“It’s a dent, not a break”: an exploration of how care co-ordinators understand and navigate boundaries in early intervention in psychosis services

ABSTRACT

Background

Early Intervention in Psychosis Services (EIP) offer a unique service model for people experiencing a first episode of psychosis. They are intensive case-management services which adopt an assertive outreach approach, employing flexible boundaries to meet clients in the community, and support them towards recovery and holistic goals. Current boundary theory is therefore not easily applied to this clinical context.

This study aimed to explore how care-co-ordinators in EIP develop their understanding and practise around navigating boundaries.

Methods

Participants were 13 EIP care co-ordinators. Semi-structured interviews with participants were conducted and analysed using grounded theory.

Results

A concentric model emerged, defining different layers of influence impacting care co-ordinators’ navigation of boundaries. This included higher governing levels (e.g. professional codes of conduct), factors and challenges specific to EIP culture, client characteristics and individual care co-ordinator decision-making processes. A flexible, case-by-case approach was adopted in response to challenges.

Discussion

Findings suggest clinical implications for how care co-ordinators can jointly construct appropriate boundaries with clients and families, and discuss dilemmas using the team resource. This model can be used as a tool in clinical practice to aid clinicians’ thinking and reflection around boundaries within EIP.

KEYWORDS:

• Boundaries
• early intervention in psychosis
• care co-ordinators
• case-managers
• assertive engagement
• assertive outreach

Background

People encountering a First Episode of Psychosis (FEP) often experience high levels of fear, mistrust, confusion, and distress. Understandably, this can lead to a reluctance to engage with services (Doyle et al., 2014; Tindall et al., 2018). To try to combat this, an “assertive outreach” (or “assertive engagement”), intensive case-management model is usually adopted (Anderson et al., 2010), endorsed by the Early Intervention in Psychosis Network (EIPN). Assertive engagement refers to a style of working whereby clients are not discharged quickly if they disengage, and case-managers (often termed “care co-ordinators”, or “lead practitioners”) adopt a flexible and committed approach with a focus on engagement, building rapport at the client’s own pace, and establishing a relationship over time (Early intervention in Psychosis Network, 2018).

The case-manager role is vital within EIP, with case-managers supporting clients across a holistic range of domains (including social recovery, education/occupation, and family intervention; Wong et al., 2019). Small caseloads and high levels of contact allow for meaningful relationship building, close support, and continuity of care. There is a culture of “standing alongside” clients, flattening the traditional patient-clinician hierarchy, and case-managers positioning themselves as a supporter, aiding the client to achieve their goals (Early intervention in Psychosis Network, 2018).

In order to facilitate this more assertive approach, EIP services promote flexibility with regards to more traditional therapeutic boundaries. For example, case-managers will often meet with clients outside of the traditional clinic room setting, such as in clients’ homes, or public settings such as parks, cafés, or even gyms, and will support clients not only emotionally but also practically (e.g. taking a client to a food bank in their car). These boundaries are flexed to encourage engagement and build relationships, or in relation to other client goals such as increasing confidence, social interaction or independence in the community (Farrelly & Lester, 2014). Case-managers also typically use more flexible communication methods than many services, such as texting or emailing with clients (from work phones or emails), to enhance accessibility.

This way of working inevitably pushes “traditional” clinical boundaries, and can render personal/professional boundaries particularly porous for case-managers. For example, there can be more clinician self-disclosure, and dilemmas navigating the role of “supporter” versus “friend”. This can render more traditional psychological models of therapeutic relationships and boundaries (e.g. Hinshelwood, 1999) less applicable.

However some level of boundary setting is essential in order to facilitate a safe therapeutic frame, to protect the wellbeing of both client and clinician, and to aid clients in the development of boundaries in their own lives, leaving dilemmas for professionals about where to draw a line. This lack of clarity in the theoretical and research literature regarding navigating boundaries with clients with psychosis in community case-management settings has been noted (Farrelly & Lester, 2014). However, to date, no research has explored how clinical boundaries are navigated by case-managers in EIP services.

Considering the clinical issues unique to people with FEP, and lack of applicable boundary theory (Farrelly & Lester, 2014) it is proposed that research which enables us to better understand how boundaries are navigated with clients within the EIP service model would be a useful addition to the evidence-base.

This study aimed to develop a grounded theory (GT) to explore how case-managers in EIP services develop their understanding and practice around navigating boundaries.

Methods and materials

Design and ethical approval

Qualitative, semi-structured interviews were undertaken via videoconferencing (due to COVID-19 restrictions). The interview schedule was devised by the authors in consultation with an advisory group of experts by experience, who work with Salomons Institute for Applied Psychology.

A GT methodology was used (Glaser & Strauss, 1967), approached from a constructivist epistemological position, in line with Charmaz’s GT methodology (Charmaz, 2014). This position acknowledges that “knowledge” is constructed through influences such as language, social discourses, and power (Clarke, 2005). The researcher therefore collected and analysed the data whilst being cognisant of their own experiences, assumptions and biases.

This study was granted ethical approval from Canterbury Christchurch University ethics panel and NHS Health Research Authority (IRAS ID 274036). The study was also approved by the recruiting trust’s research and development department.

Data collection and analysis

Care co-ordinators in participating EIP teams were provided with information about the study. Email addresses of interested potential participants were collected, and the researcher met with eligible participants. Informed consent was gained prior to interviews being conducted.

All interviews were undertaken by the lead author. Due to recruitment challenges, opportunity sampling was initially used. However, within GT, data collection and analysis are undertaken concurrently, using theoretical sampling (Glaser & Strauss, 1967). Based on data from initial interviews, theoretical sampling guided subsequent data collection as much as possible. This meant that decisions about what data to collect next (and from whom) were based on the ideas emerging from the concurrent analysis, for example, recruiting participants who had been qualified for longer. Semi-structured interviews were guided using an interview schedule which covered the following main areas:

• How did case-managers in EIP services navigate boundaries with clients in their clinical practice? What informed their decision- making?

• What were the benefits and challenges around how boundaries were navigated that case-managers experienced?

• How did they overcome these challenges?

Questions were open and non-leading, allowing participants to respond freely and with depth. However, prompts were available to aid conversation where needed. Questions and conversation often deviated from the interview schedule where this was relevant to the research questions, and in line with GT, the interview schedule was adapted over time by the lead author based on theoretical sampling, and the perceived gaps in the data. Interviews lasted between 45–82 minutes. Recruitment ended once the lead author felt “theoretical sufficiency” (Dey, 1999) had been reached.

Interviews were analysed using three stages of coding: open, selective, and theoretical coding (Charmaz, 2014; Urquhart, 2012).

Towards the end of the theoretical coding stage, through constant comparison between memos, categories and transcripts, and using integrative maps (Strauss, 1987) a theory was generated and regenerated directly from the data, resulting in the model presented in this paper.

Participants

Thirteen care co-ordinators were recruited from across four EIP teams in a trust in the south-east of England. Participants comprised a range of ages (M= 39), clinical experience and professions. Ten participants were White-British. On average, participants had been qualified 10.8 years, and had worked in EIP for 3.8 years. Table 1 outlines the demographic and professional characteristics of participants.

Table 1. Demographic characteristics of participants (page 11) (uploaded separately).

Participant	Age	Gender	Ethnicity	Profession	Years in EIPS	Years Qualified	Previous clinical experience
1	37	Male	White-British	MHN	4–6	4–6	Homeless, forensic services
2	32	Female	White-British	MHN	1–3	1–3	Inpatient CAMHS, children services
3	36	Female	White-British	OT	1–3	1–3	Hospital/community physical health
4	49	Male	White-British	MHN	10–15	16–20	Substance misuse services, CMHT
5	32	Female	White-British	SW	4–6	4–6	Children services, CMHT
6	26	Female	White-British /French	MHN	1–3	1–3	Inpatient CAMHS
7	34	Male	White-British	MHN	1–3	1–3	Inpatient – women’s “PD” Services
8	42	Female	White-British	MHN	1–3	10–15	Inpatient, female acute, dementia unit
9	27	Female	White-British	MHN	1–3	1–3	LD services
10	43	Female	Black Caribbean /White British	MHN	1–3	10–15	CMHT, 3rd sector organisations
11	44	Male	White-British/Irish	OT	4–6	16–20	3^rd sector, physical health, social services
12	56	Female	White-British	MHN	4–6	>25	Forensic services
13	53	Male	White-British	MHN	1–3	>25	Inpatient, crisis team, CMHT

Key: MHN = Mental Health Nurse, OT = Occupational Therapist, SW = Social Worker, CAMHS = Children and Adolescent Mental Health Services, CMHT = Community Mental Health Team, PD = Personality Disorder, LD = Learning Disability. Regarding Years in EIPS and Years Qualified, a range is given to help protect participant anonymity.

Quality assurance

Researcher reflexivity was ensured using a bracketing interview with a colleague prior to data collection, to illuminate the preconceptions and biases they held around navigating boundaries. This led to a researcher positioning statement. A reflective research diary was also kept by the lead a uthor, enabling them to remain cognisant of their biases and responses to interviews.

Several sections of transcripts were co-coded by the second author, and the emerging categories and model development were discussed. Transparency throughout analysis and theory development was ensured through the following processes:

• Use of memos

• extracting quotes to support emerging or reorganised categories across transcripts

• Integrative maps

Finally, involvement in respondent validation was offered to all participants. Whilst all consented to be involved, at the time of setting up feedback interviews, four participants responded to this request. The model was therefore shared with each of these participants in individual interviews. Feedback from these interviews was incorporated to create the final model.

Results

Model overview

Data analysis resulted in a concentric circular model (Figure 1) which has been numbered to enable easier navigation with the text. The model includes four layers of influence on case-managers’ navigation of boundaries with clients:

• Professional codes of conduct (17)

• EIP culture (6)

• Individual case-manager decision-making processes (2)

• Client characteristics (1)

Figure 1. A model of case-managers navigation of boundaries within EIP services (page 13).

Within each layer subcategories relevant to the navigation of boundaries in that section are identified. The model conveys that navigating boundaries in this context is a dynamic interaction; there are links between the layers of influence, and many of the subcategories are closely connected. The challenges participants faced around boundaries have been interwoven throughout the subcategories, where these were most pertinent.

Certain layers and processes (e.g. professional codes of conduct, or the “learning and calibrating cycle (18)), whilst important to case-managers’ navigation of boundaries, were arguably more generalisable clinical skills or learning processes (such as using supervision). They are therefore still included in the model, but the focus of the results will be on the layers and processes most unique to EIP culture.

The box (19) positioned outside of the circles represents case-managers’ prior experiences or influences which they carried with them into this role. It was clear that case-managers natural “stance” around boundaries fell somewhere along a spectrum; some tending towards stricter, rigid boundaries, and others towards looser, more flexible boundaries. This in turn was influenced by prior experiences, for example, during professional training.

Client characteristics (1)

This layer relates to the client themselves, and individual client characteristics which could influence or shape how participants navigated the relationship, and with it the boundaries, with individual clients.Such features included demographic characteristics, particularly age, gender, class and cultural background and ethnicity, which applied to both client and clinician, and the idiosyncratic pairing. Four participants expressed general principles around boundaries they employed relating to demographic characteristics (e.g. tighter boundaries with younger clients, and more relaxed boundaries with older clients, or preferring not to work with young females as an older male).

Other client characteristics were also considered when tailoring boundaries with clients. These included aspects such as attachment difficulties or trauma histories, forensic histories, presence of autism spectrum conditions, specific psychotic experiences, and level of engagement and risk. Certain client characteristics (such as attachment difficulties, or an autism spectrum disorder) sometimes suggested to participants that employing more explicit, firm boundaries from the outset would be most helpful.

Individual case-manager decision-making processes (2)

This layer refers to the day-to-day, or moment-by-moment decisions around boundaries participants made individually, and the factors they considered in their boundary practice with clients. Examples of these include:

• Drawing own personal boundaries (3)-for example, preferences regarding who they worked with (e.g. not sharing a GP surgery with a client), or “lines” they never crossed, even if professionally permissible (e.g. not accepting a gift of a box of chocolates).

• Using self-disclosure judiciously (4)- self-disclosure was used frequently for therapeutic benefit (e.g. normalising, or building rapport). However, each participant had their own limits, and when working with clients with “positive symptoms” of psychosis, some participants were more wary of self-disclosing, for fear of “feeding in” to overvalued ideas or delusional belief systems.

• Considering motive/rationale (5)- participants emphasised the importance of considering their rationale, or underlying motive behind making boundary decisions; ensuring their motive was for the client, not themselves. Attention was also paid to the client’s motive or needs.

EIP culture (6)

This layer represents how the culture within EIP was a significant influence on how case-managers navigated boundaries with clients.

Navigating assertive engagement (7)

A key tenet of the EIP model is assertive engagement/outreach. Participants described their primary task as engaging clients and building rapport, and the importance of this approach with people experiencing psychosis. “Some services … they didn’t turn up … we sent another appointment letter … they didn’t turn up … if we were to … strictly stick to that boundary and not assertively go and approach that client in different ways … we would lose fifty percent of our clients … they just wouldn’t get a service” (Participant 11). Nine participants employed looser boundaries to facilitate engagement, with clients who were not sure they wanted help. Eight participants felt that having explicit conversations about boundaries in strict terms hindered engagement with reluctant clients, and so would only do this if the need arose “We, won’t make those boundaries very explicit at the beginning and we kind of just fit in with wherever that person is in their lives … in the least obtrusive way” (participant 7).

However, assertive engagement sometimes came at a cost of difficulties maintaining either the clinician or the client’s boundaries. A tension was navigating the fine line between engaging versus harassing; how to assertively engage whilst respecting client’s own boundaries, or when to step back. “they felt we had really overstepped the mark … turning up unannounced to my house … how dare you … the service-user’s boundaries have been pushed … Because of our … prerogative and our agenda, when actually … maybe our perspective on what’s needed is just not aligned? … that has damaged the relationship” (Participant 9).

Meeting outside the clinic room (8)

Meeting outside the clinic room was a means through which many processes related to flexing boundaries (assertive engagement, levelling power, building close relationships) took place. Eleven participants expressed the benefits of working in this way, enabling the same “work” to take place in a way which was more suited to the client’s needs and goals (e.g. going to a café if a client wanted to become more confident in social situations, or taking exercise together if a client wanted to lose weight).

However, meeting outside the clinic room also raised questions about whether clinicians’ boundaries still held precedence when in communal spaces, and often a negotiation, or compromise around boundaries took place. “we were in this café … he wants to get a beer … is it my place to be saying … you shouldn’t be drinking alcohol when you’re meeting up with me as a professional … this person’s … living out in the community … they’re able to make those decisions … they’ve got that capacity … I said … it’d be nice to catch up with a clear head … but … I’ll leave that decision up to you. And he said … I’ll just have a half then” (Participant 9).

Levelling power relations (9)

All participants referred to the power dynamics in the client-clinician relationship. EIP advocates a flattened hierarchy, minimising the inherent power imbalance between clinician and service-user. Flexing boundaries (more tailored to the wants or needs of the client) was a way of levelling power.

Six participants tried to minimise the power imbalance by taking a collaborative approach with clients, jointly constructing boundaries. “having … open conversation about … how do you want that set up? … Be open and honest and transparent about the fact that we need to have boundaries … but actually, how do you find that and what do you want” (Participant 3). This was often complex in practice. Participants acknowledged that ultimately they still held more power, information, and duty of care responsibilities, which they could use to invoke restrictions on client’s liberties.

Inviting authentic relating (10)

This subcategory was related to using self-disclosure judiciously, although encompasses a broader sense of bringing one’s authentic, genuine self into clinical work. This often involved expressing their emotions and human responses to clients, which indirectly loosened boundaries. “Making yourself human and … a real person … just showing that” (Participant 6).

Participants felt that bringing aspects of their real selves into the relationship invited clients to reciprocate. Seven participants felt if they were not relating authentically, this led to more traditional “patient-professional” roles and boundaries, in which the client did not meaningfully engage. “I’m asking them about their kids … their life … to share it in … intimate detail and if I’m not prepared to … be a human as well who has similar experiences … it just feels a bit disingenuous” (Participant 7).

Building close relationships (11)

Due to the intensive nature of EIP treatment, and the three-year timeline, participants spoke about building closer relationships with clients over time; being part of their recovery journeys and personal growth. This was linked with the notion of standing alongside clients, positioning themselves as an ally or supporter. “we’ve been on a journey here … I’ve seen you grow … a couple … they were street homeless … we’ve been able to support them with housing and, employment and, getting their driving licence back” (Participant 9).

Closer relationships meant participants could struggle with implementing boundaries when participants had a naturally strong alliance with a client, or when transference processes, or other complex interpersonal dynamics were apparent. “she’d say things at the end of conversations like, ‘love you’ … like you would with a close friend … it’s really hard … ‘cause I did feel a bit of love for her! … it was very sort of maternal relationship” (Participant 8). This reflected the fine line between being a supporter versus a friend (or parental figure), and was complicated by use of more informal communication methods, such as texting or WhatsApp. “we’re … maybe more relaxed about what we’re actually saying and sharing and how often you’re communicating and texting on WhatsApp” (Participant 8). This concern was particularly present when clients did not have any other support or social networks.

Participants spoke about remaining cognisant of professional boundaries, whilst still allowing for a rapport. However, there was a sense for four participants that even if they felt they had drawn the boundary successfully, they could not always control how the client perceived them. “if I’m in somebody’s life in a professional capacity for three years and they feel that they want to share a lot with me … the boundaries may be blurred for them … even if I feel I’m managing it” (Participant 5).

Negotiating boundaries with families (12)

There could be tension in managing families’ expectations of boundaries, and marrying this with the assertive outreach way of working. “it’s quite hard … to explain to parents about building rapport … parents might expect you to go in and sit there with that clipboard and be seen by them to be doing something” (Participant 1).

As a result of becoming involved in many aspects of clients’ lives, determining where to draw the boundaries around confidentiality and how much to share with family members was a challenge. Sharing information could be helpful for families, but have repercussions for clients in feeling as though their personal boundaries had been overlooked (even if they had consented to information being shared).

Sharing decision-making (13)

All participants spoke about the culture of using the “team brain” (participant 9) when facing boundary quandaries. Dilemmas were discussed with the team, different perspectives offered, and decision-making was shared. Where opinions differed on how to navigate a boundary, a middle ground was often reached in which everyone (especially the case-manager) felt reasonably comfortable.

If a clinical decision was made that it was not appropriate to have explicit conversations with clients around boundaries (perhaps not well-received or understood by a client), more active collaboration with colleagues and support occurred; e.g. involving a co-worker to reduce level of risk or discomfort for the case-manager, or to prevent a client becoming too attached to a case-manager, or vice versa.

“you take it to the team … “ooh how, how do I manage this?” … they’ll say … “why don’t we introduce someone else in the team” … or “why don’t you have a conversation about it … make that explicit” (Participant 6).

Processes transcending multiple layers

Employing flexibility (14)

Thinking creatively and flexibly about boundaries was encouraged and supported within teams (e.g. ensuring a buddy system if meeting a client outside of work hours), and by EIP culture.

All participants described the necessity of employing flexibility when working with clients “you’re … drawing on every possible avenue for engagement with that person … that … requires you to think quite creatively and flexibly about how you’re gonna work with somebody” (Participant 9)

Common ways in which boundaries were flexed included: timings of meetings (during clients’ lunch breaks or after they finished work), communication methods, being creative in where participants met with clients and what they did together, which was often aligned to clients’ personal goals.

Employing flexibility around boundaries involved a higher level of cognitive burden for participants, as they were constantly assessing and reassessing the boundaries; judging whether they had got the balance right.

Transparency (15)

Participants aimed for a transparent and empathic approach with clients and colleagues concerning boundaries, particularly when boundary issues arose. Ten participants spoke about being honest about their client interactions, and routinely discussing boundary dilemmas with the team. “you’re being so transparent … we … check in so often about how we’re working with people … No one’s kind of holding secretly … well this is what I do with my patient” (Participant 6).

Several participants could recall occasional instances where they were not entirely open with the team regarding boundary dilemmas. However, because this was an exception to the norm, it signalled to them a warning that they had perhaps crossed too far over a boundary.

Finding a balance (16)

An undercurrent running through all interviews was finding a balance when navigating boundaries. As boundary dilemmas arose, often participants had to “recalibrate” or alter their relationship with clients. This often involved implicitly or explicitly introducing firmer boundaries, or deciding where to draw the line. “I had to … bring in boundaries which he found a bit odd because that hadn’t been how we’re working? but it hadn’t presented an issue until that point … So I then had to … say, well … that’s where the line is” (Participant 7).There were multiple balances to be struck, but most tensions fell within three key areas:

• balancing authenticity and humanness with professionalism

• balancing flexibility with professionalism

• balancing clients’ rights and personal boundaries with assertive engagement, or duty of care/safeguarding duties.

When considering how negotiating these balances related to professional codes of conduct, there was acknowledgement that certain actions or decisions take around boundaries within EIP, whilst usually thought through in terms of motives and rationale, would not always be endorsed by professional bodies or organisational policies.

“I think some of us … make a bit of a dent in that … When it seems to be the best therapeutic outcome … But it’s a dent rather than a break” (Participant 12).

The perforated red line surrounding the interface between organisational policy and EIP culture reflects this.

Discussion

This study has generated an original model conceptualising the navigation of boundaries within EIP services. Findings highlight how case-managers navigate and design boundaries with clients using a flexible, case-by-case approach. The benefits of this approach were that case-managers were able to present their authentic selves, build meaningful relationships, and better support client-centred, goal-oriented recovery for clients experiencing high levels of distress. This was facilitated by features of EIP culture, and team processes. Challenges were overcome through consultation with colleagues, with shared decision-making, reflection, and collaboration around boundary decisions, meaning clinicians felt supported holding this complexity.

The EIP feature of meeting outside of the clinic room, and the looser boundaries this usually entailed raised interesting points around levelling power, and who can set the boundaries, or hold power, in what context. Many case-managers observed that physical environment had an impact upon power dynamics: when meeting clients in their homes, clients’ ownership over this space increased their power regarding boundary setting. Clients could set their own rules (e.g. requesting that case-managers remove their shoes) in a manner which they could not have done within a clinic room. When meeting with clients in public spaces, as case-managers frequently did, dilemmas were described of how, or whether, traditional boundaries could hold, particularly in the context of levelling power. For example, a client and a case-manager having lunch in a café which serves alcohol: does the case-manager still have the right to set the boundaries and wield power? Dictating to a client (who has mental capacity) the choices they make within that public setting, such as not to drink a beer during their meeting (which arguably greatly alters the frame and tone of the meeting). A reasonable solution appeared to be a negotiation, or compromise, in which the participant would convey the preferred boundaries and reasoning for this, but not categorically impose these. This was reflective of the broader principles of employing flexibility, and finding a balance, which case-managers used in their everyday navigation of boundaries in EIP. There was a line however, beyond which more traditional professional boundaries were enforced. Where this line was drawn was dependent on factors such as the “greyness” of the scenario, the individual clinician, and the fragility of the engagement or relationship.

This research raised interesting questions around whose interests or agenda were prioritised. Some participants recognised that their professional boundaries or responsibilities were, perhaps unconsciously, held in higher regard than their clients’ personal boundaries. Sometimes this was justified under current legal frameworks (e.g. if there were legitimate grounds for safeguarding concerns). However, case-managers acknowledged occasions where clients had felt that under the guise of assertive engagement, their personal boundaries had been, at best, encroached upon, or at worst, disrespected, without significant cause for concern but simply because they did not wish to engage with the service. Engagement was a key focus for participants, and assertive engagement/outreach was often the guiding principle behind why, and to what degree, boundaries were flexed. Farrelly and Lester’s review (Farrelly & Lester, 2014) found a similar key focus on engagement as the goal, and the problematic aspects of this, for example, leading to the assumption that clients should engage with services at all times, and “disengagement” portrayed as problematic. Client studies reported this could come at the cost of their own choices, or personal boundaries being overridden, with detrimental impacts on their self-determination and independence.Although there were similar findings in this research, case-managers recognised this danger, having reportedly listened to clients, and learnt from such experiences (although it should be acknowledged this was in the view of case-managers, not the clients themselves). There was reported caution in not disrespecting clients’ own boundaries, and some case-managers mitigated this risk by collaborating with clients to jointly construct appropriate boundaries, and understand where clients drew their personal boundaries. Despite this, knowing where to draw the boundary with the initial stages of engaging, or with duty of care responsibilities was still a challenge.

Another parallel finding with Farrelly and Lester’s review was the dilemma regarding navigating the boundary between supporter and friend. Many of the studies included in the review reported clients expecting to develop friendships with clinicians, perhaps to replace missing social contact. Clinicians in these studies did not always challenge this perception, believing this had therapeutic benefits for the client. However, they avoided perceiving the relationship in these terms themselves, feeling this violated their professional boundaries. However, the more relaxed boundaries in this service context made this pull towards friendship or other forms of relating (such as mothering) harder to resist, particularly when case-managers had good relationships with clients. Whilst participants usually explicitly implemented boundaries when a client’s request for friendship or overstepping a boundary was obvious, when this was done in more subtle ways it felt more difficult to navigate, and participants often responded in more subtle ways too, drawing boundaries more implicitly.

Clinical implications

Several participants sought permission from the client before a potential boundary crossing, for instance, permission to self-disclose, or hug the client. Undertaking a collaborative approach to boundary setting by jointly constructing appropriate boundaries is likely to be helpful where possible. However, it should be acknowledged that given the power differentials in relationships between clients and case-managers, it may be more challenging for clients to assert their wishes meaningfully.

These clinical implications are supported by Gutheil and Gabbard (1998) who note that the difference between a boundary violation or a boundary crossing may lie in whether it can be discussed within the therapeutic relationship, and/or whether permission is first sought for the crossing. Additionally, some participants apologised in instances where clients felt they had overstepped their personal boundaries. On these occasions reparation of the therapeutic relationship was often possible, and this is reflective of Gutheil and Gabbard’s (1998) notion that boundary crossings or violations can sometimes be undone, and appropriate boundaries reinstated through further consideration and discussion with clients.

The challenges associated with navigating assertive engagement when difficulties arise, and the practice of collaborative constructing boundaries is consistent with an Open-Dialogue approach, which has a good evidence-base for FEP (Bergström et al., 2018). Open Dialogue promotes an equal dialogue between clients, families and clinicians to enable a sense of agency. Clinicians work flexibly, and the approach heavily engages the client’s social network (Aaltonen et al., 2011).Through creating a dialogue, or sense of “with-ness” rather than “about-ness”, responding to the needs of the whole person (rather than just symptoms or risk), and listening to what clients and their families say, boundaries and service engagement could potentially be more openly discussed, and decisions reached more equally and collaboratively.

Case managers can come from a range of professional disciplines and the findings of this study suggest that different mental healthcare professions may have different perspectives on boundaries shaped by their training, and the saliency of the consequences of boundary violations (Scott, 2011; Valente, 2017). This, for some participants, led to an initial preference for clear, rigid boundaries to negate any potential risk. An important application to clinical practice was the utility of consulting and collaborating with other professionals, who provided different perspectives on boundaries; offering useful insights into the client’s viewpoint (Davidson et al., 2012), or encouraging participants to reflect on their clinical practice around boundaries, for example, to consider their underlying motives.These processes not only took place within formal structures, such as team meetings and supervision, but also through informal, “watercooler” conversations. Several participants spoke about the loss of this aspect whilst working remotely during COVID-19. This loss might be particularly felt, given the additional cognitive and emotional burden which case-managers experienced in EIP because of the case-by-case, almost meeting-by-meeting, decision-making process around boundaries. The impact of this loss in day-to-day clinical practice around boundaries should be considered, and if remote working is likely to become a greater feature of case-managers’ practise in future, how this could be mitigated. Possible solutions could involve regular check-ins via videoconferencing, as well as ensuring peer supervision or reflective practice spaces are accessible online.Participants in respondent validation interviews as part of this study recognised the value of this model as a clinical tool to aid case-managers’ thinking when making difficult boundary decisions, and for reflection around boundaries, both on an individual basis and within teams.

Limitations and future research

Whilst participants appeared to be very open during interviews, sharing many real-life complex dilemmas, and openly recognising times they may have overstepped or mismanaged boundaries, it is also recognised that had there been significant boundary violations, it is unlikely that participants would have chosen to share such incidents. Participants were only recruited from one NHS trust. It is likely that differences in EIP culture within this particular trust, or other factors such as the broader trust culture, differences in service pressures or the demographic populations in these areas, will impact how case-managers navigated boundaries.

Most participants were White-British. This is not reflective of the clinical population accessing EIP, where there is a disproportionately high number of clients from ethnic minority groups (Gov.uk, 2017), nor is it reflective of the cultural diversity of the NHS workforce in many areas of the country (Gov.uk, 2021). It would have been useful to further explore the impact of cultural norms and cross-cultural interactions between participants and clients in relation to boundaries, particularly when case-managers are visiting clients’ homes. Additionally, how case-managers’ navigation of boundaries was influenced by a shared ethnicity with clients was a point several participants touched upon, as a form of inadvertent or deliberate self-disclosure that aided therapeutic alliance. Unfortunately, the researcher was unable to recruit more participants from ethnically diverse backgrounds to explore this question further.

Findings also suggested a possible relationship between clients with more “positive symptoms” (e.g. unusual experiences or belief systems), and the impact on how case-managers navigated boundaries (feeling a greater protection over clients’ personal boundaries, assertively engaging more, or self-disclosing less). These relationships were complex, and further research could explore them in greater depth.

Most importantly, future research should examine the navigation of boundaries within EIP from the client’s perspective; what is helpful or unhelpful regarding how case-managers navigate boundaries with them, and further exploring strands around who boundaries are for, and how clients’ personal boundaries are understood within this model.

Conclusion

This study aimed to better understand how case-managers in EIP services develop their understanding and practice around navigating boundaries, given the unique service model and client group. Using grounded theory, a concentric circular model emerged, defining different layers of influence impacting case-managers’ navigation of boundaries, including many factors specific to EIP culture. Navigating boundaries within EIP was a dynamic process, and involved many complex challenges. A flexible, case-by-case approach was adopted in response to such challenges. Findings suggest clinical implications for how case-managers can jointly construct appropriate boundaries with clients, and discuss dilemmas using the team as a resource. The model created can be used as a tool in clinical practice to support clinicians’ thinking and reflection around boundaries within EIP services.

Acknowledgments

Thank you to all my participants, who shared their time and valuable expertise with me. I am incredibly grateful to the second author, Rachel, for your encouragement, guidance and support throughout the project and publication process, and to the third author, Becky, for your enthusiasm with the initial idea for the project, support with recruitment, and contributions to this article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Canterbury Christchurch University