“If you don’t have a word for something, you may doubt whether it’s even real” – how individuals with borderline personality disorder experience change

Abstract

Objective:

This study explored how psychological change was experienced and what treatment-related factors or events were perceived as supporting or hindering their process by individuals with borderline personality disorder.

Methods:

Eight BPD sufferers attended a 40-session psychoeducational group intervention at a community mental health care center. At intervention end, personal experience of meaningful change was explored in an in-depth interview and data were content-analyzed. Change in BPD symptoms was assessed by the Borderline Personality Disorder Severity Index IV interview.

Results:

The qualitative content analysis on subjectively perceived meaningful change yielded three core categories: (1) improved ability to observe and understand mental events, (2) decreased disconnection from emotions, emergence of new or adaptive emotional reactions and decrease in maladaptive ones, and (3) a new, more adaptive experience of self and agency. Accordingly, (1) learning and (2) normalizing emerged as the main categories of helpful treatment factors. In turn, treatment-related factors perceived as obstacles were: (1) aggression in the group, and (2) inflexibility. With respect to symptom change, four participants were considered clinically as remitted, and two showed a reliable change.

Conclusions:

Long-term psychoeducational group therapy seems to enhance mentalization / metacognitive functioning and promote self (or personality) integration in BPD patients.

Keywords:

• cognitive behavior therapy
• group psychotherapy
• integrative treatment models
• personality disorders
• qualitative research methods
• process research

Clinical or methodological significance of this article: Acquisition of conceptual knowledge seems to facilitate self-observation in BPD sufferers. It was found that learning about BPD can aid in making sense of and organizing of experiences. Psychoeducation might provide the initial impetus that activates deep cognitive-emotional processing. Groups can have unique benefits such as providing opportunities for collaboration in learning and normalizing participants’ sense of self.

Introduction

The last 30 years has seen enormous progress in the treatment of borderline personality disorder (hereafter BPD). Research has shown that symptoms of BPD are treatable, primarily by psychological, psychosocial, and relational approaches (Choi-Kain et al., 2017). Highly specialized treatment programs have been developed and tested in randomized controlled trials (Cristea et al., 2017). Today, dialectical-behavior therapy (DBT), mentalization-based therapy (MBT), schema therapy (ST) and transference-focused psychotherapy (TFP) are the established “big four” evidence-based treatments for BPD.

In addition to treatment trials, data from two well-designed naturalistic prospective longitudinal studies indicate a high symptom remission rate (Gunderson et al., 2011; Zanarini, Temes, et al., 2018). However, a disparity exists between symptomatic and functional improvement in BPD. Soloff (2019) hypothesizes that one reason for the significantly lower rates of psychosocial recovery compared to the high rates of diagnostic remission may be that the standardized quantitative reports of diagnostic remissions do not fully capture the clinical reality of BPD. Gunderson et al. (2018) compared four different theories underlying therapies for BPD: emotional dysregulation (DBT), mentalization failure (MBT), excessive aggression (TFP), and interpersonal hypersensitivity (good psychiatric management; Gunderson & Links, 2014). Karterud and Kongerslev (2020) added insecure attachment to this list, and also postulated that these features are dynamically intertwined (Karterud & Kongerslev, 2020). Although schema therapy was not included in the comparison by Gunderson et al. (2018), it has been suggested that insecure attachment and deprivation regarding the child’s emotional needs also underlie BPD (Young et al., 2003). Thus, viewed from the perspective of personality development and integration, the relative slowness of functional improvement is not surprising.

Moreover, the first-person perspective of clients themselves may usefully expand the existing framework by furthering our understanding of the therapy process and illuminating processes of which therapists may be unaware. A recent major advance in this domain was the publication of a meta-synthesis of 14 qualitative studies exploring clients’ experiences of their treatment for BPD and their recovery perceptions (Katsakou & Pistrang, 2018), which concluded that clients make changes in four main areas: developing self-acceptance and self-confidence; controlling difficult thoughts and emotions; practicing new ways of relating to others; and implementing practical changes and developing hope. Clients experienced change as an open-ended journey, a dynamic and gradual process that consisted of small steps, including setbacks as well as achievements.

However, little is known about how improvement is achieved. The processes and specific mechanisms through which treatment characteristics facilitate or promote change remain poorly understood (Katsakou & Pistrang, 2018; Silberschatz, 2017). To understand what works for whom, and how and under what circumstances is only possible at the level of the individual patient. The identification of critical processes of change is likely to be clinically relevant, since it can help therapists recognize and foster unique opportunities for patient change as these occur during psychotherapy (Elliott, 1983). The examination of such events provides a direct window into what can, in the eyes of the therapy participants, facilitate or interfere with change; this in turn may lead to a better understanding and, ultimately, improvement in psychotherapy (Castonguay et al., 2010).

Katsakou and Pistrang (2018) also conclude that although the studies included in their meta-synthesis identified areas where people with a diagnosis of BPD made progress, they provided little information about how those improvements were reached. They suggest that more detailed accounts of change processes are needed in order to provide rich and nuanced descriptions of how therapeutic change occurs. The present study responded this call by conducting in-depth interviews with BPD sufferers who had attended a 40-session cognitive therapy group intervention.

Study Aims

The aim of this study was to explore participants’ subjective experience of meaningful development and change and how they experienced the present intervention and events in therapy. A secondary aim was to investigate change in BPD symptoms. The research questions were: How do participants with BPD perceive meaningful change in themselves after attending a long-term psychoeducational group intervention? If they experience change in themselves, what processes do they highlight? Alternatively, how do they describe the lack of change? How do patients experience the intervention as a group intervention? What elements of the intervention or events during the intervention do they find helpful or unhelpful?

Method

Study Design

This process-outcome study was conducted in community mental health care services in the City of Jyväskylä, Central Finland. Applying a mixed methods research design, the qualitative component of the study aimed, through interviews, to trace and describe patients’ first-person experiences of meaningful development and change and how they experienced the effect of different treatment factors on their change process. The quantitative component assessed change in BPD symptom scores at the end of the 40-session psychoeducational group intervention.

Qualitative content analysis was the method chosen to explore subjective experiences owing to its data sensitivity, i.e., it allows the relevant themes to emerge from the data and is thus suitable for the study of idiographic experiences (Kyngäs et al., 2020). Change in BPD symptom severity was measured by the BPDSI-IV (Borderline Personality Disorder Severity Index-IV) interview. The assessments were conducted between June 2017 and October 2018 at the community mental health care center in Jyväskylä.

Recruitment and Setting

Participants were recruited from the community mental health care outpatient services of the City of Jyväskylä, which despite its name, forms part of the municipality’s secondary, specialized psychiatric services. Professionals working in community mental health care outpatient services were approached, informed about the study, and asked to refer patients aged 18–65 years with BPD symptoms for potential recruitment. The study design was naturalistic. Professionals, as part of their routine work, informed patients with BPD diagnosis about the possibility to participate in the present study. The intervention that was part of the study was one that is routinely offered for BPD patients being treated at the community mental health care center and was not controlled for in the study. Hence, patients were simultaneously recruited for the study and the group treatment. Potential participants were assessed in order of referral. Since, owing to financial constraints, only one treatment group could be studied, recruitment ceased when the number of eligible patients reached eight.

The inclusion criterion was the Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM–5) diagnosis of BPD. Exclusion criteria were a DSM–5 diagnosis of a psychotic disorder or a substance abuse disorder necessitating detoxification prior to treatment. Exclusion criteria were assessed only clinically; no other structured evaluations were performed. The referred patients were assessed for eligibility using the Finnish version of the Borderline Personality Disorder Severity Index IV interview (BPDSI-IV). No other diagnostic evaluations were performed. Based on the eligibility interview, one patient with a primary diagnosis other than BPD (post-traumatic stress disorder) was excluded. All the eligibility assessments were performed by the first author (MK), except one (performed by SL).

Treatment

Group Intervention

The intervention, which consisted of 40 weekly 2-hour psychoeducational group sessions conducted from August 2017 to June 2018, was originally developed in Northern Finland to meet the needs of public mental health services (Oulu BPD model; Leppänen et al., 2016). The group was led by two experienced psychiatric nurses who delivered the treatment as part of their routine work at the community outpatient mental health care center. The framework integrates elements drawn from cognitive and behavioral treatment models designed to treat BPD. One of the main components of the intervention is patient education in schema therapy using the concept of schema modes (Online Supplement 1). A table delineating the content of the sessions is presented in Online supplement 3.

Adjuvant Treatment

In addition to group treatment, all patients continued their pre-existing treatment as usual (e.g., weekly individual sessions with their psychologists or psychiatric nurses as well as medication) at the community mental health care center. This treatment would, if needed, also continue after intervention end. Adjuvant treatment was not controlled for in the present study, and hence it was not integrated or coordinated with the group intervention. While some of the individual therapists were familiar with BPD treatment or with the Oulu BPD model, others were not.

Participants

Seven of the eight patients included in the study were female. Patients were aged 23–42 (mean 30, median 26) at study start. At baseline, the participants’ average BPDSI-IV (Borderline Personality Disorder Severity Index IV) score was 31.1 indicating moderate to severe symptoms. On average, the participants suffered from marked functional impairment, as reflected in the fact that only two were working or studying at entry into the study. One patient was attending a work try-out as occupational rehabilitation and five were receiving disability payments. No structural assessment of functioning was performed.

Researchers

All the present authors are psychiatrists and cognitive-integrative psychotherapists specialized in the treatment of BPD. TM has a PhD and is also a psychodynamic psychotherapist. SL is a professor in psychiatry and one of the developers of the intervention, while MK and TM had nothing to do with the development of the intervention or the organization that delivered the treatment. MK conducted all the interviews except for one eligibility interview, which was conducted by SL. MK and TM analyzed the data and had no communication with the treatment providers.

Data Collection Method

All the interviews were carried out at the community mental health care center. The in-depth interviews exploring participants’ first-person experiences and the BPDSI-IV interviews measuring symptom change were conducted soon after intervention end. These interviews were executed in close succession, the BPDSI-IV immediately after the in-depth interview. As the same interviewer conducted all the interviews, there could be no blinding during the data collection. All eight participants (100%) were interviewed, and all the in-depth interviews were videotaped.

The BPDSI-IV interviews were conducted twice: pre- and post-intervention. All the BPDSI-IV interviews conducted to ensure eligibility were audiotaped except for one, which was due to technical error. The post-intervention BPDSI-IV interviews were videotaped.

In-depth Interview

The major part of the data consists of responses to a semi-structured in-depth interview in which patients were asked to reflect on their experience of personal development or meaningful change (or lack of it) over the past year during the group intervention. The questions used in the in-depth interview are presented in Online supplement 2. In this study, we were interested in responses to questions 1-4, i.e., to the questions that asked participants to reflect on their experience of personal development or meaningful change over the past year during the group intervention and the contribution of treatment-related factors or events to their change process.

BPDSI-IV Interview

The BPDSI-IV (Borderline Personality Disorder Severity Index-IV) interview scores comprise the quantitative data. The BPDSI-IV is a clinical interview assessing the frequency and severity of BPD symptoms during the previous three months. The purpose is to provide a quantitative index of current symptom severity. The BPDSI-IV is based on the DSM criteria for BPD and consists of 70 items organized into nine subscales: (1) abandonment, (2) unstable relationships, (3) identity disturbance, (4) impulsivity, (5) parasuicidality, suicide plans and attempts, (6) affective instability, (7) emptiness, (8) outbursts of anger, and (9) paranoid ideation and dissociative symptoms. The frequency of occurrence of each item over the previous three months is rated on an 11-point scale from 0 (never) to daily (10). Answers are then scored from never (0 point) to daily (10 points) or rated on a 5-point Likert scale (Giesen-Bloo et al., 2010; Leppänen et al., 2013). Previous research has found a cut-off score of 15 between patients with BPD and controls, with a specificity of 0.97 and a sensitivity of 1.00 (Giesen-Bloo et al., 2010). Recovery is defined as achieving a BPDSI-IV score of less than 15. Reliable change, which reflects individual clinically significant improvement, is achieved when the improvement is at least 11.7 points (Nadort et al., 2009). No systematic measurements other than the BPDSI-IV interviews were performed.

Data Analysis

Qualitative content analysis was applied to the in-depth interview data. This methodological approach was chosen because it allows both inductive and deductive classification and interpretation of data (Kyngäs et al., 2020). The preliminary approach to the data was inductive. MK immersed herself in the videotaped interview data and transcribed the interviews verbatim. The parts of the text that covered responses to questions 1–4 were extracted and compiled into a single text. This text, which forms the unit of analysis, was then divided into meaning units, e.g., words, sentences or paragraphs that describe a single idea. The meaning units were then condensed, and the condensed meaning units were abstracted and coded. All the text fragments from subsequent interviews sharing the same meaning were then assigned the same code. Units were then clustered based on their shared meaning, to form larger categories.

MK frequently went back over the transcribed in-depth interview data. MK and TM reviewed 80% of the videotaped in-depth interview data together and discussed and revised the preliminary codes and clustering decisions made by the first author. SL read the transcribed data and negotiated the clustering.

The BPDSI-IV was scored soon after the interview. As they were not blinded, the researchers were aware of the participants’ remission status when analyzing the qualitative data.

Reflexivity

In qualitative research, researchers are interpreters of basically ambiguous human experience (Binder et al., 2012). In trying to understand research participants’ experiences, some reconstruction of meaning is necessary and unavoidable (Morken et al., 2019a). Researchers’ pre-assumptions inevitably have some influence on the findings in qualitative research meaning that the results of phenomenological exploration are co-created.

The exploration phase in qualitative research is often carried out in an interview context where experiences are recalled and relived in an interpersonal situation. The interview is thus much more than a data-gathering method, reflection on the interview relationship being an essential part of the research process, as the quality of this relationship determines what parts of the participant’s experience become accessible and what parts remain unarticulated (Binder et al., 2012). For example, in the exploration phase of the current study, the interviewer noticed how subtle signals on her part influenced the interviewees. If, for example, her response was delayed due to a focus on note taking, some interviewees might start second-guessing their experience or even shut down. She also noticed that to be able to reflect upon their experiences and deepen their descriptions, some participants needed a lot of validation. She was concerned that, by so doing, she might be in danger of facilitating talk about what interested her personally and thereby disproportionately intrude her own mindset on the interview. She therefore sought to adopt the stance of a benevolent follower who would, nevertheless, structure the interview.

In the data abstraction phase, we noticed a major tension between our desire to remain close to the participants’ lived experience while in part interpreting this by applying the theory and language of psychotherapy. We provide excerpts from the data both to increase transparency and help the reader follow and evaluate our reasoning.

Ethics

This study was approved by the ethics committee of the Central Finland Health Care District on 9th May 2017. All participants provided a written informed consent after receiving a full description of the study procedure.

Results

In this study, the quantitative part focused on change in BPD symptoms while the qualitative part explored participants’ subjective experience of meaningful change and their views on how different treatment factors and events were related to their individual processes.

I. BPD symptoms

Four patients (50%) were considered remitted based on their BPDSI-IV interview, conducted soon after the intervention had ended. Two patients (25%) were considered to have experienced a reliable change and two (25%) to have remained unchanged.

II. Subjective experience of meaningful change

Based on the in-depth interviews, a total of 22 areas of change were initially coded. These codes were grouped into larger categories based on their shared characteristics. Three main areas of change were identified: (1) improved ability to observe and understand mental events in oneself and others, (2) decreased disconnection from emotions, and emergence of new, adaptive emotional reactions and decrease in maladaptive ones, and (3) a new, more adaptive experience of self and agency.

1. Improved Ability to Observe and Understand Mental Events in Oneself and Others

All eight participants, regardless of their remission status, reported having experienced changes in their ability to perceive their experiences with increased accuracy in the present moment and to make sense of them. Most participants perceived this emerging ability to obtain a meta-perspective, i.e., being able to “mentalize” what happens in the moment at hand, either in their own mind or in relation to others, as a major and very meaningful change. Importantly, this improved ability was often informed by a kind and compassionate tone:

Íve become kind of very mindful of what I’m feeling and why is it that I’m feeling that way … I learned like to relate to myself in a way … like I would relate to a little child or baby, like why are you feeling bad like are you hungry? Or do you feel some discomfort or are you hurting and so on? Are you tired and so on? … Like you become mindful of this kind of stuff … That was like really, really, like really helpful. (becomes moved).

When able to understand mental events as representations as opposed to absolute truths, participants became less incapacitated by them and better able to engage in functional behavior:

What was crucial for me in the group, was to get a grasp of ‘I am not a valid person’ which has affected every aspect of my life. Previously, I wasn’t able to set any goals because it was a fact for me back then when the group began … During the course of the group, I got a steadily growing sense that this might not be a truth or that it might be a distorted view.

One patient felt that learning to deliberately focus her awareness on the present moment had been curative for her. This ability had ended her longstanding suicidal ideation and urges that had resulted in a serious suicide attempt a few months before the group started. For her, the improved ability to be present also seemed to serve as a basis for better understanding herself and others.

Four participants described how their relationships with partners and friends had improved. For example, improved mentalizing helped patients to understand that each person has his or her own thoughts and feelings, and thus they became more capable of allowing others to freely experience their own thoughts and feelings. They were able in more nuanced ways to interpret situations where they sensed some disagreement. This, in turn, diminished their formerly strong sense of threat and desperation.

… realization that erm every person has those (schemas, modes, and coping strategies), like everything that works for me can be applied to others, too. My relationships with others have improved, that might be one (of the most meaningful changes).

For this patient, this kind of new understanding translated into more functional relationships. For example, if he realized that a friend of his was angry, he could choose to pause, postpone his own reaction and reflect on how it would be wise to respond in contrast to unmodulated knee-jerk responses that typically made things worse.

Half of the patients reported having gained new understanding of themselves in terms of their past. Being able to mentalize how their ways of experiencing, coping strategies, or symptoms had evolved helped them question their previous learning, which was no longer viewed as the only truth but understood as an imprint of their life experiences. Understanding how the past had affected oneself was associated with increased self-compassion, sense of agency and hope. For example, if the schema ´Ím not a valid persoń is learned, it can also be unlearned.

The capacity to regulate emotions was associated with improved ability to observe and understand mental events. All eight patients described how they had become more able to modulate their own behavior to match the requirements of the situations they encountered in their daily lives. Many patients described how their developing ability to stop and take a step back and to adopt an observer perspective towards one’s mental states resulted in a better ability to regulate distress. As a result of this kind of intentional awareness, emotions did not last as long as they previously did. Patients also reported improved capacity to resist acting on emotional urges and that they did not lose their ability to function even when emotional.

You don’t feel the need to do something when you are anxious nearly as often as earlier. You can just stop and analyze it a bit and then it doesn’t last that long any more … And I feel that it might not be that crippling as it used to be. I can do things even if I’m anxious, like my whole life doesn’t fall apart anymore.

2. Decreased Disconnection from Emotions, and Emergence of new, Adaptive Emotional Reactions and Decrease in Maladaptive Ones

The content analysis yielded a second core category, the theme of which was the processing of emotions. Seven out of eight participants, i.e., all but one who remained unchanged according to the BPDSI-IV, described an increased ability to stay connected to their emotions without having to cut them off. In addition, participants reported new, adaptive emotions and a decrease in less adaptive emotions.

a) Decreased disconnection from emotions

Five patients described how experiential avoidance had been one of their main regulatory strategies for as long as they could remember. They also reported that their awareness of the various strategies they used to disconnect from their mental contents had increased. Now that they were willing to attempt to establish more contact with their emotional experiences, they were working to implement change in the conscious parts of their avoidance strategies. The resulting emerging ability to feel more and feel oneself to be authentic was welcomed, although some participants also described feeling acutely sensitive and aggrieved when trying to allow themselves to experience what was in their minds. One patient also described how she had lost her previous level of functioning when trying to stay in contact with her experiences, as she did not yet have the skills to deal with overwhelming emotions.

Besides detaching from emotions, another strategy that had buffered participants against painful emotions was angry protection:

I feel this is new in me: I can apologize for doing something and I dare to make an apology and I now dare to really admit ‘I was wrong’.

This patient offered examples of how the ability to allow oneself to be more vulnerable in relationships and a decreased need to defend oneself had exerted a positive influence on her relationships.

b) Emergence of new, adaptive emotions and decrease in maladaptive emotions

Learning about the development of BPD and further elaborating on the theme during group discussions triggered memories and emotions.

What I’ve noticed is that previously I kind of had very few memories of my childhood but now that I’ve recalled the bad stuff, I also have recollection of some nice memories … Yesterday, when I was at my boyfriend’s place, he asked if he could have a closer look at my teddy bear … and then I remembered somehow very vividly the moment when I was buying that teddy bear when I was ten … It was interesting, ‘cos somehow, previously I haveńt been able to recall almost anything at all (of childhood) … It’s nice to notice how my whole childhood kind of erm opened up my memory to more … at first, bad things and then good memories, too.

Participants described sorrow over what had been missing in their lives and what was still missing, and adaptive anger towards those who had exploited them or failed to meet their needs. They also described decrease in self-hatred, guilt, and shame:

The most crucial thing was that this is not my fault, that it makes sense I have this (BPD), there’s a logical reason for this. I’ve done what I’ve done, my behavior has been what it’s been, it makes sense. ‘Cos I’ve always wondered why I behave like this. Why do I fuck everything up, why do I do reckless things … I kind of compensated for what I had to endure myself back then (in childhood) … Now that my awareness has increased, now that I know why I behaved like I did, it also means I can forgive myself.

Self-compassion and pride were mentioned as new emotions. Sometimes waning of the internalized punitive self-concept enabled emergence of these new emotions. However, self-compassion increased even in those who had not, at least explicitly, suffered from harsh self-criticism.

Hopelessness is a very prevalent secondary emotion in BPD. Three patients explicitly mentioned the emergence of a new counteracting feeling, hope. While others did not explicitly name hope as a discrete emotion, a decrease in hopelessness or an increase in hope was indirectly evident in their accounts, e.g., in how they expressed their new willingness to live, in the cessation of their previously unrelenting suicidal ideation, in their increased trust in their own competence to deal with daily hassles or in their ability to plan for the future.

3. A new, More Adaptive Experience of Self and Agency

The content analysis yielded a third core category that reflected change in the experience of self. Importantly, only patients who were classified as either remitted or having achieved reliable change described experiences in this category.

a) Attenuation in internalized harshness and emergence of one’s own voice

Three patients described attenuation in the harsh way of relating to oneself. One of them reported this to be among the most important experiences in her change process. As one example, she described a recent moment in a fitting room when she noticed that her previous self-berating attitude was no longer coloring her self-observation:

I didn’t have thoughts like “ugh, how ugly you are, you’re no use to anybody, your belly is ugly, ugh” … instead … I felt sorrow … Maybe for the first time in my life, I saw myself in the mirror as I really am or I think I saw myself as I really am … And somehow, I have a growing sense of … like I have to take better care of myself. For example, I smoke and I genuinely now have a constantly growing sense that, for example, I have to quit smoking, that it’s bad for me … Íve always known that it’s bad for me but now there is a sense that it’s no good to me and I can decide whether I do this for myself.

In addition to enabling more realistic self-observation, change in the harsh self-concept resulted in feelings of sorrow, self-compassion and need for self-care. Throughout the interview, this patient’s narrative suggested that the waning of internalized punitiveness had played an essential role in enabling her self-actualization:

I’m not drifting any more … Previously, I didn’t feel I was able to choose, I didn’t have the possibility to choose, to make decisions concerning me … Gee, I suddenly realized … for example, I can start studying. I can become something if I work towards that. I can do that because I’m adequate, I’m valid … ’cos I’m worth it. And one thing: I can do it even if it all went wrong … I don’t need to succeed … It’s ok to fail. ‘Cos previously, failing at something was like < makes a gesture of cutting her throat >.

Some descriptions revealed how the change in internalized harshness was related to positive change in the other-oriented, self-forgetful, or submissive coping strategy:

Previously, in my life, I didn’t actually do anything I myself kind of wanted to do … I always kind of tried to understand what the other person like wants to do and for example <laughter > I made food only if someone else was also going to eat it and then I did it in just the way the other person wanted it done and then I might even make something I didn’t even like but it was the right thing to do ‘cos the other person preferred it … I’d like learn to relate to people in a completely new way … like somehow genuinely and … I feel I’ve begun to hear my own voice that I somehow I haven’t heard earlier what I’d like and so on kind of erm I feel this (the group) gave me a good start for something like kind of being able to build the kind of life I would like to have.

b) Self as continuous and existent

One patient described how the ability to identify her own opinions, interests and preferences was related to her becoming more consistent and stable across different situations. Previously, she had experienced a weak sense of self that even made it difficult for her to be alone. She described how her previous need to accede to other people’s wants and opinions changed as she grew stronger:

I’m able to stick to my own … or I’m able to identify what is like my own opinion and … I’m not like a chameleon any more, at all, that has almost completely ceased … There was a time, when I didn’t know at all what I like and because of that, I was unable to be alone ‘cos if I was alone, there was no mirror. There was no possibility to mirror what I might be interested in or what I should do at the moment … That was quite bad earlier … In a way, I feel like I exist or I am able to see the future, too. There’s not only this moment that’s going on right now … that’s quite a major insight.

For her, the ability to identify and validate her own perceptions seemed pertinent to the process where the self became more stable and continuous.

III. Treatment factors and events perceived as helpful or hindering

We also asked participants to describe what treatment-related factors they had experienced as either facilitating or hindering their process. The qualitative content analysis yielded two main categories of helpful factors: (1) learning and (2) normalizing. Accordingly, two main categories that were perceived as hindrances were found: (1) aggression in the group and (2) inflexibility of the treatment. In some instances, participants also described how different factors or events had affected them. Thus, when the data allowed, we sketched links between treatment-related factors and areas of change.

Treatment Factors or Events Perceived as Helpful

1. Learning

Acquiring information, especially about the development of BPD and mental states, the so-called schema modes typically encountered in BPD, but also about interpersonal cycles and various skills was regarded as helpful. The concepts facilitated self-observation, expressed in utterances such as “This is the voice of the Punitive authority mode”, “Now I’m acting from the Compliant surrender mode”, “I want to activate my observer self” or “Are there factors that increase my vulnerability to emotions right now?” Conceptual knowledge aided in making sense of and organizing experiences that were elusive: “If you dońt have a word for something, you may doubt whether it´s even real”. It helped patients when they aimed to take a healthy distance from their mental states and regulate them.

In my opinion, it’s indeed information that is the curative thing here … because it helps me to get a clearer sense of my experiences. Organizing is the correct word for how the group affected me. I can organize things in my mind, that’s the point … When I realized that there are different self-states, I realized that the voice of the Punitive parent isn’t my own. It´s her voice. I realized I don’t need to listen to it anymore. It was awareness of that self-state that enabled me to expel it … to literally push it at arm’s length (demonstrates with a gesture) … It’s this distance that enables me to feel that it’s outside of me, it isn’t part of me anymore. I can disagree with it, but what’s even better, I can engage in a dialogue with it … But in order to change, in order to be able to put it at arm’s length, you first need to do a lot of ground work, realize many things (gestures towards the group workbook).

Obtaining information on BPD and various skills also resulted in understanding that BPD is a treatable condition and that one can be an active agent in the change process. This seemed to be associated with an increased sense of self-efficacy and hope: “I can learn, I can practice, I can recover”. Learning about the development of BPD triggered early memories and various emotional reactions described under the rubric Emergence of new, adaptive emotions and decrease in maladaptive emotions. Thus, learning was not merely a cognitive or a passive process where the patient was simply receiving information. Rather, it seemed that psychoeducation provided the initial impetus that activated cognitive-emotional processing, and that patients were active in processing new information. They were also very eager to learn from and with peers. For example, participants highly valued the detailed analyses of problem situations conducted in the group and collaboration in learning how challenging situations or emotions could best be approached, handled, and endured:

… if somebody had had some hassles, we figured out what had triggered the situation and what the factors underlying it were and then we kind of dissected the situation and figured out how it could be approached normally.

2. Normalizing

It seemed that the conceptualizations offered in the group were experienced as normalizing and kind. This seemed to set the tone for more compassionate self-observation. Besides information acquisition, peer experiences were commonly mentioned among the treatment factors or events perceived as helpful. Participants learned that others also experience difficult emotions, are sensitive or hypervigilant in interpersonal contexts, may distort information when emotional, etc. Being able to share, connect with and be understood by others seemed crucial:

It’s such a strong feeling when you realize that first, you’re not alone and then, that someone else has also been through that … I’m not defective like I used to believe, I’m not too weak … If these folks have experienced the same thing, my experience gets … validated and it becomes true … It’s not that I’ve just imagined it, it’s not that Íve just aggrandized everything in my mind … Ím not completely crazy as I used to believe.

This patient identified the above-described event, where she had shared her experience of annihilation of the self in the group and found that two peers were able to relate to it, as one of the most powerful episodes in her change process. Finding out that others were like oneself was healing:

Others’ experiences (had a major influence) … We had many similarities … very many and … for example, when someone told us that she had got a job or something like that, I started to feel like “I may be able to start working as well” … I’ve been receiving disability payments for about five years or at least four and a half years now … I started to think that I might also be able to do that (start working) … You start to feel like normal, you don’t feel yoúre like … kind of, some kind of problem case, in any way … You feel that it’s normal to have feelings like that from time to time … The group actually had a major role in that I’ve now been able … This fall, I’ll begin a work try-out and next fall, I’m determined to start studying.

Peer experiences seemed to help participants to validate and normalize not only their emotions, but their whole self. This could translate into agency. As one’s self-concept became increasingly positive and patients no longer perceived themselves as fatally flawed, they were able to act in a new way.

Treatment-related Factors or Events Perceived as Hindering Change

1. Aggression in the Group

Besides being beneficial, peer experiences were also experienced as the most important hindrance to development and change, inducing hurt or even harm. Half of the patients reported being troubled by aggression expressed in the group. Two patients who described experiencing a particularly strong reaction to their peers’ behavior, reflected on how aggression reminded them of their own former aggressive behavior. In addition, for some participants, displays of aggression in the group triggered early traumatic memories, hypervigilant scanning for potential aggression in others, and strong avoidance reactions:

I suppose one instance was enough for me … It cut too deep inside and made me recall my time at junior high school … Exactly the same feelings surged, directly from the time at junior high … so I felt I won’t take this anymore, this issue is concluded now … I felt insecure, and that made me retreat into my shell, like into that Protector mode … like “I dońt really have anything to say” … If I’m in that Protector mode, I feel “ok, this was enough, I can leave right away” … I feel I can’t get anything out of this (group treatment) anymore … In fact, if there was some argument, I kind of avoided it because re-opening my wounds doesn’t aid me in healing myself, but rather drags or actually dragged me down again.

Importantly, the participants who reported being most disturbed by aggression in the group also reported no benefit from the treatment in the BPDSI-IV interview.

One member’s behavior was sometimes experienced as aggressive by some but not all group members. This participant differed from the others in that she was emotionally more constrained and more prone to rely both on rational processing and on overcompensation strategies. She pondered whether, due to these qualities, she might have appeared to others as having no problems. In return, the group pressured her to express more vulnerability and to open up more than was possible for her at the time. However, she recognized that her holding back was at least partly fear-based:

For me, it (change) means that I analyze my thoughts and, and like cognitively process them … I’m not sure whether I’d ever been ready to go there (to sharing emotions at the expressional level in the group) … Be it any situation, if I broke down there in the group, if I couldn’t cope with talking about something or the like, that wouldn’t bring me any further, rather backfire … One thing I very often brought up there (in the group) was that I don’t want to lose control because I’m not able to tell whether I am the person who cries gracefully or whether I’m someone who shrieks and goes red in the face. I can’t tell, and I somehow don’t want to know.

This patient, too, was left with the feeling of not being understood and not able to connect with others. Consequently, she resorted to an even stronger intellectualization and distancing of emotions in the group context.

To conclude, the participants in this group remained insufficiently encouraged to engage in the further exploration and management of aggression. This outcome seemed to strengthen their reliance on their old coping strategies. The feeling that they had to protect themselves or overcompensate for their vulnerabilities seemed to block their sharing of their underlying adaptive emotions or needs.

2. Inflexibility of the Treatment

Some participants experienced the wordings of the mindfulness exercises practiced at the beginning and end of each session as aversive. They had previously attended another group where the exercises varied from session to session and also wanted to discuss this possibility in the present group. However, the original wordings were retained, which left some group members with the feeling that genuine negotiation was not possible. For one participant, the most difficult aspect of this situation was her feeling that the other participants’ wishes were valued more highly than hers. She felt that the group leaders sided with those in the group who wanted to limit the amount of time spent on discussing potential revisions of wording. For her, this episode resulted in a rupture in the therapeutic alliance that was never fully repaired. She described the strengthening of her old coping strategy, namely, acting compliant while hiding her true feelings such as disappointment and anger:

At first, my approach was, that I’ll try to be myself in the group (but then) I noticed that some stuff (in the group) was allowed here while some was not … I was able to present an edited version of myself in good time so that I was accepted (compares herself to the above-mentioned group member) … I made use of the same coping strategy I had used at home … I’m able to behave in a way that I can survive … It has always been easiest not to be myself, but to behave as others want me to behave.

Discussion

In this study, we explored (1) what factors BPD sufferers themselves considered meaningful in their process of personal development and change and (2) what treatment-related factors or events contributed to or hindered this process. We contrasted these narratives with their medical recovery status as assessed by the BPDSI-IV.

The main area in which participants perceived themselves to have made progress, was in their improved ability to monitor and understand mental events in oneself and others. Second, participants reported increased ability to be in contact with their own emotions as a major positive change. We also noticed a decrease in maladaptive emotions such as unjustified guilt, shame and hopelessness and the emergence of new, adaptive emotions such as hope, pride and self-compassion. Third, only patients who had experienced change in their BPD symptoms (either reliable change or remission) described changes that could be understood as reflecting a more adaptive self-experience. Most importantly, the waning of harsh or punitive internalizations and the ability to identify and validate one’s opinions and preferences seemed a meaningful part of their more adaptive self-experience. Fourth, we observed complex inter-relationships between meaningful changes.

To illustrate the inter-relatedness of meaningful change, participants had first, for example, to adopt a meta-perspective towards, or to be able to mentalize the internal voice that tells “you are not a valid person”, since in the state of psychic equivalence (Bateman & Fonagy, 2012) “you are not a valid person” is reality and no alternative perspectives are possible. In other words, participants had to be able to distinguish between truths and mental representations. Importantly, this improved metacognitive awareness of mental states or the ability to explicitly mentalize was associated with a more compassionate attitude towards oneself. It also seemed that enhanced mentalization in combination with kindness towards oneself enabled an approach orientation instead of the previous avoidance behavior that had functioned as an attempt to feel safe. Acting - and possibly failing - became possible because they were no longer so closely coupled with punishment. In fact, it was found that some processes seemed to cut across nearly all domains, most importantly, self-validation, self-compassion, and agency.

Our findings on the relationship between internalized harshness and agency resemble those of Donald et al. (2019), who found that harsh self-criticism and punitive self-concept may impede the recovery process by preventing individuals from acting. They found a strong positive correlation between self-compassion and recovery from BPD and a strong negative correlation between self-criticism and recovery. Likewise, Katsakou et al. (2019) found that moving from shame to self-acceptance and compassion is central to the recovery process. Previous findings on self-compassion group interventions for BPD suggest that such interventions have some utility (Feliu-Soler et al., 2017; Lucre & Corten, 2013). Donald et al. (2019) postulate that these findings may reflect what Krawitz (2012) highlighted: interventions that explicitly ask BPD patients to cultivate greater self-compassion often provoke negative reactions, as they may be perceived as invalidating. Donald et al. (2019) suggest that the trauma therapy approach, where the “Wise adult self” empathizes with the “Child part”, showing compassion towards the child’s suffering, has the advantage of being client-specific rather than generic. The schema therapy model of BPD, on which the present intervention was primarily based, shares this same advantage, and enables an individualized conceptualization of patients’ problems and history. Schema therapy also adds the Punitive authority (or Critic) mode to the conceptualization (Online Supplement 1). The present participants became skilled in recognizing this harsh, critical voice and in doing so, achieved a healthy distance from it. This seemed to associate with a host of positive effects, such as self-validation, better self-care, and agentic action.

Our results accord with those of previous studies. In the areas of change, we identified the same themes as found in a recent meta-synthesis of qualitative studies that explored BPD sufferers’ perceptions of recovery (Katsakou & Pistrang, 2018), although we categorized them somewhat differently. For example, our first core category “improved ability to observe and understand mental events in oneself and others” encompassed themes from three of the four main categories identified by Katsakou and Pistrang, namely “developing self-acceptance and self-confidence”, “controlling difficult thoughts and emotions” and “practising new ways of relating to others”. Our patients reported on mindfulness experiences and gave rich accounts of their enhanced ability to observe mental events in a new, more compassionate way, to reflect on them, to understand others’ minds and to regulate their emotions and impulses without losing their ability to function. This divergence in the categorization of the same themes may reflect different researcher backgrounds and theory-guided analysis despite the initial use of an inductive approach. Using study design similar to ours, Morken et al. (2019a) explored personal experiences of psychological change processes in 13 female patients with BPD features and comorbid substance use disorder after attending mentalization-based treatment. From the patients’ perspective, their central change processes involved new ways of perceiving and feeling emotions, new ways of thinking about mind-states, new ways of self-reflecting in interpersonal encounters and new ways of exploring others’ intentions in interpersonal encounters. The findings of Morken et al. (2019a) not only resembled those of the present study but the change processes also seemed to demonstrate complex mutual interaction.

With respect to helpful and unhelpful treatment factors or events, the first main finding was that learning about BPD was helpful. Second, normalization emerged as a beneficial factor. Furthermore, two unhelpful treatment factors or events were found: aggression expressed in the group, and inflexibility of the treatment.

Starting with learning, participants described how learning about BPD helped them to relate to themselves in a more normalizing and compassionate way and gave them hope. They learned concepts that aided self-observation and helped them to organize their experiences. Their accounts often revealed a multi-faceted learning process that involved the retrieval of memories, activation of deep emotional processing, and the ability to make use of their recently acquired psychoeducation in new situations. Thus, we assume that psychoeducation provided the initial impetus for this learning. At best, learning was collaboration, as patients were especially eager to learn from and with peers how difficult situations and emotions could be approached, dealt with, and endured. Some narratives revealed innovative moments of collaborative learning that were perceived as very meaningful.

Our findings on the benefits of acquiring information on BPD support those of previously published studies showing that psychoeducation can reduce BPD symptoms (Ridolphi et al., 2019; Zanarini, Conkey, et al., 2018). According to Zanarini, Conkey, et al. (2018), non-disclosure of a diagnosis of BPD in clinical practice often leaves patients thinking that they are “bad” people or the only one suffering from these symptoms. The present qualitative findings support and extend findings from quantitative studies. Specifically, we found that the feeling of inner badness and the associated guilt or self-hate decreased as patients learned about their disorder.

With respect to helpful treatment-related factors and events, our second main finding was that normalization was experienced as healing. Although it was clearly evident that the educational material contributed to normalization, participants often referred to peer experiences when discussing this phenomenon. Specifically, listening to others who had experienced something similar provided comfort and relief by normalizing and validating one’s experiences and even the whole self. Patients described how this normalization was related to a decrease in the sense of aloneness and an increase in feeling connected with others. Normalization also promoted agency. We found that when self-concept becomes more positive and patients no longer identify themselves as flawed, they become able to act in a new way, to take constructive steps towards building a life worth living. In sum, patients started to perceive themselves as more normal, more like others and capable of enduring ordinary disappointments and failings as part of life and to proceed working towards their goals when simultaneously experiencing emotions. This is a significant change, as people with BPD have a propensity to seriously lose their capacity to function when strong emotions are activated.

Previous qualitative research on BPD patients’ subjective experience of recovery has identified connectedness as a relevant dimension in the change process. For example, reporting on a thematic analysis of interviews with five women diagnosed with BPD, Agnew et al. (2016) described how the participants understood their suffering as having relational origins and therefore, relational solutions. Consequently, finding ways of connecting constructively with others was regarded as important in recovery. In a similar vein, Kverme et al. (2019) who interviewed 12 female BPD patients about their experiences with recovery and treatment, identified “moving toward connectedness” as a key dimension in the recovery process. They described how, across their participants, “connectedness implied feeling ‘I am like others and others are like me’, feeling human amongst other humans and part of a community, part of a whole”, in contrast with the feeling of separateness they experienced in the outside world. A subcategory in the Kverme et al. (2019) study was “learning to hold one’s own”, by which they referred to being an agent in the process of change. Agency implied being able to believe that change could come about through action, by changing old patterns and habits. This, too, accords with our observations: hearing others’ experiences normalized the group members’ experiences and, even more fundamentally, the self, and this change facilitated agency.

Our findings on normalization also support Morken et al. (2019b), who explored how patients with BPD features and substance use experienced mentalization-based treatment, and Farrell et al. (2009), who published the first randomized controlled trial on schema therapy groups. Morken et al. (2019b) found that by identifying with others with similar problems, patients who had suffered from shame and a sense of being bad achieved a sense of self-worth. After listening to co-patients, they felt normalized and less alone and bad. According to observations reported by Farrell et al. (2009), groups uniquely possess important curative factors stemming from supported peer-to-peer-interactions such as universality, sense of belonging, vicarious learning and opportunities for in vivo practice. Farrell et al. (2009) also stated that patients accept peer responses as more genuine than those of professionals, who they may believe “have to respond positively”. The qualitative part of the study by Farrell et al. (2009) also identified decrease in the sense of defectiveness, captured in the same words as used by one of the present participants: “I’m not alone, I’m not crazy”. Interestingly, participation in a schema-oriented intervention yields almost the same experiences as participation in mentalization-based therapy.

However, serious obstacles to a helpful therapeutic process, such as problems in the processing of aggression, also emerged. An important finding was that the two participants who showed the strongest reaction to aggression expressed in the group were also the only ones reporting no change in their BPD symptoms. Our findings concerning feeling exquisitely vulnerable in the group and even bullied are consistent with findings from a recent study exploring recovery processes in BPD. Katsakou et al. (2019) also reported that individuals with BPD could feel exposed when sharing personal information and that they could experience peers as dismissive or bullying. If participants’ interpersonal schemas are very insecure and epistemic trust low (Fonagy & Allison, 2014), a single episode experienced as too painful may induce a response that is difficult or even impossible to repair, as happened when one patient experienced another patient as resembling her former school bully and decided to not open up anymore. Failure to process aggression must be regarded a limitation of the intervention investigated in the current study. The group was structured such that psychoeducational material was first presented, after which the participants could discuss their own experiences on this theoretically introduced topic. It is not, however, an easy task for group leaders to strike a balance between covering the educative content and attending to the group process. In an educative group, there may not be enough time to respond to participants’ experiences in a way that encourages further exploration. In addition, patients may be far from ready to admit their more primary woundedness and the insecurities that contribute to how they experience each other, a process needed to help prevent them from projecting onto peers something that is within the self. As Farrell et al. (2009) put it, a group per se can play an important curative role in the treatment of people with BPD if it is structured to avoid invalidating and schema-perpetuating experiences. Feedback needs to be solicited and constantly actively processed to avoid an escalating group process. The intervention was also limited in that individual therapy was not an integrated component of the treatment model. We do not know if the fate of the non-responders who experienced a strong negative reaction to aggression would have been different had the group facilitators and individual therapists had structured opportunities for collaboration, as is the case in DBT and MBT treatments.

With respect to helpful and unhelpful treatment factors and events, our last finding was that inflexibility of the treatment hindered the patient’s ability to benefit from it. Most importantly, inflexibility triggered a feeling of not being heard and a feeling that the facilitators are not on “my side”. Our findings accord with the meta-synthesis findings reported by Katsakou and Pistrang (2018). They found that feeling that one was not an equal partner was mentioned as an unhelpful treatment-related factor in 9 of the 14 studies included in their meta-synthesis of qualitative studies investigating BPD clients’ experiences of treatment and recovery. Consistent with our findings, they noted that therapy that was too rigid and inflexible induced feelings of powerlessness and anger. Recently, Katsakou et al. (2019) also found that individuals with BPD felt coerced and disrespected when they experienced therapists as rigid in following therapeutic agendas. According to Linehan (1993), many battles in psychotherapy have to do with the maldistribution of power and patients’ attempts to rectify it. She explains how borderline patients are quick to detect power differences and are intolerant of arbitrariness in the therapeutic relationship, perhaps because they have suffered in the past from an unequal distribution of interpersonal power. It is not surprising that most psychotherapies for BPD nowadays underscore the need for flexibility as opposed to rigid rules (e.g., Bateman & Fonagy, 2012; Linehan, 1993).

Study Strengths

Perspective triangulation, i.e., the combination of different frameworks may be considered a strength of this study. Combining the medical framework (change in BPD symptoms) and patients’ subjective experience of meaningful change and treatment enabled us to compare narratives between those who achieved remission and those who achieved reliable change with those who did not change in medical terms.

The credibility of the results is increased by investigator triangulation, i.e., the involvement of multiple observers and interpreters. In addition, the trustworthiness of the study is supported by the provision of representative excerpts from the data.

Limitations

This study has its limitations. It was predetermined that only eight patients could be studied. We were, therefore, unable to take saturation into account in the sampling protocol.

Regarding the trustworthiness of this study, SL is one of the developers of the intervention while both MK, who interviewed the patients, and TM, who interpreted the data with MK, had nothing to do with the development of the intervention or the organization that delivered the treatment.

A major limitation concerns the transferability of the results, as some of these results may apply only to group treatment. For example, while it may be quite easy to deliver psychoeducation in the group context, an educative approach in the individual therapy setting is a more complex issue and may meet with a very different response.

In sum, psychoeducation, sometimes rejected as a superficial approach, seems capable of providing impetus for very meaningful change. Groups can have unique benefits such as providing opportunities for collaboration in learning and normalizing participants’ sense of self, as participants can feel understood and feel that they are like others and others are like them. However, to prevent serious obstacles to recovery, it is of paramount importance to understand how an optimal group process may be facilitated.

Acknowledgements

We are grateful to the participating patients and therapists (Merja Saarela and Nina Forsman) at the community outpatient mental health care services in the Central Finland Health Care District without whose cooperation the study would not have been possible.

Supplemental Data

Supplemental data for this article can be accessed https://10.1080/10503307.2021.1883763

Additional information

Funding

Maaria Koivisto has received a grant from the Finnish Association for Cognitive and Behaviour Therapy.