Open dialogue in the UK: qualitative study.

BACKGROUND: Open dialogue is an integrative approach to the organisation of specialist mental health services and therapeutic meetings.AimsThis qualitative study sought to explore service users' and clinicians' experiences of network meetings during the implementation of open dialogue in a modified version, for a UK-based mental health service.
METHOD: In total 19 participants were interviewed (8 service users and 11 clinicians) and an inductive thematic analysis of the data was conducted.
RESULTS: Four dominant themes were identified: (1) open dialogue delivery, (2) the impact of open dialogue principles; (3) intense interactions and enhanced communication, and (4) organisational challenges. Clinicians considered open dialogue as a preferred, but challenging way of working, while being therapeutic. The data indicated that service users' experiences of network meetings were mixed. There was a wide variety of service user views as to what the purpose of a network meeting was and for some witnessing reflective conversations felt strange. However, the majority described feeling listened to and understood, excluding one service user who described their experience as distressing. Clinicians expressed an authentic self in their interactions with service users and both service users and clinicians described network meetings as emotionally expressive, although this was described as overwhelming at times.
CONCLUSIONS: The results of this thematic analysis indicate that service users' and clinicians' experiences of open dialogue warrant further investigation. The intensity of interactions in network meetings should be carefully considered with service users before gaining consent to commence treatment. Implementation of open dialogue should be monitored to assess clinician- and service-level adherence to the principles of the approach.Declaration of interestNone.

Open dialogue is an approach to specialist mental healthcare that originated from the need-adapted approach in Western Lapland, Finland during the 1980s and is both a way of being with people and of organising services. Open dialogue is an integrative approach that embodies systemic family therapy[1] and mobilises the psychosocial resources in a service user's network during a mental health crisis.[2] Open dialogue favours working with service users in the community over admission to hospital and aims to redress power imbalances between the service user and clinician by facilitating autonomy and transparent decision-making. The approach is being implemented in the USA, UK, Scandinavia, Germany, Italy, Austria, Australia, Japan and Poland. Implementation of open dialogue presents particular challenges, as it requires changes to the way services are organised and delivered. Published studies suggest that open dialogue provides benefits for service users on a wide variety of outcomes, however, the approach has yet to be evaluated in a randomised controlled trial, and further studies are needed to draw conclusions about its effectiveness.[3] Further robust studies evaluating the utility of open dialogue are ongoing. Presently, there is limited understanding of clinicians’ and service users' experience of implementing open dialogue or the therapeutic meetings, termed network meeting, which form a central part of open dialogue treatment. The current study aims to inform open dialogue implementation, by conducting a thematic analysis of qualitative data collected from clinicians and service users regarding the application of open dialogue. Open dialogue as delivered in this programme was a version modified for the UK's National Health Service (NHS), which included peer-support workers, mindfulness practice for clinicians and a specific supervision framework.[4] Given the uptake of open dialogue, there is a need to better understand the barriers to implementation and how service users and clinicians experience the approach.

Method

Service context

The data for the current study were gathered from an NHS trust community mental health service situated in a borough of London, UK. The study was conducted in the early phases of a programme grant to evaluate the effectiveness of open dialogue. As this study was conducted early on in the implementation phase of the approach, it was new to both service users and clinicians. The service comprised of four specialist care psychosis teams. Each team was multidisciplinary, and professionals predominantly provided care-coordination and medication management. The service enrolled 13 of its clinicians (psychologists, family therapists, psychiatric nurses and psychiatrists) on a foundation diploma training course on open dialogue. At the point of the data collection the clinicians were halfway through training. One clinician was a peer worker and worked across all teams. All the open dialogue clinicians came together once a month for supervision facilitated by an external open dialogue practitioner.

Design

A qualitative design was used to cultivate a rich understanding of how service users and clinicians experienced open dialogue. An inductive analysis guided by the principles of thematic analysis was undertaken.[5]

Ethics

This study was performed in accordance with the Declaration of Helsinki. Ethics approval was not required for this study because the research and development office of the NHS Trust in which it was carried out approved it as an audit. All participants provided written informed consent for the publication of the material arising from the study and were informed of their right to withdraw their consent. The procedures specified by the British Psychological Society's Code of Ethics and Conduct were also followed.

Procedure

In total 5 service users, 3 network members and 11 clinicians participated. Sampling was guided by purposive and convenience approaches.[6-8] R.H.T. and A.M.F. contacted all 13 clinicians in open dialogue training inviting them to be interviewed about their experiences of network meetings and delivering open dialogue, two were unavailable for an interview. The mean age of clinicians who participated was 44.80 years (s.d. = 8.66), and their ethnic backgrounds included: White British (n = 8), Black British (n = 2) and White other (n = 1). R.H.T. and A.M.F. asked clinicians for details of all the service users who had participated in network meetings since its introduction to the service (n = 18).

R.H.T. and A.M.F. invited each service user to participate in the study following a case-by-case discussion of their suitability during clinical supervision with S.L. Inclusion criteria stipulated that the participant was a network member and had participated in at least one network meeting in the past six months. Service users were excluded if they were unable to consent to take part in the audit. Four potential participants were excluded from the study using these criteria following a deterioration in their mental health (for example, because they were being assessed under the Mental Health Act for an involuntary admission to hospital). Five service users were unable to be contacted or declined to take part. Eight service users including three family members were interviewed; their mean age was 39.75 years (s.d. = 16.73). Their ethnic backgrounds included: White British (n = 4), Asian British (n = 2) and White other (n = 2). The results from service users and family members were combined during analysis because of the size of the sample, and both groups are referred to as service users hereafter.

Interview schedule

Scoping discussions were held with two experienced external open dialogue clinicians to construct the semi-structured interview schedule. This addressed the topics of network meetings, critical moments within network meetings, positive and negative aspects of network meetings, relational dynamics between staff and implementation issues. All interviews were audio-recorded and transcribed. Clinicians commented on their own adherence to the model without formally rating or recording it.

Data analysis

Thematic analysis was used to identify and categorise key themes in the data.[5,9] An inductive theoretical approach was taken, meaning that the themes were strongly linked to the data. Where possible the authors suspended prior knowledge of the phenomenon under investigation, in a process termed ‘bracketing’.[10] The thematic analysis followed the six-phase method outlined by Braun and Clarke.[5] Each transcript was read by R.H.T. and A.M.F. several times and relevant ideas were hand-coded using open coding, checked across the entire data-set, and then collated to form higher-order themes and subthemes. The themes were then identified and reviewed against the raw data. Analysis continued until no new themes emerged, indicating that saturation had been reached. Coding reliability was checked by R.H.T. and A.M.F. by independently coding several transcripts and cross-checking results.

Validity

Evaluative guidelines[11] suggest that the authors' personal orientation to open dialogue be disclosed: R.H.T., A.M.F., J.C.H.S. and S.P. were not open dialogue practitioners. S.L. was an open dialogue trainee clinician at the time the study was conducted. Thorough credibility checks[5] were undertaken by using multiple analysts and refining themes and subthemes through discussions between R.H.T., A.M.F., J.C.H.S. and S.P. Emergent themes were presented to clinicians for member checking and were supported; specific feedback was given and incorporated into the analysis.

Results

See Appendix for an overview of the themes and subthemes, including the variations within each for service users and clinicians.

Dominant theme 1: open dialogue delivery

Subtheme 1.1: open dialogue as a positive change

The data indicated that the majority of participants experienced network meetings as different to usual care. Service users spontaneously made comparisons between previous experiences of mental healthcare and referred to their negative experiences of meetings within the service before network meetings were introduced; these included not feeling listened to, being coerced or experiencing an unwelcome focus on medication. There was a sense among clinicians that open dialogue aligned with their professional values and improved the quality of care they could provide compared with treatment as usual (TAU). Clinicians appreciated the opportunity to talk with service users under less time pressure, and to work in a way that felt comparatively more respectful of service users.

‘I mean I do really enjoy listening to people and hearing their stories and yeah it's definitely a really nice way to work it's definitely more humane, really, definitely it's kinder, I think it's kinder because… because you, what's the word, you devote the time to it, you don't just rush off, the meeting comes to a natural end.’ (Participant 11, clinician)

Subtheme 1.2: impact of reflections

The reflective conversations held between clinicians within network meetings stood out as a new way of delivering treatment and were experienced differently by clinicians and service users. Clinicians felt that the reflections they made had powerful therapeutic potential through the amplification of ideas and emotions, and made it possible for service users to hear their own perspective from different positions. The reflections were also a useful way of sharing new perspectives and topics that had not yet been disclosed. The data indicated that for half of the service users the reflections felt ‘difficult to get used to’, ‘weird’, ‘odd’, and ‘strange’ but not ‘unpleasant’. Another service user described the reflective conversation as a ‘meta meeting,’ and others said that it led to a ‘parent/child dynamic’ and grappled with the authenticity of the reflective communications. Clinicians had an awareness of the powerful and varied ways reflections could be interpreted, and some noticed that service users found this way of working strange:

‘Yeah, I found that [reflective conversation] a bit weird, yeah it didn't, I wasn't prepared for it, I didn't really know, it just felt strange, like you're in an open meeting where two people were just having a conversation themselves and we were all just sitting there sort of not knowing what to do, so it was just a little bit, that might be a function of it being a new thing or not being explained to the people, or we're not used to having those kind of meetings.’ (Participant 13, service user)

‘It was almost nice to kind of hear what they were thinking because I trust that they were being sincere about what they were saying.’ (Participant 18, service user)

‘They've [service users] seemed quite disarmed by this different approach and have looked a bit, kind of, what's going on? It's obvious they're experiencing that this is a different dynamic.’ (Participant 6, clinician)

One service user had a distressing experience within network meetings and found the reflections ‘embarrassing’. For this participant, clinicians were talking ‘among themselves’ and ‘feigning compassion’, highlighting how vital it is to continually monitor the impact of reflections on service users.

Dominant theme 2: impact of open dialogue principles

Subtheme 2.1: impact of uncertainty

Tolerating uncertainty is a basic principle of open dialogue[12] and describes how clinicians try to avoid premature decisions and treatment plans.

‘Listening in a way that is different in the sense that there is no real following of agenda, so it's not like that's my bit that's their bit.’ (Participant 1, clinician)

Clinicians reported that taking a step back from having an agenda or offering solutions in meetings was important, but challenging at times, noting that this meant stepping out of the ‘expert role’ and allowing service users to come up with solutions. Clinicians suggested that removing the agenda from meetings slowed the pace and created a space for something else to happen. Perhaps as a response to this, there was a wide variety of service user views as to what the purpose of a network meeting was, some were unsure about what was expected of them and what to expect:

‘I don't really know what they [the clinicians] wanted out of it.’ (Participant 18, service user)

‘I just don't think it achieved anything, it wasn't really clearly defined what we were trying to do, it probably was at the start but it just didn't end up that way.’ (Participant 13, service user)

A lack of focus appeared to contribute to the feeling that the meetings were unfocused and unproductive. Other service users described the network meetings as a space to seek advice, to be evaluated or to have therapy. One service user stated that the purpose of the network meetings was therapeutic therefore oriented his level of expression accordingly and expected to get ‘emotionally involved’ to address the core of his issues. Another stopped talking about his past at times because he ‘knew it was not therapy’.

Subtheme 2.2: impact of dialogism

Dialogism,[13] a core principle of open dialogue, describes how clinicians attempt to find a common language for distress by actively listening and responding to every utterance. Meanings are generated and transformed from response to response leading to richer possibilities for understanding and adopting new points of view.[2] Unsurprisingly, clinicians referred to the use and impact of dialogism, and felt that the approach enabled them to actively listen to service users. The data from both service users and clinicians suggested that multiple perspectives were given space within network meetings. For the majority of service users, this appeared to result in a feeling of being listened to and understood. Participant 14, for example, had a clear sense that he was being heard:

‘He was a good listener for a doctor… you get old school doctors they can be, a little bit bombastic they don't really let you tell them everything, they tend to, if you like take over. He didn't do that he sat down and listened to what was happening in the family.’ (Participant 14, service user)

This participant later noted that he felt respected throughout the meeting as the doctor ‘wanted to understand’ and this had a large impact on how he felt within the meeting and about the clinicians themselves.

Dominant theme 3: intense interactions and enhanced communication

Large portions of the data were organised around the theme of communication. Authenticity and emotional expression were identified as subthemes and at times facilitated intense interactions within the network meetings, surprising both service users and clinicians.

Subtheme 3.1: authenticity

Clinicians discussed how they were able to bring their authentic opinions, emotions and selves into meetings, and talked about how this created a sense of openness. Some service users noted that clinicians seemed more open in network meetings than they had been in other meetings. When talking about these experiences, both service users and clinicians described meetings as ‘emotional’ and ‘intense.’ Clinicians posited that their increased sense of openness was associated with their ability to be more authentic in network meetings. However, at times some service users questioned the sincerity of these interactions:

‘I assume that whenever they were taking time out kind of processing things that they were kind of putting on a bit of a show. And being like were gunna' talk about you quite nicely to sort of prove to you that we do think nicely about you but we don't really think nicely about you – but that was all me. And yeah most likely they were just, they were literally just trying to process what happened which I appreciated. They seemed really engaged in what both me and my sister were saying and just trying to make sense of our story which was nice.’ (Participant 18, service user)

For others, the new, unexpected levels of authenticity and emotional expression felt uncomfortable. Service users noted that the clinicians were more open and actively encouraged openness within the meeting, as well as ‘honesty’; some of the participants oscillated between wanting to be open and wanting a return to privacy. The service user who experienced the network meetings as distressing described feeling too unsafe to be open and believed that the clinicians were pretending to be open-minded. Clinicians perceived their interactions with service users as being more authentic as they brought more of themselves into the conversation by drawing on personal as well as professional experience. The clinicians appeared more struck by the change in themselves than in the service users, whereas service users tended to reflect on the shift in authenticity in both themselves and clinicians.

‘What I really value is that people [clinicians] step out of the shadow that they start becoming more, more active that they start being more real, a real person than actually in a role “oh I'm the care coordinator, just sitting here, taking notes, showing my face”.’ (Participant 9, clinician)

By connecting more with who they were as people, the data suggested that the clinicians entered into a more authentic form of communication with service users:

‘I think we are bringing a lot of ourselves, and I think somehow that is being appreciated because I think some of those very honest reflections and reactions are touching somewhere.’ (Participant 8, clinician)

Subtheme 3.2: emotional expression

The data suggested that the sense of openness in meetings included the fluid exchange of emotions. Both clinicians and service users frequently commented that feelings were expressed more freely in network meetings in comparison to usual care. Another clinician described the process of containing the reciprocal sharing of emotions as a ‘curative element’ in meetings. Clinicians generally said that the expression of emotions in network meetings was positive and had therapeutic value, as did half of the service users, even though they also described this experience as overwhelming at times. More than half of the service users described network meetings as emotional, as the content of the meetings connected to issues that were meaningful for the network.

‘It seems to have led to quite a lot of expression of emotion I suppose, erm and I suppose I tend to think if people are getting emotional in sessions then it means that we are talking about things that are important.’ (Participant 10, clinician)

‘It got quite emotional I think, it was quite hard because I guess, my partner and I were talking in quite a much more… candid way because I guess we felt we could.’ (Participant 17, service user)

The environment created within the network meeting appeared to act as a vessel for the exchange of meaningful, emotional content. This was not always welcomed; a smaller proportion of the service users found the emotional openness of clinicians uncomfortable, the purpose of such expressions was often unclear. It left one service user feeling ‘wary’ and ‘quiet’. Two service users also found bringing forward their own opinions and emotions difficult because of the presence of family members.

‘None of us felt like we could be as open as we wanted ’cos we were basically discussing my mum's illness while she was there and she hijacked the meeting so it was hard to be honest.’ (Participant 13, service user)

Dominant theme 4: organisational challenges

Subtheme 4.1: lack of resources and process

Eight clinicians reported that further organisational changes in line with the principles of open dialogue were needed to support the delivery of network meetings. Issues raised included a lack of sufficient resources, limited or disjointed systems for monitoring service users and a lack of clear policies and procedures. This was discussed by clinicians who referred to the competing priorities in the context of stretched NHS mental health trusts.

Subtheme 4.2: challenging hierarchical structures

Open dialogue was perceived as challenging the traditional hierarchical structure across individual, organisational and professional levels. Some clinicians felt that the introduction of open dialogue was an opportunity to change oppressive hierarchical structures in the service. Divergent and conflicting discourses emerged depending on whether the implementation of open dialogue to the service was seen as successfully able to challenge hierarchical structures. Some clinicians felt that open dialogue would allow more voices in the teams to be heard. However, opinions differed about how successfully the service had started to change:

‘I really am not sure about this flattening of hierarchies, I really can't see the NHS or any organisation even families [flattening hierarchies]. Let's face it. Hierarchies exist.’ (Participant 2, clinician)

‘I understand why the strategy group needs to be a bit more hierarchical, but things like the events, you know to hear about something after its happened it undermines the very nature of open dialogue and I think it has put some people off, some people are shying away now and thinking is this just a vanity project for the seniors in the trust, sadly that is the way that some people are talking.’ (Participant 7, clinician)

This contrasted with the views of some clinicians who felt that hierarchies could be flattened, thereby providing an opportunity for a major shift in mental health services that may otherwise be missed. Clinicians also reported conflicting priorities between the ‘politicised’ NHS and the open dialogue approach:

‘I think if I miss something, or drop a KPI [key performance indicator] and I'm spending time doing network meetings I don't feel that the Trust will support me, you know I feel I will be hammered, you know, because KPI's rule.’ (Participant 11, clinician)

Subtheme 4.3: lack of wider implementation

The resistance of non-open dialogue trained clinicians to embrace the open dialogue approach was evident in the data from clinicians, who suggested that other team members had questioned the aims of the project. Around half of the clinicians interviewed reported that the management and implementation of the approach were not as dialogic as it could be:

‘I think they do, the clients do, the family members do, the non-trained team members, clearly feel the buzz and excitement of doing things differently and being asked to do things differently and they enjoy that but it's also, you know, it does transpire, it changes the dynamic with the other team, the home treatment team and so on, there is a sense of things changing but not uniformly, [it's all] a relief and positive – it's not like that, it has quite a lot of jealousy and quite a bit of resistance around.’ (Participant 9, clinician)

‘[There needs to be] more emphasis around bringing people in and, listening to people's apprehension.’ (Participant 1, clinician)

Discussion

This study aimed to explore clinician and service-user experiences of network meetings in the early implementation phase of open dialogue within an NHS setting to inform further implementation. The study led to the identification of four dominant themes. These were: (1) open dialogue delivery, (2) the impact of open dialogue principles, (3) intense interactions and enhanced communication, and (4) organisational challenges. The data demonstrated links between clinician experience of delivering open dialogue and how service users received this. There were a number of variations and contradictions within the subthemes between the experiences of clinicians and service users that will now be expanded upon.

Delivery and monitoring open dialogue

Participants experienced network meetings as different from TAU. Clinicians, in particular, were extremely positive about the approach and used powerful language to denote their preference of open dialogue in comparison with TAU. Service users frequently referenced their previous negative experiences within mental healthcare and reflected on the increase in positive communicative experiences in network meetings.

Clinicians were mostly positive about the level of openness that the structure of network meetings afforded them, however, the service users who also noted the openness within network meetings, sometimes found it uncomfortable. Some clinicians spontaneously described carefully monitoring the effects of their own practice and raised concerns about causing harm if the approach was not delivered appropriately and sensitively. It is important to note that one service user found network meetings distressing, highlighting the need for formal evaluation of service-user experience, clinician adherence and fidelity to the model, and the suitability of the approach on a case-by-case basis.

The open dialogue model includes the principle of flexibility that encompasses this type of case-by-case approach. Flexibility is also a strongly held principle in the need-adapted approach to treatment in which open dialogue originated. For both approaches, flexibility ensures that the therapeutic intervention is individually planned and adapted to each service user to meet their changing needs, therefore adherence to this principle in particular is extremely important.

Overall, there was a sense among clinicians that open dialogue aligns with their professional values and improves the quality of care they can provide, a finding echoed by Stockmann and colleagues in their thematic analysis of clinician experience of open dialogue training.[14] The high level of investment and commitment to open dialogue from clinicians in the current study may be an important factor for the sustainability of the approach in the service. Similarly, a Norwegian study by Neilsen cited by Buus and colleagues[15] demonstrated that clinicians ‘believed that open dialogue had improved their professional attitude.’

Impact of open dialogue principles

The open dialogue principles of dialogism and tolerating uncertainty had a large impact on participants. The clinicians referenced how they were able to tolerate uncertainty by not having a focused agenda, which may have left some service users confused as to the purpose of meetings. Service users experienced network meetings as unusual and clinicians noticed that service users at times seemed disarmed. Clearly outlining the purpose of meetings before initiating treatment may help service users to make sense of the approach and their interactions with clinicians. Dialogism was associated with multiple perspectives being elicited and heard, and data suggested that both groups experienced this as missing in TAU. For the majority of service users, the data indicated that they felt listened to. This may have resulted from the space created by clinicians to share multiple perspectives and understandings, which captures the aims of dialogism. Tolerating uncertainty and dialogism were experienced as positive aspects of the approach by clinicians.

Intense interactions and enhanced communication

Clinicians felt they had made large changes to the way they had communicated with service users, and correspondingly service users felt a shift in the depth of communication achieved. There was a strong sense that the network meetings involved a level of self-disclosure and emotional expression that neither service users nor clinicians were familiar with. For some service users and clinicians this led to a sense of connection and a level of intensity that felt overwhelming. The compelling nature of the relationships formed through open dialogue and the intensity of the interactions in comparison with TAU should be carefully considered and outlined with service users before gaining consent to commence treatment. A crucial element of open dialogue was the felt and perceived authenticity of interactions within network meetings. It will be important for further studies to look at how authenticity has an impact on the therapeutic relationship and whether this fosters trust. Previous studies of open dialogue have shown that the increased trust between service users and therapists may be an essential mechanism of change within open dialogue.[16-18]

Organisational issues

It may be important to consider whether additional resources are needed for an open dialogue approach to service delivery. Clinicians described the implementation of open dialogue as challenging and called for a more dialogical approach to be taken. They suggested that leadership take a more transparent and inclusive approach to changes in service delivery, which would adhere more closely to the open dialogue approach. When implementing complex clinical interventions, it is important for those delivering the service to collectively define the procedures and actions needed to maintain changes, in order to encourage continued involvement and commitment.[19]

Critical to implementation is the culture of the service, which is heavily influenced by where power lies and how it is used. In the current study, several clinicians identified challenges in delivering a healthcare approach where there is a flattening of hierarchical structures. Clinicians referenced the fixed hierarchy and power dynamics in NHS mental health services as an integral part of the way teams are structured and a function of how services monitor work and divide responsibilities. It is therefore important to further explore how to adopt a radically different approach to service delivery in the NHS and other international healthcare providers. Razzaque and Wood[20] suggest that a cultural shift in NHS services is needed to allow open dialogue to be incorporated into practices and that open dialogue may offer a framework to develop services in a clinically meaningful way.

A previous review of the implementation of open dialogue in Norway[21] suggested that a dialogical approach to the implementation process itself may be necessary for its successful integration into a service. This would require a bottom-up approach by including and acting on the opinions and concerns of service users, peer workers, paraprofessionals and professionals lower in the hierachical structure. Clinicians reported that discussions and decisions needed to be more open and transparent, and that the concerns of team members needed to be voiced and heard. In previous studies outside of Western Lapland,[22] open dialogue challenged the traditional working roles and hierarchies that were already in place.

There is little published information about how the teams in Western Lapland, Finland managed these challenges in the implementation phases of open dialogue. Further studies of open dialogue need to actively assess and evaluate whether the implementation phase adheres to its principles. Careful consideration is required to elicit feedback about the leadership and delivery of the service-level changes from clinicians. A process of external consultation may be important to assist this process.

Research recommendations

An important finding of this study is that one service user experienced network meetings as distressing and half described the reflective conversations as strange and sometimes uncomfortable. It will be important for future research to evaluate the potential negative impact of open dialogue informed network meetings and the potential contributing factors, including issues relating to the implementation of the approach. These questions need to be formally addressed in larger qualitative and quantitative studies. The response from some service users who noted that they felt heard and understood in network meetings indicates the potential therapeutic value of open dialogue and suggests that a randomised control trial is required to compare open dialogue with TAU. Importantly, clinicians said that open dialogue allowed them to behave in ways that they felt were more closely aligned with their professional values than current structures and methods of operating allowed. In this context, there is value in examining how the culture of services can be shaped to enable clinicians to act more consistently in line with their professional values to improve levels of staff retention and reduce burnout. Research is required to determine whether the challenges to implementation highlighted in the current study are a result of the culture at this specific site. Additionally, issues related to the peer-supported element of the approach were not prominent in the interview data. This is an interesting finding and may reflect the number of peer-support workers at the site which was only one; further research is required to determine whether there are barriers in actualising this element of open dialogue.

More broadly, research is needed to explore the relationships between the experience of open dialogue and service users' cultural, social and community context. The variation in cultural perspectives on the aetiology of mental health distress,[23] approaches to healing this distress and the levels of deprivation in treatment catchment areas may have an impact on service users' ability to ‘engage’ in treatment and ultimately ‘recover’.[24] Access to one's network may also differ dramatically depending on complex contextual factors such as one's cultural beliefs or residential status in the country where treatment is delivered.

Strengths and limitations

The study was conducted in the context of the early phases of a programme grant to evaluate the effectiveness of open dialogue. As such, the clinicians facilitating network meetings were part-way through their training in open dialogue; therefore, caution is needed before generalising to other well-established open dialogue services. The sample size is small, particularly for service users, and generalisations to the wider population are not intended. The sample also lacked gender diversity as service users were predominantly men and clinicians were predominately women, although this reflects the demographics of the service users and mental health professionals more generally in the service. Clinicians identified participants, which may have introduced bias into the sample.

Dominant themes and subthemes

Dominant theme	Subthemes	Clinicians and service-users variation on subthemes
		Clinicians	Service users
1. Open dialogue delivery	1.1 Open dialogue as a positive change	Open dialogue as a preferred way of working	Previous experiences of treatment unhelpful
	1.2 Impact of reflections	Reflections as therapeutic	Reflections as strange
2. Impact of open dialogue principles	2.1 Impact of uncertainty	Uncertainty as challenging	Uncertainty as unsettling
	2.2 Impact of dialogism	Eliciting multiple perspectives	Feeling listened to and understood
3. Intense interactions and enhanced communication	3.1 Authenticity	Authenticity of self	Authenticity of self and perceived authenticity of clinician
	3.2 Emotional expression	Heightened self-disclosure	Heightened emotional expression
4. Organisational challenges (clinicians only)	4.1 Lack of resources and processes
	4.2 Challenging hierarchical structures
	4.3 Lack of wider implementation