Exploring service users’ and practitioners’ priorities regarding outcomes of cognitive behavioural therapy for distressing voices: a thematic analysis

ABSTRACT

Background

The outcomes of Cognitive Behavioural Therapy for distressing voices and their measurement have mostly been determined by professionals, with little focus on the outcomes that matter to people who hear voices. The aim of this qualitative study was to explore views about the outcomes of Cognitive Behavioural Therapy for distressing voices from the perspectives of lived and professional experience.

Methods

Fifteen service users and fourteen practitioners were recruited from mental health services. Semi-structured interview schedules were used to guide individual interviews with service users and focus groups with practitioners. All interviews were audio recorded and transcribed verbatim. Thematic analysis was applied to generate themes.

Results

Four themes were identified: “I want what I can’t have”, “Feeling empowered”, “Understanding and managing my emotions”, and “Doing more of what I want to do”. Two themes contained sub-themes.

Discussion

Valued outcomes by service users and practitioners appeared to reflect changes in voice characteristics, improvements in self-confidence and self-respect, acceptance of voices, understanding and managing emotions and personal outcomes. Findings suggest discrepancies between service users’ and practitioners’ priorities in regard to changes in voice characteristics and understanding and managing negative emotions. Recommendations include a greater focus on managing emotions by practitioners.

KEYWORDS:

• Auditory hallucinations
• voice hearing
• cognitive behavioural therapy
• outcomes
• qualitative research

Introduction

Cognitive Behavioural Therapy (CBT) aims to help people with psychosis to make sense of their experiences and to change unhelpful beliefs that contribute to distress (Johns et al., 2020). It has been recommended as a first-line psychological therapy for the treatment of distressing voices (National Institute for Health and Care Excellence, 2014). However, recent trials of symptom-specific CBT for distressing voices have used a range of outcomes and measures indicating differing perspectives about what should be targeted and measured, in both research and routine clinical practice (Lincoln & Peters, 2019; Loizou et al., 2022; Thomas et al., 2014). It should be noted that these outcomes have mostly been determined by professionals, with little focus on lived experience perspectives; consequently, they may not be representative of the changes or benefits sought by the voice hearers themselves. Hence, it is important to determine a set of measurable outcomes, valued by both voice hearers and practitioners, which can facilitate the comparability of findings across trials and inform the effectiveness of these interventions (Lincoln & Peters, 2019; Loizou et al., 2022; Thomas, 2015; Thomas et al., 2014).

Previous research has shown that service user perspectives can be overshadowed by professionals (Berry & Hayward, 2011). This is particularly important as there are often disagreements between these stakeholder groups regarding priorities for treatment, emphasising the need to take the views of service users into consideration (Bédard et al., 2005; Byrne et al., 2010; Hansen et al., 2004). Qualitative research into outcomes and experiences of CBT for psychosis identified a range of outcomes and mechanisms of change including increased independence and self-esteem, improved understanding and acceptance of the experience and improved social and functional ability (Byrne et al., 2010; Greenwood et al., 2010; Kilbride et al., 2013; Wood et al., 2015). Qualitative research into process of recovery in relation to psychosis also identified several key aspects such as increased agency and overcoming stigma (Wood & Alsawy, 2018). However, these variables may not be applicable to the experience of hearing distressing voices across diagnoses. In this regard, a recent systematic review of qualitative literature has found similarities between distressing voices and psychosis literature (Hall et al., 2022). Important aspects of recovery in voice hearing included making sense of experiences, changing beliefs and relationships with voices, connecting with others and feeling empowered. Yet, there are existing gaps as to what outcomes are important to voice hearers that may have not been addressed in CBT and whether service users and practitioners hold similar views.

Therefore, it is important to conduct a qualitative study with the aim of identifying a set of outcomes that are prioritised by voice hearers and practitioners. The current study aimed to explore views and perspectives of voice hearers and practitioners about the outcomes of CBT for distressing voices that should be prioritised. The following questions were investigated:

1. What outcomes of CBT for distressing voices are prioritised by voice hearers?

2. What outcomes of CBT for distressing voices are prioritised by practitioners?

Methods

Design

Individual and focus group semi-structured interviews were conducted with service users and practitioners, respectively. Individual interviews with service users were chosen to provide a private space to allow for in-depth exploration of their experiences and focus groups with practitioners were chosen to allow for collaborative discussions. Focus groups were small to manage group dynamics and to allow for in-depth discussions among practitioners.

Participants

Twenty-nine participants (15 service users, 14 practitioners) were recruited from secondary mental health services across Sussex NHS Foundation Trust (SPFT). Service user and practitioner characteristics are reported in Tables 1 and 2. Service users met the following criteria: a) aged 18 years and above, b) attended at least 50% of a course of CBT for distressing voices within the past 12 months, and c) were able to talk about their experience of CBT for distressing voices. Service users were excluded if they were unable to sufficiently communicate in English or could not give informed consent. There was no restriction on diagnosis, as voice hearing is often seen across a range of mental health difficulties (Waters & Fernyhough, 2017).

Table 1. Service user characteristics.

Participant	Age	Gender	Ethnicity	Diagnosis	Therapy
1	36	Female	WB	EUPD	4 sessions GIVE
2	58	Male	WB	Psychosis	6 sessions GIVE
3	48	Female	WB	Psychosis	16 sessions Relating Therapy
4	53	Male	WB	Psychosis	12 sessions GIVE
5	32	Female	WB	Psychosis, EUPD, Autism	9 sessions GIVE
6	36	Female	WB	EUPD	8 sessions GIVE
7	26	Female	WB	Psychosis	8 sessions GIVE
8	32	Female	WB	Psychosis	8 sessions GIVE
9	48	Male	WB	EUPD, Psychosis	11 sessions PBCT
10	50	Male	WB	Psychosis	9 sessions PBCT
11	58	Female	WB	Psychotic Depression	12 sessions PBCT, 3 booster sessions of Relating Therapy
12	66	Female	WB	Psychosis	8 sessions GIVE
13	56	Female	WB	Depressive Disorder, Post Traumatic Stress Disorder	4 booster sessions Relating Therapy
14	46	Female	Mixed	Paranoid Schizophrenia	6 sessions GIVE
15	34	Female	WB	Psychosis, Autism	8 sessions GIVE

EUPD = Emotionally Unstable Personality Disorder, GIVE = Guided self-help cognitive behavioural therapy, PBCT = Person-based cognitive therapy, WB = White British.

Table 2. Practitioner characteristics.

Focus groups	Practitioners	Gender	Role
1	1	Female	Clinical Lead
1	2	Female	Clinician
1	3	Female	Mental Health Practitioner
2	4	Female	Research Assistant
2	5	Female	Mental Health Practitioner
2	6	Female	Mental Health Practitioner
2	7	Male	Support Worker
2	8	Female	Support Worker
3	9	Female	Research Assistant
3	10	Female	Principal Clinical Psychologist
3	11	Female	Clinical Psychologist
3	12	Female	Trainee Psychological Therapist
3	13	Female	Occupational Therapist
3	14	Male	CBT therapist

Practitioners were deemed eligible if they had been trained and had varying experience of delivering CBT-based interventions for distressing voices to ensure that there were diverse views. These included highly trained therapists with extensive experience of delivering CBT for distressing voices and briefly trained therapists who had been trained in an effort to increase access to evidence-informed treatments (Clarke et al., 2021).

Materials

Two semi-structured interview schedules were developed by the first author to guide individual interviews with service users and focus groups with practitioners (interview schedules can be obtained on request by email from the corresponding author) and were revised accordingly based on feedback from the last author. A self-report Equality Diversity and Inclusion (EDI) survey was also adapted to identify participant characteristics. The team sought consultation from lived experience experts on the Service User Interview Schedule. Following feedback, a preamble was included to check the person’s understanding and minor changes to the wording were made.

Procedure

Service users were referred to the study by staff from the Sussex Voices Clinic (SVC), a specialist service for the psychological treatment of distressing voices within SPFT. The Participant Information Sheet (PIS), Consent Form and the EDI survey were sent either electronically or via post to those who were interested in taking part. Service users were given at least 24 hours after receiving the PIS to consider whether they would like to take part in the study. Eligible service users were given the opportunity to provide either written (electronically or via post) or verbal (over Zoom) informed consent.

Practitioners from SPFT secondary care mental health services who met eligibility criteria and were known to the SVC team were identified by the last author and contacted to inquire regarding their potential interest. The PIS, Consent Form and the EDI survey were sent to those who expressed an interest in taking part in the focus groups. Eligible practitioners provided informed consent electronically.

Separate virtual meetings were arranged between service users and the lead researcher over Zoom to conduct individual interviews (range 30–65 minutes). Virtual meetings were also arranged between practitioners and the lead researcher over Zoom to conduct three focus groups (range 50–60 minutes). All interviews were recorded and transcribed verbatim. Following data analysis service users who consented to be contacted again at a later stage were asked to provide their feedback on the generated themes. Practitioners were not asked to provide their feedback due to time constraints from both the authors’ and practitioners’ standpoints. All service users were reimbursed for their time taking part in the study.

Analysis

Reflexive thematic analysis was used to identify themes in the data (Braun & Clarke, 2006, 2021). NVivo version 1.6.2 was used to code and analyse data. Thematic analysis was informed by a critical realist approach (Fletcher, 2017) to centre the professional and/or lived experience of the participants and to consider the mechanisms that influence the identified themes. Critical realism integrates realist ontology and epistemological relativism, recognising that although there is one “true” reality, multiple perspectives exist that are socially constructed.

Reflexive positioning

The first author, who was involved in the design, data collection and analysis, is a PhD student in the area of CBT and voice hearing. The remaining authors are Clinical Psychologists and supervisors of the PhD and have provided input throughout the study. The second author is an expert in psychosis and CBT. The last author is an expert in CBT for distressing voices. Whilst we largely took an inductive approach, all authors have a cognitive-behavioural background. Hence, it is likely that the knowledge, biases, background and views of the authors influenced the interpretation of the results such as the focus on outcomes relating to voice- and self-related appraisals.

Results

Thematic analysis generated four themes, two of which contained sub-themes.

Theme 1: I want what I can’t have

Eight out of 15 (53%) service users reported desiring changes in voice characteristics, as the voices were interfering with their lives and causing them distress, whilst none of the practitioners highlighted this as an important outcome, as they all had a background in CBT, where changes in voice characteristics are not considered a desired outcome. Many service users wanted their voices to be less loud or for them to go away:

To hear the voices a lot less, to get more days off than days I hear them and for them not to scream at all, because they always start off screaming at me and then they quieten down to shouting. (Service user 13)

Whilst some service users sought changes in the content of the voices:

I’d like to turn the negative into positive with this voice. (Service user 15)

However, the majority of service users recognised that this was not a feasible goal, arising from conversations about the limits of the therapy and their long-standing experience of battling with voices, which influenced their expectations:

I’d like them to go away permanently, but that’s not going to happen. I would like for them not to come to the foreground at all […] but realistically I don’t think I’m going to get that. I’m a bit of a realist. I know it’s very unlikely to happen. (Service user 9)

Theme 2: feeling empowered

Both service users and practitioners described several important benefits or changes that they observed either during or following therapy, leading to feelings of empowerment in relation to the self and voices. Most participants were hoping to further build upon these improvements.

Sub-theme 1: taking control of life

Six out of 15 (40%) service users reported that taking control of life was an important outcome, as they were dependent on family members and mental health teams. Voices took up space in people’s lives and this resulted in giving up work and experiencing difficulties dealing with simple tasks. Participants reported that the therapy equipped them with skills to live alongside their voices. This helped them feel more independent, in charge of their lives and gave them reassurance in their ability to handle tasks and manage their mental health:

In the last few years I’ve had absolutely no control over my own life. So having those little things that I can control to a degree kinda gives me a bit of reassurance in myself. I mean, they are not massive things; other people think that’s pathetic, but to me that’s the only thing I got and if I can maintain that and keep practising, I might be able to get a handle on stuff like my anxiety, my panic attacks, my self-harming, my day-to-day routine, making sure my bills are paid, my impulsive behaviour, you know, this is a little stepping stone. (Service user 1)

Similarly, practitioners in all focus groups (100%) reported that a sense of agency is pertinent, as the intervention teaches people skills to gain some autonomy and independence, and leads to a newfound understanding that they can create choices for themselves:

I like the idea of agency because they think if you have to go and take medication, if a pharmacological intervention is the only one that’s been offered, that’s all you’ve ever had. So, you always got to go and take the pills, go to the doctor to get the prescription, go to the pharmacy – you have to keep doing this sort of treadmill, whereas this gives people a set of skills where they don’t need to rely on anyone else. (Practitioner 2, focus group 1)

Sub-theme 2: taking control of responses to voices

A greater sense of control and choice over how to respond to the voices, through assertively challenging and negotiating with the voices and thus weakening their potency, was identified as a key outcome by 12 (80%) service users. This had a positive impact on day-to-day life, as service users felt less threatened and distressed by their voices, felt more in control and found it easier to stand their ground:

I’m starting to get my life back for me, rather than sit here and feeling scared that something’s going to happen. I’m actually going out into the world and experiencing things. They are still there; they still sort of say “cross the road now” because there is a lorry coming but I can control that now, while before they were so loud that I didn’t have a chance and I didn’t know how to control them and that’s what I learned. (Service user 15)

This was also highlighted as an important outcome by practitioners in all focus groups (100%), noting that CBT teaches people to challenge their beliefs and thus develop a sense of control over their voices and a sense of choice over their actions:

I suppose thinking with CBT how it’s helping people see that there’s a step between hearing the voice and having to do what they say, or assuming its accuracy, and seeing that there is a step in-between there, that the person can have some agency and some choice over, and some of the successes I’ve experienced is where people see there is sort of a middle step and that’s given them more of a sense of control and choice […] they don’t have a choice until they have more information and awareness about another way of looking at things which is more CBT, so for me, yeah, enabling people to see that they do have a choice over their actions and how they respond to voices. (Practitioner 10, focus group 3)

Sub-theme 3: being confident and respecting oneself

Ten (67%) service users recognised self-confidence and self-respect as being crucial for managing their voices and their mental health. Many suggested that an increase in self-confidence led to them challenging the accuracy and power of the voices and through that they felt less distressed by their voices, enabling them to enjoy and experience life more:

When it all first started, all I used to do was really sit in front of the telly all day. I never thought I could do anything again. I was really down on myself […] The voices love telling me that nobody likes me, but when you meet all these lovely people that I’ve met you realise that actually that’s not true […] So yeah, I’ll say the confidence was the biggest thing. (Service user 11)

While for others, a sense of worth and value was important to manage their mental health, as they tried to be less critical and more accepting of themselves:

Service user: Probably the kindness, like being kinder to myself.

Interviewer: Why would you say that’s important?

Service user: I think because I’ve been so negative to myself for so long, it would help my own mental health and the voices if I could be kinder, letting myself make mistakes (Service user 7)

This was also corroborated by practitioners in all focus groups (100%). Individuals could identify their strengths through evidence building, and thus feel more confident and more capable to cope with their voices, and subsequently be more open to other experiences:

Confidence in their ability to cope with their voices and therefore cope with other things. I think the best case you see is someone who’s been battled, who’s been struggling and battling with their voices on their own, feeling alone, feeling isolated, to someone who’s feeling stronger and more capable and therefore more open to other experiences because they’re feeling more confident. (Practitioner 8, focus group 2)

Sub-theme 4: demystifying and accepting voices

Eleven out of 15 (73%) service users reflected on the value of understanding and accepting their voices. This led to them feeling less stigmatised and helped them to improve their relationships with other people and their voices by being more honest and open about their voices. While for some service users, developing a better understanding of their experiences was a key priority to challenge the power of the voices:

I think the only one is to make them less apparent – sort of demystify them and try and be more rational about it. I think that’s what I’m trying to do in relation to the voices, of what the voices are saying. (Service user 2)

Accepting the voices through having open conversations was an important achievement for most. Being honest with people helped to break down the stigma of hearing voices and feelings of shame, which was key to their recovery journey:

I suppose I’ve accepted – which I think is very important – I’ve accepted the fact that I do hear those voices and not sort of like I used to, pretend that it wasn’t happening, because I was too scared to admit to it, so I think I’ve learned to be open and honest as much as I can be to people, and they seem to understand. It is that stigma thing again. (Service user 11)

This disclosure was perceived as fundamental by practitioners in all focus groups (100%), giving voice hearers the space to understand more about their relationships with their voices and other people. The disclosure could be facilitated by decreased service users’ perceived stigma and a sense of safety in voice hearers to open up about their experiences:

Breaking the secrecy is a big thing, isn’t it? (Practitioner 8)

Huge. (Practitioner 7)

That’s a really big one. The number of people I had saying “I can actually talk to people a little bit about it”, not necessarily full disclosure but more like “I can share some things and it’s ok”. (Practitioner 6, focus group 2)

Theme 3: understanding and managing my emotions

Understanding and managing emotions was a prominent outcome for 13 service users (87%), but not for practitioners. Some service users described significant improvements during or following therapy in stress and worry, resulting from being able to cope better with their voices and feeling more in control:

I love the fact that I don’t feel quite so stressed. It’s a lot easier to even deal with the stresses in life that you have to deal with. (Service user 4)

However, others identified mood as a desired outcome, as they either observed no improvement in their mood or this was only somewhat targeted in therapy:

I want to be able to not be depressed no more. (Service user 10)

Service users recognised that their mood was having an impact on how they were responding to their voices, as they appeased or complied with them:

When I’m low I kind of give into them, when I am high, I fight back. (Service user 8)

They also suggested that their mood could influence the intensity and content of their voices, as exacerbated distress can trigger voice activity:

I think the mood [is important], because it’s the mood that has an impact on everything else. There is a definite correlation between the intensity and the theme of the voices when my mood is higher and when is lower. When is lower they are generally more derogatory, angry and intense. When I’m high they are generally nicer, more pleasant, encouraging me to do good things. (Service user 6)

Theme 4: doing more of what I want to do

Service users and practitioners discussed the importance of personal outcomes. Whilst practitioners valued overall personal outcomes, service users reflected in particular on what was important to them.

Sub-theme 1: lead the life they want to lead

Practitioners in all focus groups (100%) discussed the value of personal outcomes, highlighting the importance of focussing on what matters to the individual, allowing them to lead the life they want to lead:

They are really individual. It’s what’s important to that person […] There’s stuff that I might think is important. I’m not important and society is not important. It’s what’s important to them. (Practitioner 2, Focus group 1)

Practitioners also emphasised that outcomes should be aligned with individual goals, whilst recognising that these may not necessarily be related to the voices and that the voice hearing experience may be a barrier to achieving those goals:

Interviewer: We’ve talked about a couple of different changes. Which ones would you consider the most important?

Practitioner: I think the ones that align with the individual goals […] We can have all the hopes and dreams for these people and everything, but it might not be what they would have wanted from this situation. (Practitioner 5, Focus group 2)

Sub-theme 2: (re)connecting with family and friends

Five service users (33%) described changes in their relationships with family and friends. Feeling safer and less affected by them allowed people to focus on their relationships with others:

My mate, we can sit and talk for hours now, whereas before I struggled to speak to him for half an hour and last time I went over there, which was last week, I was there for seven and a half hours almost, so the quality in general is a lot better ‘cause I’ve actually got things I want to talk about now and I’m not listening to the voices sort of telling me how “he doesn’t really want to be here, you get out there as soon as you can”. (Service user 9)

Some were interested in spending more time with them:

I’d like to make some more friends, but I’d just like to spend more time with the friends I’ve got as well because they have been really good to me through everything. (Service user 9)

Whilst others observed improvements in the quality of time spent with family and friends:

Even just interacting with my son and playing with him and things, that wouldn’t happen because sometimes it would be so bad. So now, that rarely happens that it’s like that, rather than being all of the time, I feel that I can actually, you know, have like a nice game together and things like that, so it has had a really positive impact. (Service user 4)

Sub-theme 3: pursuing work and education

Five service users (33%) found it important to prioritise work and education, which gave them a sense of purpose, as the voices became less intrusive and people felt more confident. This allowed them to have more space in their lives and motivation to pursue a vocation that they are passionate about. Some wanted to complete their education:

Service user: I think the most important benefit would be to start living again really.

Interviewer: Can you tell me more about that?

Service user: Because of the voices I put my life on hold for a long time, so it was like restarting the machine and the therapy was the start of that […] The first thing would be to finish university; that would be a big thing for me and then to be able to go out more and hopefully get a job. (Service User 7)

Whilst others felt that the intervention gave them the confidence to consider returning to work:

Service user: I got an interview lined up for real work, paid work with a company that is working with the job centre, so I think it’s in 2 weeks’ time, so me maybe going back to paid work parttime.

Interviewer: Is this something that you would like to do?

Service user: Yeah, definitely, because I worked all my life, so to suddenly go from working 40 hours a week to nothing, even though you are not well, doesn’t do your confidence any good.

Interviewer: Would you say the therapy helped you with that?

Service user: Yeah, it’s given me confidence to say “ok I’ll meet with you, and we’ll have a chat”. (Service user 15)

Sub-theme 4: engaging in meaningful activities and hobbies

Six service users (40%) found it important to start engaging in meaningful activities that the voices stopped them from partaking in as they felt distracted or unsafe to go outside:

I’m hoping to start writing again ‘cause I used to write a lot [...] My mind is a lot clearer, and I can focus on things I need to focus on to write things. (Service User 9)

One service user wanted to join clubs:

I’d like to go out and join more clubs. (Participant Service user 12)

Whereas another service user wanted to prioritise their physical health:

I’d just like to be healthier […] I just want to be able to feel that I’ve had a nice workout in the gym because I know that if you go to the gym and have a good work out it helps a lot of people. (Service user 10)

Discussion

Main findings

Four themes were identified, two of which encompassed sub-themes. Service users and practitioners valued a sense of agency, control over the voices, self-confidence, self-respect, acceptance of voices and personal outcomes. This is in line with the cognitive model of voices postulating that self- and voice-related appraisals are key factors in determining emotional and behavioural responses to voices (Birchwood et al., 2000, 2004; Chadwick & Birchwood, 1994). Our findings also share a significant overlap with previous research demonstrating that perceived power, self-concept, independence, empowerment and understanding and acceptance are important outcomes of CBT and aspects of recovery (e.g. Byrne et al., 2010; Greenwood et al., 2010; Hall et al., 2022; Hayward et al., 2018; Wood & Alsawy, 2018; Wood et al., 2015). These variables have also been targeted in psychological therapies for distressing voices including perceived power, acceptance and self-esteem (Thomas et al., 2014).

Most outcomes that are prioritised by both voice hearers and practitioners (e.g. self-confidence, acceptance of voices and voice controllability) have been previously measured in trials of psychological interventions for distressing voices (Lincoln & Peters, 2019; Loizou et al., 2022; Thomas et al., 2014). However, sense of agency has not been explicitly measured, although it may be captured to an extent using the Process of Recovery Questionnaire (Neil et al., 2009). Additionally, some measures used in trials may lack good psychometric properties or single items may be less sensitive to change, which can hinder outcome evaluation. For example, voice controllability has mostly been measured using the “controllability” item of the Psychotic Symptom Rating Scale for Auditory Hallucinations (Haddock et al., 1999), which may lack sensitivity to change (Thomas et al., 2014) and the Rosenberg self-esteem scale (Rosenberg, 1965) is considered outdated and was found to have poor psychometric properties (Keith & Bracken, 1996).

Despite similarities in service users’ and practitioners’ priorities, there were some clear discrepancies in the outcomes prioritised by service users and practitioners. Whilst service users sought changes in voice characteristics, practitioners did not highlight this as a key outcome, which could be explained by the fact that these interventions aim to reduce associated distress, rather than change the frequency or content of the voices. Crucially, many service users understood that this was not a feasible outcome, which could be attributed to the setting of reasonable expectations at the start of the therapy. Furthermore, while most service users were concerned with understanding and managing their emotions, this was not identified as a key priority by practitioners, which may be because this was not targeted in therapy. Indeed, few symptom-specific interventions aimed to improve negative emotions in voice hearers such as depression (van der Gaag et al., 2012), suggesting that this is not considered a priority by practitioners. However, our findings, along with previous quantitative research, suggest that negative emotions such as anxiety and depression can play a detrimental role in voice hearing (Hartley et al., 2013). Their key role has also been emphasised in the cognitive model of psychosis, suggesting that negative self-related appraisals (e.g. low self-esteem) can lead to negative emotions, which can maintain distressing voices (Garety et al., 2001).

Practitioners’ foremost priority appeared to reflect personal outcomes, with service users identifying several areas of importance including relationships, occupation and meaningful activities. Whilst several instruments of functioning, quality of life and goal attainment have been employed to measure either performance in those areas or achievement of goals (e.g. Social Functioning Scale, Birchwood et al., 1990; CHOICE, Greenwood et al., 2010; EQ-5D-5L, Herdman et al., 2011), these either do not capture how important these outcomes are to the individual, indicate how these outcomes change over time (e.g. is it the quality or amount of time spent that has changed?) or account for changes in goal priorities over time.

Strengths and limitations

The current study included service users’ and practitioners’ views about valued outcomes. Practitioners with varying levels of experience were included in the focus groups and service users met diagnostic criteria for a range of clinical disorders, which ensured that there was diversity within the sample. Most service users were White British and female, overlooking views of voice hearers from black and ethnic minority backgrounds and other gender identities. It should also be acknowledged that most service users received guided self-help CBT for distressing voices, which aims to target self- and voice- related appraisals. Hence, this may have influenced our findings, as other interventions may have a different therapeutic target (e.g. trauma-focused CBT). Additionally, service users and practitioners were recruited from a single NHS Trust, thus priorities may not be reflective of the wider population.

Although service users were included if they had completed at least 50% of a course of CBT for voices, most service users appeared to have either completed or engaged well with the therapy. Therefore, the priorities of those that completed less than 50% of a course may be different. Indeed, the reasons for not completing the course could include dissatisfaction with the goals that were being worked towards. A final limitation concerns the analysis and interpretation of the data by professionals, some of whom are part of a community that advocates for the benefits of CBT for distressing voices. However, all three service users, who provided feedback on the findings, reported that the themes reflected their experience.

Implications and future directions

The current qualitative study identified several outcomes (i.e. changes resulting from the intervention) that are important to service users and practitioners. Some of these could also be perceived as processes that could be targeted in CBT (e.g. perceived power). Indeed, outcomes and processes are interconnected, as changes in processes can generate changes in other outcomes (Thomas, 2015). A number of CBT targets (e.g. control, self-confidence, acceptance) (Thomas et al., 2014), were valued by both service users and practitioners. Such emphasis suggests that they may be of high importance compared to other treatment targets. Additionally, understanding and accepting voices appeared to be a priority for both service users and practitioners, suggesting that processes such as the use of normalisation and formulation is important in CBT and could reduce service users’ perceived stigma, allowing people to share and be more accepting of their experiences. The findings also highlight some discrepancies between service users’ and practitioners’ priorities for treatment. Whilst a balance should be struck between service users’ and practitioners’ priorities and the aims and targets of CBT, it is evident that service user perspectives have received less attention. Hence, there is a need to allow for greater consideration of service users’ priorities, which can lead to improved outcomes and increase engagement with the therapy (e.g. Kilbride et al., 2013). In this respect, greater focus should be placed on understanding and managing emotions, as they appear to directly influence voices.

Future research should further explore the relationship between negative emotions and voices and seek to target negative emotions prior to the commencement of CBT for distressing voices, as they appear to persist following treatment, and this should be co-produced with people with lived experience. There is also a need to identify measures of personal outcomes, that can provide a more comprehensive understanding of people’s experiences. Qualitative methods can also provide an avenue to capturing and understanding nuances and changes over time.

Discussion

Service users and practitioners valued several changes relating to issues of empowerment and personal outcomes. Despite many similarities in service users’ and practitioners’ priorities, there were some notable differences. Service users valued changes in voice characteristics and managing their emotions, whereas these were not highlighted as important outcomes by practitioners. Recommendations include developing measures in collaboration with service users and a placing greater focus on understanding and managing emotions

Ethical statement

The study was sponsored by the University of Sussex and has received ethical approval by the London-Stanmore Research Ethics Committee and the Health Research Authority (21/LO/0257).

Acknowledgements

We would like to thank all study participants. We would also like to thank SPFT and SeNSS for the doctoral training grant.

Disclosure statement

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

Data availability statement

Data are available from the corresponding author by email upon reasonable request.

Additional information

Funding

This was supported by the South east Network for Social Sciences (SeNSS) Supervisor-led Collaborative Studentship sustained by the ESRC and SPFT (grant number: ES/P00072X/1). SeNSS has no role in in the design, analysis, write-up of the manuscript or the decision to submit for publication.