a modified version of STEPPS could be a promising addition to the evolving field of psychological treatments for patients with BD.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

This study was reviewed and approved by the Ethics Committee of the VU University Medical Center in Amsterdam, The Netherlands (registration number: 2012/470). It is registered in www.trialregister.nl (trial ID: NTR4016). The patients/participants provided their written informed consent to participate in this study.

GR, NW, ER, MC, and RK: conception and design, and/or acquisition of data, and/or analysis and interpretation of data, drafting the article or revising it critically for important intellectual content, and give final approval of the version to be submitted and any revised version. All authors contributed to the article and approved the submitted version.

This research was supported by the Netherlands Organization for Scientific Research (NWO). The supporters had no role in the design, analysis, interpretation, or publication of this study.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.