Design

An exploratory, qualitative, interview-based study, assessing patients' perspectives of recovery in BPD was conducted. The study design drew on Grounded Theory and thematic analysis. Grounded theory is a method aiming to inductively build a theoretical explanation of a social phenomenon based on the study data [31]–[33]. Thematic analysis is used to identify a limited number of themes that adequately reflect the data, by comparing and refining emerging topics [34].

The core research team included researchers with academic and clinical backgrounds and service users. More specifically, CK is an academic researcher and a DBT psychotherapist; SM works clinically with forensic patients with personality disorders; KB and MS are academic researchers working with service users with a BPD diagnosis; SP has a long clinical and research experience as a psychologist and psychiatrist; HW is a service user with a diagnosis of BPD who has received MBT; KL is a consultant psychiatrist and psychodynamic psychotherapist working in an MBT-informed setting. The team met regularly to discuss the study design, implementation, and data analysis.

Service-users were also involved in various stages of the study (design, data analysis and interpretation) to ensure that their perspectives were reflected in the interpretation of the data. Their specific contributions will be described in detail in the relevant parts of the methods.

Written informed consent for participation in the study was obtained by all interviewees. The study design was approved by the East London National Health Service (NHS) Research Ethics Committee (ref: 09/H0704/14).

Sample and data collection

Participants were recruited from secondary mental health services in East London, including two specialist services for BPD (a DBT team and a therapeutic community informed by the MBT approach) and generic mental health services who offer support to service users with a range of Axis 1 and Axis 2 diagnoses. The generic services included 3 community mental health teams (CMHTs) and a psychological therapies service. The psychological therapies service offers psychological therapy, including cognitive behavioural, psychodynamic and integrative interventions. The inclusion criteria for participation in the study were: age above 18 years, a diagnosis of BPD and a history of self-harming. Self-harming was defined as self-injurious behaviour, overdosing or suicide attempts that were performed with the intention to self-harm. It should be noted that participants did not have to engage in self-harming behaviour currently, but were included if they had done so at any point in their lives. A history of self-harming was used as an inclusion criterion, as we wished to have a homogenous sample and we aimed to include those who had experienced more severe BPD symptoms at some point in their lives and who are therefore likely to use services for BPD-related problems frequently. Furthermore, in our experience, service users with no self-harming behaviour might be more likely to receive Axis 1 diagnoses, such as mood, anxiety or eating disorders, and therefore psychiatric or psychological treatments not specifically developed for BPD. Those with severe learning disabilities, those who did not speak sufficient English to participate in interviews and those unable to give informed consent were excluded.

The current and archived referrals to the two specialist services were reviewed and eligible patients were identified. Professionals from three CMHTs and one psychological therapies service were also contacted and asked to inform the researchers about eligible patients. Purposive sampling was applied to ensure that the research aims were addressed and that the sample included interviewees with a wide range of characteristics. Both patients who perceived that they have recovered and those who did not were included. Similarly, participants with various clinical and demographic characteristics were selected (i.e. co-morbid diagnoses, service use, a wide range of ethnic backgrounds, age and gender). Furthermore, service users who were engaged with services and those who discontinued their treatment were interviewed. Among those who were engaged with services, we aimed to interview service users after they had used services for at least 4 months, so that they had some time to reflect on the treatments they received. New participants were recruited on the basis of their potential similarities or discrepancies from the already participating sample. The sampling of new participants stopped when saturation of the emerging themes was reached [31]–[33].

Once patients meeting the inclusion criteria were selected, they were contacted through their key-worker and introduced to a researcher. The researcher explained the study and asked for their informed consent to take part. If consent was given, self-reported socio-demographic and clinical characteristics were collected (see table 1) and the qualitative interviews were conducted. The clinical characteristics included diagnosis as documented in the service users' files. Both specialist services used SCID-II to assess Axis 2 diagnoses, whereas in generic services all diagnoses were commonly given by psychiatrists in clinical interviews (without the use of psychometric instruments).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Participants' socio-demographic and clinical characteristics.

https://doi.org/10.1371/journal.pone.0036517.t001

The sampling and data collection process was discussed in a meeting with CK and 4 service users who had used both generic and specialist services. Their feedback informed final decisions in these areas.

Interviews and topic guide

In depth semi-structured interviews were conducted. A topic guide for the interviews was developed in a meeting between CK and two service users who had used generic and specialist services. Participants were asked to describe what they perceived as recovery, their goals and aspirations, their journey towards recovery and their reflections on their progress and achievements.

The above-mentioned core topics were covered in all interviews. Participants were also encouraged to discuss related topics that they judged significant. The interview style was flexible, guided by neutral and open questions. The interviews lasted between 30 and 120 minutes. Four researchers recruited and interviewed study participants. All interviews were recorded using a digital recorder and transcribed by a professional transcriber.

Data analysis

Data gathering and analysis were interdependent and were carried out one after the other repeatedly, until saturation on the meaning of recovery from BPD was reached. This was a dynamic process, consisting of moving from data collection, through initial analysis, to theoretical hypothesis generation that informed new rounds of sampling and data collection. Thematic interview coding and constant comparisons to identify similarities and differences between emerging themes and sampled cases guided the researchers into more abstract understandings of the themes, leading to their conceptual clarification and to the development of more holistic interpretations [31]–[34].

The process of the construction of a coding frame to capture the various emerging themes was inductive and based on the interview data. The coding frame was developed by four researchers and two service users. Following this, three researchers coded 30 interviews together to discuss practical issues around coding and refine the coding frame, a technique known as multiple coding [35]. CK then coded all interview transcripts using the MAXqda software (version 2) for qualitative data analysis.

Once the research team had agreed on the core themes emerging from the data, the data analysis was discussed in a meeting between CK and 4 service users. Their feedback helped the research team to further refine their understanding and interpretation of the data.

A. Personal goals and/or achievements during recovery

The main areas where participants felt that they have made progress or would still like to improve are presented below. These goals/achievements were seen as inter-linked and influencing each other: improvement or deterioration in one area usually led to improvements or problems respectively in achieving other goals.

B. Balancing personal goals of recovery versus service targets

Some participants thought that there was a clash between their personal aspirations and the focus of treatment. They felt that therapy did not address all problems they were struggling with. Some treatments were experienced as focusing almost exclusively on specific topics, i.e. self-harming or relationships (often as they were enacted in the group setting), leaving service users frustrated when they could not address other issues that were either equally or more important to them. Other common problems that service users felt were not sufficiently addressed were eating problems and past traumatic experiences. “DBT helped, but it didn't answer all of my questions. It didn't help me to work things through myself, it didn't help me to achieve my goals really… I was trying to get over my divorce and also my relationship with my mum and men, and I was trying to work through it but it was all about other things, it was about self-harming, it was about mindfulness…” (participant 11).

C. How recovered do people feel?

Participants described various states of recovery, as illustrated below. These were perceived either as different stages in their journey to recovery or as reflecting their overall recovery status.

D. Problems with the word ‘recovery’

Some interviewees believed that recovery might not be the right word to describe their progress. They felt that the term recovery implies a dichotomous classification of problems, suggesting that people either have problems or they are fully recovered. They believed that in BPD this is particularly inappropriate, as it might reflect and encourage black and white thinking, one of the symptoms of the disorder. They thought that full recovery in that sense is impossible, as they could not imagine not having some difficulties in dealing with their emotions and lives. Thinking that they are recovered following such a dichotomous definition was also seen as dangerous, as it could be unrealistic, indicate a lack of acceptance of their problems, and lead to not monitoring themselves and therefore relapsing. “I think recovery is a very difficult word particularly with mental illnesses and I think you can recover, but I suppose I'm naturally worried that if I go and recover, I would be worried that I could think I'm wonderful now and then all of them fall from the rails, cause I'm not keeping a check of myself… I think mentally I'm always gonna have to keep in my mind that I have these issues, I have this problem, I have this diagnosis and that's not to say it's a bad thing… I don't want to say I've recovered because I don't want to give myself an opportunity to relapse” (participant 9).

Other participants thought that especially in the context of BPD separating themselves from the disorder is particularly hard, as they have been experiencing emotional difficulties for as long as they remember. Therefore, BPD is not something they feel they can recover from, as that would mean that they would have to become a different person, which is not necessarily what they want. “I can't imagine not having BPD. I don't remember a time in my life when I didn't feel this way. So recovery, cure… no, I don't think so. Learning how to deal with it, I'm very positive about… I don't think it will every go away. I'd have to have a personality transplant for that to happen… I'm not entirely sure I want to actually let go of certain aspects of me. I like the fact that I'm a compassionate human being. I don't want to be this involved with people's emotions, but I wouldn't want to not care. And that wouldn't be me. I'd much rather learn how to deal with it than have it taken out of me” (participant 22).

More quotes on the above themes are presented in Box S1.

Main findings

For service users with BPD recovery involved developing self-acceptance, self-confidence and self-esteem, gaining control over emotions, moods and thoughts, improving relationships, getting involved in activities and employment, and making progress in clinical symptoms, such as suicidality, self-harming, eating problems, drug and alcohol consumption. Some service users felt that some of their goals were not adequately addressed and that therapies had an extreme focus on specific topics, such as self-harming or relationships. Recovery was experienced as a dynamic process with various stages. Participants described how their recovery fluctuated, with periods with marked improvements followed by times when things were particularly hard to manage. This made them feel exhausted and disheartened, although it was often seen as a natural process in their recovery journey. Although most interviewees felt that they make continuous and gradual progress, full recovery was commonly seen as a distant goal. However, people felt hopeful that they can learn how to deal with their problems in more effective ways and keep on moving forward and making meaningful changes in their lives.

Interestingly, some service users did not find the word ‘recovery’ helpful in describing their experiences of personal development and progress. Some felt that the term implies a dichotomous classification of problems, a black and white way of thinking, in which recovery becomes the unobtainable goal of having no problems at all. Others thought that, especially in the context of BPD, separating themselves from the disorder is particularly hard, as they have been experiencing emotional problems for as long as they remember.

Strengths and limitations

To our knowledge this is the first qualitative study exploring perceptions of recovery among people with BPD. The sample size was large (48). Interviewees were recruited from a range of specialist and generic mental health services and represented various levels of recovery. The research team consisted of researchers with various academic and clinical backgrounds and service users were involved in all stages of the study.

Nevertheless, participants were recruited only from services in East London and we do not know to what extent their experiences can be generalised to users of other services. Furthermore, the specialist services investigated included DBT and MBT and we do not know whether service users' views of these approaches apply to other specialist psychotherapies for BPD, such as transference-focused therapy (TFT) or schema therapy. Similarly, therapists' perspectives' of recovery and their views on how treatments address these were not explored. Moreover, although the response rate for participation in the study was high, we do not know whether the findings could be generalised to service users who were not able or willing to take part in research. Lastly, participants' expressed views might to an extent reflect the philosophies of the treatments they received.

Findings in the context of previous literature

The perception of recovery expressed by service users with a BPD diagnosis, where recovery is seen as an open-ended journey and involves learning how to cope with problems and developing a meaningful life with the limitations of the disorder, reflects the definition of recovery within the wider recovery literature [27]–[28]. Similarly, some of the aspirations service users described in this study are in line with recovery goals among users of general mental health services with mainly Axis I disorders. Improvements in self-acceptance, relationships, activities and employment have been widely documented as reflecting service users' perceptions of recovery [28], [36].

Other goals, however, such as gaining control over difficult emotions and specific symptoms (i.e. self-harming) are more specific to the nature of BPD. Similarly, the meaning of self-acceptance and the development of self-confidence in people with BPD might be different to re-claiming identity after a diagnosis of mental illness among users of general mental health services [36]. Interviewees described how they struggle with shame and guilt, not as a result of having a diagnosis of mental illness, but because they find it hard to come to terms with who they are. This might mirror enduring problems in developing a sense of identity and self-compassion, which often reflects a lack of secure attachment relationships and a history of abuse or neglect among people with BPD [37]. In this context, improving relationships for this group might also be more complex than solely addressing social isolation, which is commonly discussed in recovery literature [36]. More specifically, it might also involve developing relationship skills, such as building trust, tolerating fears of abandonment and allowing themselves to feel vulnerable in close relationships. Therefore, findings from this study support the view that what constitutes recovery and the meaning of specific recovery goals might be different for service users with different diagnoses [30], [38].

A key question arising from our findings is to what extent treatments for BPD address service users' personal goals of recovery. This may be more realistically achievable by specialist psychotherapies than general psychiatric services. The two specialist psychotherapies received by the study participants – DBT and MBT– both directly address some of the goals identified in this study. DBT places an emphasis on reducing self-harming and on emotion regulation [15], whereas MBT focuses on understanding and developing relationship skills [17]. Both therapies implicitly facilitate self-acceptance and self-confidence by offering a theoretical explanation of factors leading to the development of BPD and by fostering processes like mindfulness and mentalisation respectively, which are intended to help people understand and accept themselves, their emotions and behaviours [15], [17]. DBT also targets more practical goals, like employment, but only if self-harming has resolved. It is therefore encouraging that both these therapies address at least some of the targets that are important to service users. At the same time, we believe that each therapy focuses on one main area, i.e. DBT on self-harming and MBT on relationships. (It should be noted that this interpretation of the main areas of focus for each treatment might not coincide with the views of those who developed these therapies). While this might be an effective way for therapy to stay focused and help service users make significant improvements at least in one domain, it might fail to take into account other important goals. Such overfocus on single problem areas has been highlighted as a limitation of some specialist psychotherapies for BPD [8], [26], which may contribute to treatment dropout and/or slow therapeutic progress [39].

Other specialist psychotherapies for BPD, including schema and transference-focused psychotherapy, claim a wider impact on personality structure, which goes beyond improvements in individual BPD symptoms, and a higher likelihood of full recovery [25]. Schema therapy believes that by working on negative self-schemas, service users with BPD develop a more positive sense of self and agency, which then lead to further improvements in functioning and overall quality of life [26]. TFT claims that service users achieve meaningful changes by developing their self-control and healthier representations of self and others, through working on the therapeutic relationship [22]. Whether service users agree with such claims was not assessed in the present study and needs to be explored in future research. Conclusions

This study identified recovery goals that service users find important. Treatments focusing on these targets may increase users' motivation and engagement with services and facilitate recovery. The findings may also guide research in this area and ensure that outcomes that are relevant to service users are evaluated and processes of achieving such outcomes are explored.

The specialist therapies for BPD that were investigated in this study (DBT and MBT) explicitly address some of the recovery goals that are important to service users. However, other goals might only be indirectly or poorly addressed. It might be unrealistic for specialist treatments to offer solutions to all problems, as these treatments might naturally have to focus on some of the most important areas to be effective. Similarly, the fact that service users did not feel that these treatments addressed all their goals might also reflect an overall slow pace of recovery in BPD and the need for long-term and comprehensive care.

In this context, service users' specific priorities might determine which specialist service might be more appropriate: i.e. for those whose main aim is to stop self-harming DBT might be more relevant, whereas those who want to work on relationships might find MBT more appropriate. Other treatment models, like schema and TFT, could also be explored based on service users' preferences. Future research should focus on understanding specific processes of change within each therapy, which might also guide treatment choice by service users [24].

Professionals might also need to work with service users towards devising comprehensive individualised case formulations, including all treatment targets that are important to them, their priorities and long-term plans on how their targets might be met and which services might be involved. Specialist services for BPD might be a starting point, but other services, including eating disorders, drug and alcohol, trauma services, or employment schemes might also be involved at different stages of recovery.