Journal of Psychiatric Research 174 (2024) 289–296

Contents lists available at ScienceDirect

Journal of Psychiatric Research

journal homepage: www.elsevier.com/locate/jpsychires

Effects of a novel, brief psychological therapy (Managing Unusual Sensory

Experiences) for hallucinations in first episode psychosis (MUSE FEP):
Findings from an exploratory randomised controlled trial
Robert Dudley a, b, *, Guy Dodgson a, Stephanie Common c, Emmanuel Ogundimu d, James Liley d,
Lucy O’Grady a, Florence Watson c, Christopher Gibbs a, Bronia Arnott e, Charles Fernyhough d,
Ben Alderson-Day d, Charlotte Aynsworth a
a
Cumbria, Northumberland, Tyne and Wear NHS Foundation Trust, St. Nicholas Hospital, Jubilee Road, Gosforth, Newcastle Upon Tyne, NE3 3XT, United Kingdom
b
Department of Psychology, University of York, YO10 5DD, United Kingdom
c
Tees, Esk & Wear Valley NHS Trust, Wessex House, Falcon Court, Stockton on Tees, TS18 3TX, United Kingdom
d
University of Durham, Stockton Road, Durham, DH1 3LE, United Kingdom
e
Newcastle University, Population Health Sciences Institute, Baddiley-Clark, NE2 4AX, Newcastle Upon Tyne, United Kingdom

A R T I C L E I N F O A B S T R A C T

Keywords: Hallucinations are a common feature of psychosis, yet access to effective psychological treatment is limited. The
Psychosis Managing Unusual Sensory Experiences for First-Episode-Psychosis (MUSE-FEP) trial aimed to establish the
Hallucinations: Treatment outcome research feasibility and acceptability of a brief, hallucination-specific, digitally provided treatment, delivered by a non-
Digital
specialist workforce for people with psychosis. MUSE uses psychoeducation about the causal mechanisms of
hallucinations and tailored interventions to help a person understand and manage their experiences. We un­
dertook a two-site, single-blind (rater) Randomised Controlled Trial and recruited 82 participants who were
allocated 1:1 to MUSE and treatment as usual (TAU) (n = 40) or TAU alone (n = 42). Participants completed
assessments before and after treatment (2 months), and at follow up (3–4 months). Information on recruitment
rates, adherence, and completion of outcome assessments was collected. Analyses focussed on feasibility out­
comes and initial estimates of intervention effects to inform a future trial. The trial is registered with the ISRCTN
registry 16793301. Criteria for the feasibility of trial methodology and intervention delivery were met. The trial
exceeded the recruitment target, had high retention rates (87.8%) at end of treatment, and at follow up (86.6%),
with good acceptability of treatment. There were 3 serious adverse events in the therapy group, and 5 in the TAU
group. Improvements were evident in both groups at the end of treatment and follow up, with a particular benefit
in perceived recovery in the MUSE group. We showed it was feasible to increase access to psychological inter­
vention but a definitive trial requires further changes to the trial design or treatment.

1. Introduction therapists, the time taken to train them, and CBTp usually entailing 6 or
more months of weekly appointments (Morrison, 2017)
Hallucinations are a common feature of Psychosis, with many Owing to these rate limiting factors a number of approaches have
reporting hearing (Australian Schizophrenia Research Bank, 2017) or been used to increase access. One solution is to use briefer treatments for
seeing things others do not (Dudley et al., 2023a,b). For some, these psychosis (Hazell et al., 2018). Another is to broaden the provision by
experiences can be extremely distressing, and can contribute to higher training non-specialist staff to deliver therapy as part of routine clinical
rates of admission and relapse (Waters et al., 2018). Cognitive Behav­ practice (Garety et al., 2018; Turkington et al., 2002). However, CBT for
ioural Therapy for Psychosis (CBTp) has been recommended for some psychosis encompasses a wide range of treatment targets (Morrison,
time (National Institute of Clinical Excellence NICE, 2002) but access to 2017) which can cause challenges in training and delivery of a complex
psychological therapy remains limited. This is owing to a lack of trained intervention by a less trained workforce. One method to overcome this

* Corresponding author. Department of Psychology, University of York, York, YO10 5DD, United Kingdom.
E-mail address: [email protected] (R. Dudley).

https://doi.org/10.1016/j.jpsychires.2024.04.031
Received 26 November 2023; Received in revised form 3 April 2024; Accepted 15 April 2024
Available online 16 April 2024
0022-3956/© 2024 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
R. Dudley et al. Journal of Psychiatric Research 174 (2024) 289–296

issue of complexity is to focus on targeting specific symptoms such as 2. Material and methods
auditory hallucinations (Hayward, 2018; Hazell et al., 2018) or a limited
number of key causal mechanisms (Foster et al., 2010; Myers et al., 2.1. Design
2011) with the aim of achieving meaningful results more efficiently. An
additional strategy is to reduce complexity by standardising the treat­ This was a single blind, pragmatic randomised controlled trial
ment using digital technologies; an area where there has been a rapid (ISRCTN registry: 16793301; registered December 7, 2021) comparing
growth in digital innovations to augment mental health care, as rec­ MUSE plus treatment as usual (TAU) with TAU alone. Assessment
ommended in the NHS Long-Term Plan (Hollis et al., 2018; NHS Long occurred pre-randomisation, two months post-randomisation (post-
term plan, 2019). Accordingly, treatments for psychosis are increasingly treatment) and three-four months post-randomisation (follow-up).
delivered using novel digital approaches (Freeman et al., 2022; Garety Full consideration of sample size requirements, randomisation,
et al., 2021; Craig et al., 2018; Yiend et al., 2022) which provide blinding, measures, data monitoring and assessment of safety were
treatment in an accessible and engaging way. described in the trial protocol published before completion of data
This present study combines these approaches by offering a brief, collection (Dudley et al., 2022). No significant changes were made to the
targeted therapy focussed solely on helping hallucinations, using a methods after trial commencement. Qualitative interviews captured
widely available workforce with a treatment that is delivered using a service-users’ experience of therapy and clinicians’ experiences of the
digital platform. The approach used, called Managing Unusual Sensory training and supervision in MUSE. Clinicians were asked about factors
Experiences (MUSE), focusses on causal mechanisms that lead people to affecting uptake, adherence, and facilitators/barriers to implementa­
hear and see things others do not. It uses engaging technology such as tion. Thematic analysis of the qualitative interviews will assess the
videos, and animations over 4 to 6 sessions to explain how the mind acceptability of the training, intervention, and trial procedures (in
works to make sense of the world around; how processes like inner preparation) further informing any future definitive study.
speech (Fernyhough, 2004); vigilance (Dodgson and Gordon, 2009;
Dudley et al., 2014); and the consequences of trauma (Dudley et al., 2.2. Participants
2023a; Stevens et al., 2019) can lead to hallucinatory experiences. These
explanations link to key interventions and coping strategies that target Eligible participants were recruited from two UK mental health
these causal processes, such as interrupting the phonological loop as a Trusts (CNTW and TEWV NHS Trusts). The participants were; aged 16
way of affecting inner speech. This process draws on psychoeducation to years and over; met ICD-11 criteria for schizophrenia, schizoaffective
help a person change their understanding of their experiences as well as disorder or entry criteria for an EIP service; current hallucinations for at
to manage their experiences better and cope more effectively. A full least four weeks; considered these an important issue to work on; had a
description of the rationale for this approach, as well an illustration of care coordinator; were able to provide written, informed consent; and
the modules and examples of the interventions offered is provided in the judged to be clinically stable at the time of the assessments and for the
trial protocol (Dudley et al., 2022). past month (for example had no medication changes or reported in­
Two uncontrolled studies with people with Psychosis (MUSE PSY­ crease in self-harm ideation or incidents within the past month).
CHOSIS; Dodgson et al., 2021a) and those at risk of transition to Psy­ Exclusion criteria were: known organic illness; primary diagnosis of
chosis (MUSE ARMS; Dodgson et al., 2021b) provided preliminary substance misuse non-English speaking; currently (or in the past 6
evidence of acceptability and impact when treatment was offered by months) engaged in CBTp.
trained psychological therapists, particularly for those service users in
the earlier stages of their psychosis. However, there is a clear need to 2.3. Randomisation and blinding
refine such results before MUSE can be considered a viable treatment
option. As such, the aim of the current study was to inform the design of Participants were randomly allocated in a 1:1 ratio to receive either
a definitive clinical and cost-effectiveness trial by evaluating the feasi­ MUSE and TAU, or TAU alone. Stratified block randomisation was
bility, safety and acceptability of MUSE therapy compared to treatment completed using an online randomisation service (Sealedenvelope.com).
as usual (TAU) amongst people with hallucinations in the early stages of It was stratified by site and employed randomised-permuted blocks of
psychosis. Moreover, given the need to increase access, the aim was to 4–6. All post-randomisation assessments were completed by research
train a more widely available frontline workforce (rather than psycho­ assistants who were blind to participant allocation. There were 5 full
logical therapists), to deliver MUSE in routine community services. In blind breaks with 2 in TAU and 3 in MUSE. When breaks in blinding
the UK, this workforce typically comprises of psychiatric nurses, occu­ were reported, assessments were completed by another research assis­
pational therapists, and social workers who provide care coordination tant ensuring the assessor remained blind to allocation.
which is a clinical case and crisis management function. Past attempts
have noted the challenges in training staff to deliver psychological 2.4. Procedure
treatments in busy clinical services (Garety et al., 2018) but Early
Intervention in Psychosis (EIP) teams have lower case-loads encour­ Eligible participants were identified by members of the clinical team
aging greater opportunity for provision of evidence-based treatments and asked for verbal permission to be contacted by a research team
(Bird et al., 2010; Brabban and Dodgson, 2010). Also, as MUSE is highly member. Research assistants provided information sheets, obtained
standardised it both reduces the complexity of delivery, and increases written informed consent, and administered baseline measures.
consistency of delivery. Recruitment began in June 2021 and was completed in May 2022.
The present study is a feasibility randomised controlled trial of the Follow-up assessments ran from August 2021 to September 2022.
MUSE intervention. The primary aim was to establish the clinical Participants were allocated to receive either treatment as usual
feasibility of the intervention, in terms of participant recruitment, up­ (TAU), or up to 8 1-h sessions of MUSE plus TAU over a two-month (8
take, safety and satisfaction with treatment, retention in the trial, and week) period. In the UK, TAU for service-users with psychosis is based
uptake of training and delivery of therapy by the non-specialist work­ on the principles of the Care Programme Approach and comprises a
force. An additional aim was to estimate the parameters to calculate the range of interventions, including psychiatric medication, care coordi­
effect size on key outcomes to inform a future definitive trial. nation, social or vocational support, family interventions, outpatient
follow-up care, and access to CBTp. The MUSE intervention was deliv­
ered by care coordinators who acted as research therapists having
completed three days of training on MUSE prior to taking on cases.
Therapy was conducted in participants’ homes or at clinical bases. The

290
R. Dudley et al. Journal of Psychiatric Research 174 (2024) 289–296

initial sessions generally focused on engagement, psycho-education indicates that samples of 35 per arm or more give a reliable estimate of
about how the mind works and on normalising the experience of hal­ the standard deviation of the outcome measure (Teare et al., 2014). An
lucinations. Subsequent sessions focussed on identification of key causal estimated attrition of 12.5% was based on past research of psychological
processes and the use of coping strategies. Optional sessions were therapy with people with psychosis (Morrison et al., 2018) and similar
available for understanding and managing visions and sleep. Adherence brief interventions (Freeman et al., 2022) meaning 70 people were ex­
checklists were utilised to maximize fidelity, with any protocol di­ pected to complete the study.
vergences monitored during therapist supervision. Supervision sessions When considering outcomes, the focus was placed on descriptive
occurred fortnightly using a group supervision format and were facili­ statistics, point estimates, and associated 95% confidence intervals
tated by psychological therapists with clinical experience of MUSE. rather than tests of statistical significance. Effect sizes are reported based
on Cohen’s d (Cohen’s d = mean difference/(pooled standard devia­
2.5. Outcomes tion)). Descriptive baseline and follow-up data were summarised as
mean (sd) for continuous variables and frequencies/percentages for
The primary purpose of the study was to consider the feasibility of a categorical variables. Analyses followed a pre-specified plan approved
future definitive trial. Consequently, there was a focus on recruitment by the chief investigator, the trial statistician, and the trial steering
and retention rates, adherence to allocation, trial and treatment committee (available to view online at ISRCTN registry) and was based
acceptability (assessed through discontinuation rates and a qualitative on intention-to-treat principles at the participant level. A linear mixed
study). model was used to estimate the effect of MUSE on the outcome at end of
Progression criteria based on ADEPT guidelines (Bugge et al., 2013) treatment and follow up while controlling for study site. The analysis
were developed based on discussion with the Patient Public Involvement was repeated adjusting for baseline variables (age, sex, number of hal­
group (PPI), trial steering and management groups in advance of the lucinations, duration of hallucinations, length of time engaged in the
final data collection. The progression criteria were divided into three service, site, and PSYRATS delusions score). All available data was used
categories (green, amber and red). from each timepoint, with missing data imputed with pro-rating. The
main analyses were all conducted in R (version 4.1.2) using the ‘lme4’
2.5.1. Measures package (version 1.1–30).
Auditory hallucinations were assessed using the Psychotic Symptom In line with the recent CONSORT - Social and Psychological In­
Rating Scales (PSYRATS (Haddock et al., 1999)) which is an 11 item terventions (CONSORT-SPI; Grant et al., 2018) guidance, which rec­
semi-structured interview assessing frequency, duration, loudness, ommends minimising the distinction between primary and secondary
distress intensity and control of hallucinations. The delusions subscale outcomes for psychological therapies trials, all outcomes are reported at
was also completed. In addition, the self-report voice-impact subscale on all assessment time points.
the Hamilton Program for Schizophrenia Voices Questionnaire (HPSVQ;
Van Lieshout and Goldberg, 2007) was used. Additional items asking 2.8. Ethical considerations
about hallucinations in non-auditory modalities (visual, somatic, olfac­
tory and tactile) were assessed as well (Dudley et al., 2023b). The investigation was carried out in accordance with the Declaration
Levels of anxiety and depression (Depression, Anxiety and Stress of Helsinki. Ethical approval was provided by the NHS Yorkshire and the
Scales (DASS) Lovibond and Lovibond, 1995), as well as perceived re­ Humber-Sheffield Research Ethics Committee (21/YH/0090), Health
covery (QPR process of recovery questionnaire; Neil et al., 2009) were Research Authority (HRA/HCRW) approval (IRAS 292150). Informed
assessed. The perceived impact of the intervention (The CHoice of consent of the participants was obtained after the nature of the pro­
Outcome In Cbt for psychosEs (CHOICE) Greenwood et al., 2010) was cedures had been fully explained. The trial was funded by National
used to assess progress towards therapy-related goals. To determine Institute of Health Research (NIHR201078).
therapy acceptability and alliance, the Satisfaction with Therapy and
Therapist Scale (Oei and Green, 2008) and Working Alliance inventory 2.9. Patient and public involvement
were used (Horvath, 1986). In addition, at each session, a short
self-assessment form comprising items adapted from the main measure As detailed in the trial protocol (Dudley et al., 2022) people with
of hallucinations monitored variations in voice frequency and distress. lived experience of hallucinations were involved all aspects of the MUSE
Therapists completed a therapy adherence checklist each session treatment development as well as in the trial set up, delivery of the study
reporting on what modules of the MUSE package had been used. To help and in the interpretation and dissemination its findings. We held
establish the feasibility of collecting information on health economics, monthly PPI meetings and had PPI representation on the Trial steering
self-report measures of service use and quality of life (Short Form-36; committee.
Ware and Sherbourne, 1992), EQ-5D (EuroQol Research Foundation,
2019), perceived capability (Investigating Choice Experiments Capa­ 3. Results
bility Measure for Adults (ICECAP-A; Flynn et al., 2015) were collected
as well as case record review using a tool developed for the study. 3.1. Feasibility considerations

2.6. Adverse events 3.1.1. Recruitment and retention

132 individuals were referred to the trial, with 94 people consented
Serious adverse events (SAEs) and adverse events (AEs) were and 82 recruited and randomised with 40 allocated to MUSE and 42 to
recorded via participant self-report to therapists and/or research assis­ TAU. 11 participants of the 132 referred declined to participate from
tants during the trial. Screening of electronic medical records was also referral to baseline assessment. Others were excluded owing to not
conducted at follow-up. All SAEs were reported to the Chair of the Trial meeting entry criteria or not being contactable. Recruitment into the
Steering Committee and trial Sponsor for independent monitoring. trial exceeded the target (green zone). In terms of retention, 72 (87.8%
green) completed end of treatment and 71 completed follow up assess­
2.7. Statistical analysis ments (86.6% amber). Of the treatment group three withdrew owing to
experiencing bereavements and one was lost to follow up. (Fig. 1)
A target sample of 80 participants was deemed sufficient to both Baseline characteristics are reported in Table 1. The groups were
demonstrate feasibility and to obtain reliable parameter estimates for similar in age, sex, and ethnicity. Participants reported high levels of
sample size in a definitive trial. Guidance on external pilot studies distress and negative consequences from their hallucinations

291
R. Dudley et al. Journal of Psychiatric Research 174 (2024) 289–296

Fig. 1. CONSORT diagram for flow of participants.

comparable to that seen in other psychological therapy trials for hallu­ 3.1.2. Staff training
cinations (Craig et al., 2018). The MUSE group reported somewhat 25 care coordinators from 15 EIP teams attended a three-day training
higher levels of unemployment, and psychiatric admission but lower course on MUSE. Most had little or no previous therapy experience. Of
levels of current antipsychotic medication use. The MUSE group re­ the 25 trained, 16 went on to work with one or more supervised therapy
ported hearing voice for longer as well. cases. The training was run on three occasions owing to attrition of the

292
R. Dudley et al. Journal of Psychiatric Research 174 (2024) 289–296

Table 1 3.1.5. Adverse events

Demographic characteristics of participants at baseline. There were 14 adverse events documented across the study period
Total sample (N TAU n = 42 MUSE n = 40 reported by 11 individuals randomised into the trial. 8 were classed as
= 82) serious adverse events and there were 5 in the TAU group and 3 in the
Mean SD or N Mean SD or N Mean SD or N MUSE + TAU group, and all were assessed as unrelated to the inter­
(%) (%) (%) vention by the independent TSC Chair. In addition, a person sadly died
Age 30.4 (10.34) 29.21 (10.27) 31.65 (10.4)
after they had consented, but before any assessments were undertaken.
Sex Details are provided in Table 2.
Female 38 (46%) 20 (48%) 18 (45%)
Male 44 (54%) 22 (52%) 22 (55%)
3.2. Outcomes by trial arm and assessment time point
Ethnicity
White 78 (95.5%) 39 (93%) 39 (97.5%)
Asian 2 (2.5%) 1 (2%) 1 (2.5%) Outcome at each time point is shown in Table 3 as well as the
Black 1 (1%) 1 (2%) intention to treat analysis. There was considerable improvement in both
Mixed 1 (1%) 1 (2%) groups over a relatively short period of three months (within subject
Employment status
Unemployed 41 (51%) 16 (39%) 25 (64%)
effect sizes of Cohen’s d = 0.8–0.9 on PSYRATS AH at 12 weeks for
Working full or part time 23 (29%) 12 (29%) 11 (28%) example). The reduction in scores on the PSYRATS in both arms at 12
Student 11 (14%) 9 (22%) 2 (5%) weeks was greater than that reported in successful treatment studies for
Health related benefits 5 (6%) 4 (10%) 1 (3%) distressing hallucinations (Craig et al., 2018) delivered over similar
Highest educational level
timescales. The same pattern was evident on all the outcome measures
No qualification 4 (6%) 2 (5%) 2 (6%)
GCSE 33 (47%) 16 (43%) 17 (52%) meaning that at the end of treatment and at follow up there were
A levels (or sixth form 26 (37%) 15 (41%) 11 (33%) negligible differences between groups (effect sizes all trivial to small as
equivalent) reported on Table 3), with a possible exception of the improvements in
Undergraduate 7 (10%) 4 (11%) 3 (9%) perceived recovery but even this was a small effect. Analyses run con­
Past Psychiatric admission
Yes 19 (24%) 6 (15%) 13 (33%)
trolling for baseline variables (age, sex, number of hallucinations,
No 61 (76%) 34 (85%) 27 (67%) duration of hallucinations, length of time engaged in the service, site,
Current antipsychotic medication and PSYRATS delusions score) made no difference to the findings.
Yes 59 (79%) 34 (87%) 25 (69%) Health economics data (EQ5D, ICECAP and SF-36) were collected to
No 16 (21%) 5 (13%) 11 (31%)
establish feasibility of data collection and are not reported here but the
Duration of time in EIP in 10.54 (9.50) 10.57 (9.99) 10.5 (9.1)
months full data is available as outlined in the data availability statement.
Duration of voice hearing in 86.23 (110.59) 70.32 (103.6) 102.1 (117.4)
months 3.3. Impact of covid
Number of hallucination 2 2 2
modalities
Midway through the trial a resurgence of Covid 19 led to 7 of the 15
teams withdrawing citing staffing pressures. The impact of Covid on trial
workforce resulting from changes in roles with a number leaving for new delivery was assessed by considering the number of sessions provided
posts or advanced training opportunities. Also, 7 of the 15 recruiting before the resurgence (mean = 5.57 sd = 0.53) in comparison to after
sites withdrew from the study following a resurgence in Covid 19 (mean = 5.3, sd = 1.24) with no significant difference found (W = 109, p
midway through the trial. Owing to this three people were treated by a = 0.53). Similarly, there was no difference in any baseline assessment
clinical psychologist rather than a care coordinator, and this protocol
deviation was recorded and reported to the sponsor and Trial Steering Table 2
Committee. Hence, the mean number of cases per MUSE therapist was Incidence of adverse events across groups.
2.35 (range 1–6). MUSE n = TAU n =
40 42
3.1.3. Treatment adherence and fidelity
Serious Adverse Event (SAE)
36 (90%) participants attended at least one session of MUSE with the Participants with an SAE 3 4
mean being 5.4 sessions (sd = 1.13, range = 0–8). Therapy was classed Number of SAEs 3 5
as ‘insufficient’ if participants had completed less than 4 sessions Types of SAE
meaning 32 (80% green) had a satisfactory dose of treatment. Session by Deatha 0 0
Hospital admission – mental health 2 1
session recording of content covered in MUSE sessions indicated good Overdose/self-injury requiring treatment in general 0 1
adherence to the protocol with participants completing an average of 4.5 hospital
of the 7 modules in the sessions available. All participants who had a Hospital admission – physical health 0 3
satisfactory “dose of treatment” (32/32; 100%) completed the module Pregnancy loss 1 0
Adverse event (AE)
on: understanding voices; 97% completed how the mind works; 87.5%
Participants with AE 1 3
the module on inner speech; 53% the trauma and memory module, and Number of AEs 1 5
66% the module on hypervigilance. The modules on visions (31%) and Types of AE
sleep (19%) were the least used. Overdose/Self-injury – no treatment sought 0 2
Supervision attendance was variable as it tended to occur when Physical injury – unrelated to mental health – no 0 1
treatment sought
working with a case. Planned audio recordings to independently assess Attended A&E – mental health – no treatment 0 1
adherence were not undertaken owing to an inability to resolve infor­ required
mation governance issues. Attended A&E – physical health – no treatment 0 0
required
Police incident 1 1
3.1.4. Participant satisfaction
The participants all reported they agreed or strongly agreed that they A&E Accident and Emergency room.
a
were satisfied with MUSE therapy and that they would recommend it to One person died after consenting to the trial, but was not contactable for
someone with a similar problem (Oei and Green, 2008). many months, and no baseline assessments were undertaken, and person was not
randomised.

293
R. Dudley et al. Journal of Psychiatric Research 174 (2024) 289–296

Table 3 digitally delivered treatment for distressing hallucinations, by a less

Baseline, end of treatment and follow up scores on outcome measures. specialist workforce. The trial recruited to target, and retained a suffi­
TAU MUSE Coefficienta 95% CI Cohen’s cient number of participants (86% at follow up). Withdrawals were
d mainly owing to suffering significant bereavements. There was good
PSYRATS AH total uptake (90%) and compliance (80% received 4 or more sessions) with
Baseline 29.81 29.3 the MUSE treatment. Satisfaction with the treatment was high. Adverse
(4.45) (5.55) and serious adverse events were similar across conditions and not
8 weeks 23.53 24.56 1.63 − 2.00–5.26 0.06 obviously attributable to treatment.
(12.89) (10.2)
12 22.39 21.91 − 0.05 − 3.71–3.59 − 0.03
The study also demonstrated it was possible, but not easy owing to
weeks (11.65) (12.05) high rates of therapist attrition, to train a wider workforce and for them
PSYRATS Distress to deliver treatment. Despite most therapists treating a small number of
Baseline 15.71 15.38 participants there was evidence of good adherence to the treatment with
(2.75) (3.44)
most participants receiving more than the minimum number of sessions,
8 weeks 12.47 12.61 0.50 − 1.65–2.65 0.02
(6.98) (5.85) and content being recorded as covering the core modules. This would
12 11.86 11.2 − 0.43 − 2.59–1.74 − 0.07 imply that standardised treatment delivered on a digital platform helps
weeks (6.67) (7.41) with adherence (Killikelly et al., 2017).
Hamilton total A future definitive trial would appear to be feasible but negligible
Baseline 23.64 23.62
(5.64) (6.04)
differences between the groups at both end of treatment and follow up
8 weeks 20.49 20.85 0.34 − 2.41–3.09 0.03 requires further consideration. The impact on perceived recovery, but
(9.43) (7.82) not on measures of hallucinations was also reported in another hallu­
12 17.94 18.12 − 0.06 − 2.81–2.69 0.01 cination focussed treatment (Longden et al., 2022). In this context of
weeks (10.06) (8.71)
general improvement any additional intervention would have to have a
PSYRATS Delusions
Baseline 13.45 14.5 very powerful effect in order to demonstrate impact. Reasons for this
(7.37) (7.02) general improvement may be owing to the participants being in the
8 weeks 11.5 10.42 − 1.65 − 4.89–1.61 − 0.09 early stage of psychosis where improvement is common (Lewis et al.,
(8.75) (8.39) 2002). We did not ask about Duration of Untreated Psychosis which has
12 9.19 8.6 (7.97) − 0.93 − 4.2–2.34 − 0.05
been associated with recovery, but instead asked about duration of voice
weeks (8.32)
DASS Stress hearing and length of time in service. These baseline variables were
Baseline 14.1 14.4 included as covariates but made no difference. Hence, it is unclear if
(4.18) (3.83) duration of hallucinations or psychosis impacted on our findings. It is
8 weeks 12.46 11.94 − 0.47 − 2.39–1.45 − 0.07
notable that EIP services have smaller case-loads help improve quality of
(6.08) (5.01)
12 11.43 11 (5.47) − 0.43 − 2.37–1.52 − 0.05 relationships developed by staff with service users which aids recovery
weeks (6.32) (Goldsmith et al., 2015; Turkington et al., 2017). It is possible that the
DASS Anxiety care coordinators trained in MUSE could have used the ideas more
Baseline 11.4 11.75 widely leading to contamination of TAU. However, a review of notes
(4.42) (4.20)
revealed no evidence of MUSE being used in TAU. In fact, the notes
8 weeks 9.83 9.91 0.04 − 1.63–1.71 0.01
(5.02) (5.29) review produced little evidence of any hallucination focussed work
12 8.571 8.5 (5.09) − 0.2 − 1.90–1.48 − 0.01 being undertaken in TAU. At most, only one or two care coordinators in
weeks (4.45) each team were trained in MUSE, meaning most staff had not had the
DASS Depression
training, reducing the risk of contamination.
Baseline 13.4 14.42
(5.75) (4.23)
The short duration of the intervention and the follow up may well
8 weeks 11.97 11.65 − 0.40 − 2.33–1.54 − 0.04 have reduced the capacity to deliver a sufficient number of treatment
(5.98) (5.67) sessions, or for any effect to be clearly revealed against the general
12 10.83 10 (5.67) − 1.27 − 3.23–0.69 − 0.10 improvement seen in both groups. If a demonstrable benefit of MUSE is
weeks (6.39)
to be established to warrant a definitive trial, then future studies could
QPR
Baseline 26.88 27.4 introduce a baseline period to the design to ensure stability of the ex­
(12.39) (9.47) periences before treatment is offered. Another strategy may be to in­
8 weeks 31 35.15 2.30 − 1.99–6.61 0.28 crease the number of sessions and extend the follow up period to enable
(11.76) (9.35)
larger differences to emerge between the groups owing to a greater dose
12 31.29 33.94 1.22 − 3.07–5.53 0.15
weeks (13.05) (11.78)
of treatment and a longer time period to consolidate the gains. Of course,
CHOICE it would also be important to consider the value of an active comparison
Baseline 71.15 71.55 group that controls for any digital placebo effect (Firth and Torous,
(19.84) (19.77) 2015).
8 weeks 65.93 59.85 − 4.65 − 12.85–3.54 − 0.18
The present study was ambitious in combining targeted, brief, digi­
(23.46) (23.54
12 64 59.77 − 4.74 − 13.10–3.61 − 0.12 tally delivered treatment by a non-specialist workforce. Future
weeks (25.25) (23.81 dismantling studies could try to consider these factors and isolate the
a active ingredients of digital interventions (in this case the modules and
The coefficient is the estimated effect on the outcome (in points) from MUSE
vs TAU, and could be interpreted as an adjusted mean difference. interventions that people found helpful) which may assist in refining the
features which yield maximum benefits, whilst improving our under­
standing of the mechanisms which drive them (Michie et al., 2017).
scores before and after the resurgence (full analysis available on
Finally, it is important to note that the group were representative of the
request).
local population (ONS, 2021) but was not very diverse. There are
different outcomes for people with Black or Asian backgrounds (Griffiths
4. Discussion
et al., 2023) involved in EIP services, and the impact of MUSE for these
groups, and its cultural relevance/acceptability is not tested here. Future
The MUSE FEP trial established the feasibility of delivering a brief,
trials may consider recruiting from teams with a greater diversity in the

294
R. Dudley et al. Journal of Psychiatric Research 174 (2024) 289–296

population. Ridge, Mike Rudd, Adam Wright, Janine Munroe, Kayleigh McManus,
Lucia Slack, Nikki Lonsdale, Sarah Hopkins, Louise Cate, Jodi
5. Conclusions McQueeny, Joanne Woolhouse, Angela Conway, Charlotte Shurben, and
Tim Jackson all who work in CNTW and TEWV NHS Trusts. Also, we
Overall, MUSE as a treatment appears to be safe and acceptable. Our thank Dr. Adetayo Kasim, Dr. Ehsan Kharatikoopaei and Professor Luke
approach of focussing on one experience (hallucinations), targeting key Vale for their input to the statistics and health economics. Dr. Kaja
causal mechanisms, providing treatment from a more available work­ Mitrenga helped with set up of the study and literature reviewing. Vic­
force, using a digital platform has promise for increasing access to toria Patton supported PPI involvement. Professor Sandra Bucci, Dr.
psychological therapies. Alison Brabban, and Dr. Valentina Short supported the DMC and TSC.

Funding statement References

This paper presents independent research funded by the NIHR under Bird, V., Premkumar, P., Kendall, T., Whittington, C., Mitchell, J., Kuipers, E., 2010.
Early intervention services, cognitive-behavioural therapy and family intervention in
its Research for Patient Benefit (RfPB) Programme (Grant Reference early psychosis: systematic review. Br. J. Psychiatry 197, 350–356.
Number: NIHR201078). MUSE was developed with support from the Brabban, A., Dodgson, G., 2010. What makes early intervention in psychosis services
Wellcome Trust (grant number WT1087201). The views expressed are effective? A case study. Early Intervention in Psychiatry 4, 319–322. https://doi.org/
10.1111/j.1751-7893.2010.00169.x.
those of the authors and not necessarily those of the NIHR, or the Bugge, C., Williams, B., Hagen, S., Logan, J., Glazener, C., Pringle, S., Sinclair, L., 2013.
Department of Health and Social Care. The study sponsor and funder A process for Decision-making after Pilot and feasibility Trials (ADePT):
have no role in study design; collection, management, analysis, and development following a feasibility study of a complex intervention for pelvic organ
prolapse. Trials 14 (1), 353.
interpretation of data; writing of the report; and the decision to submit Craig, T.K., Rus-Calafell, M., Ward, T., Leff, J.P., Huckvale, M., Howarth, E., Emsley, R.,
the report for publication. RD and CA were also supported by NIHR ARC Garety, P.A., 2018. AVATAR therapy for auditory verbal hallucinations in people
mental health research fellows awards. with psychosis: a single-blind, randomised controlled trial. Lancet Psychiatr. 5 (1),
31–40. https://doi.org/10.1016/S2215-0366(17)30427-3.
Dodgson, G., Alderson-Day, B., Smailes, D., Ryles, F., Mayer, C., Glen-Davison, J.,
CRediT authorship contribution statement Mitrenga, K., Fernyhough, C., 2021a. Tailoring cognitive behavioural therapy to
subtypes of voice-hearing using a novel tabletised manual: a feasibility study. Behav.
Cognit. Psychother. 49 (3), 287–301. https://doi.org/10.1017/
Robert Dudley: Writing – review & editing, Writing – original draft,
S1352465820000661.
Validation, Supervision, Software, Resources, Project administration, Dodgson, G., Aynsworth, C., Mitrenga, K.J., Gibbs, C., Patton, V., Fernyhough, C.,
Methodology, Investigation, Funding acquisition, Formal analysis, Data Dudley, R., Ewels, C., Leach, L., Alderson-Day, B., Common, S., 2021b. Managing
curation, Conceptualization. Guy Dodgson: Writing – review & editing, unusual sensory experiences: a feasibility trial in an at Risk Mental States for
psychosis group. Psychol. Psychother. Theor. Res. Pract. 94 (3) https://doi.org/
Validation, Supervision, Software, Resources, Project administration, 10.1111/papt.12323.
Methodology, Investigation, Funding acquisition, Data curation, Dodgson, G., Gordon, S., 2009. Avoiding false negatives: are some auditory
Conceptualization. Stephanie Common: Writing – review & editing, hallucinations an evolved design flaw? Behav. Cognit. Psychother. 37 (3), 325–334.
https://doi.org/10.1017/S1352465809005244.
Supervision, Resources, Project administration, Methodology, Investi­ Dudley, R., Dodgson, G., Common, S., O’Grady, L., Watson, F., Gibbs, C., Arnott, B.,
gation, Funding acquisition, Formal analysis, Data curation, Conceptu­ Fernyhough, C., Alderson-Day, B., Ogundimu, E., Kharatikoopaei, E., Patton, V.,
alization. Emmanuel Ogundimu: Writing – review & editing, Formal Aynsworth, C., 2022. Managing Unusual Sensory Experiences in People with First-
Episode Psychosis (MUSE FEP): a study protocol for a single-blind parallel-group
analysis, Data curation. James Liley: Writing – review & editing, randomised controlled feasibility trial. BMJ Open 12 (5). https://doi.org/10.1136/
Formal analysis, Data curation. Lucy O’Grady: Writing – review & bmjopen-2022-061827.
editing, Project administration, Data curation. Florence Watson: Dudley, R., Dodgson, G., Sarll, G., Halhead, R., Bolas, H., McCarthy-Jones, S., 2014. The
effect of arousal on auditory threat detection and the relationship to auditory
Writing – review & editing, Project administration, Data curation.
hallucinations. J. Behav. Ther. Exp. Psychiatr. 45 (3) https://doi.org/10.1016/j.
Christopher Gibbs: Writing – review & editing, Validation, Supervi­ jbtep.2014.02.002.
sion, Resources, Project administration, Methodology, Investigation, Dudley, R., Turkington, D., Coulthard, N., Pyle, M., Gumley, A., Schwannauer, M.,
Kingdon, D., Morrison, A.P., 2023a. Childhood trauma in clozapine-resistant
Funding acquisition, Formal analysis, Data curation, Conceptualization.
schizophrenia: prevalence, and relationship with symptoms. Schizophrenia Bulletin
Bronia Arnott: Writing – review & editing, Methodology, Investigation, Open. https://doi.org/10.1093/schizbullopen/sgad030/7416855.
Funding acquisition, Formal analysis, Data curation, Conceptualization. Dudley, R., Watson, F., O’Grady, L., Aynsworth, C., Dodgson, G., Common, S., Day, B.A.,
Charles Fernyhough: Writing – review & editing, Methodology, Fernyhough, C., 2023b. Prevalence and nature of multi-sensory and multi-modal
hallucinations in people with first episode psychosis. Psychiatr. Res. 319 https://doi.
Investigation, Funding acquisition, Formal analysis, Data curation, org/10.1016/j.psychres.2022.114988.
Conceptualization. Ben Alderson-Day: Writing – review & editing, EuroQol Research Foundation, 2019. EQ-5D-5L user guide. https://euroqol.org/publicati
Project administration, Methodology, Investigation, Funding acquisi­ ons/user-guides. (Accessed 21 January 2022).
Fernyhough, C., 2004. Alien voices and inner dialogue: towards a developmental account
tion, Formal analysis, Data curation, Conceptualization. Charlotte of auditory verbal hallucinations. New Ideas Psychol. 22, 49–68.
Aynsworth: Writing – review & editing, Supervision, Project adminis­ Flynn, T.N., Huynh, E., Peters, T.J., et al., 2015. Scoring the Icecap-a capability
tration, Methodology, Investigation. instrument. Estimation of a UK general population tariff. Health Econ. 24 (3),
258–269.
Foster, C., Startup, H., Potts, L., Freeman, D., 2010. A randomised controlled trial of a
Declaration of competing interest worry intervention for individuals with persistent persecutory delusions. J. Behav.
Ther. Exp. Psychiatr. 41 (1), 45–51. https://doi.org/10.1016/j.jbtep.2009.09.001.
Freeman, D., Lambe, S., Kabir, T., Petit, A., Rosebrock, L., Yu, L.-M., Dudley, R.,
RD receives payment for workshops in treating hallucinations and Chapman, K., Morrison, A., O’Regan, E., Aynsworth, C., Jones, J., Murphy, E.,
GD, SC and RD declare they are involved in running treatment studies Powling, R., Galal, U., Grabey, J., Rovira, A., Martin, J., Hollis, C., et al., 2022.
investigating psychological therapies for psychosis. All other authors Automated virtual reality therapy to treat agoraphobic avoidance and distress in
patients with psychosis (gameChange): a multicentre, parallel-group, single-blind,
declare no competing interest. MUSE is not licensed and will be made
randomised, controlled trial in England with mediation and moderation analyses.
freely available if proven to be beneficial. Lancet Psychiatr. 9 (5) https://doi.org/10.1016/S2215-0366(22)00060-8.
Firth, J., Torous, J., 2015. Smartphone apps for schizophrenia: a systematic review. JMIR
Acknowledgements mHealth and uHealth 3 (4), e102. https://doi.org/10.2196/mhealth.4930.
Garety, P.A., Craig, T.K.J., Iredale, C.H., Basit, N., Fornells-Ambrojo, M., Halkoree, R.,
Jolley, S., Landau, S., McCrone, P., Tunnard, C., Zala, D., Waller, H., 2018. Training
We express our thanks to members of the PPI Panel (the A team) for the frontline workforce to deliver evidence-based therapy to people with psychosis:
their contributions to the delivery of the trial. We thank the MUSE FEP challenges in the GOALS study. Psychiatr. Serv. 69 (1), 9–11. https://doi.org/
10.1176/appi.ps.201700268.
trial group which includes the trial supervisors Christina Thompson, and
Mark Dawson and the trial therapists Louise Maule, Janet Hood, Karen

295
R. Dudley et al. Journal of Psychiatric Research 174 (2024) 289–296

Garety, P., Ward, T., Emsley, R., Greenwood, K., Freeman, D., Fowler, D., Kuipers, E., Australian Schizophrenia Research Bank, McCarthy-Jones, S., Smailes, D., Corvin, A.,
Bebbington, P., Rus-Calafell, M., McGourty, A., Sacadura, C., Collett, N., James, K., Gill, M., Morris, D.W., Dinan, T.G., Murphy, K.C., Anthony O′Neill, F.,
Hardy, A., 2021. Effects of SlowMo, a blended digital therapy targeting reasoning, on Waddington, J.L., Donohoe, G., Dudley, R., 2017. Occurrence and co-occurrence of
paranoia among people with psychosis: a randomized clinical trial. JAMA Psychiatr. hallucinations by modality in schizophrenia-spectrum disorders. Psychiatr. Res. 252
78 (7), 714–725. https://doi.org/10.1001/jamapsychiatry.2021.0326. https://doi.org/10.1016/j.psychres.2017.01.102.
Goldsmith, L.P., Lewis, S.W., Dunn, G., Bentall, R.P., 2015. Psychological treatments for Michie, S., Yardley, L., West, R., Patrick, K., Greaves, F., 2017. Developing and
early psychosis can be beneficial or harmful, depending on the therapeutic alliance: evaluating digital interventions to promote behavior change in health and health
an instrumental variable analysis. Psychol. Med. 45 (11), 2365–2373. https://doi. care: recommendations resulting from an international workshop. J. Med. Internet
org/10.1017/S003329171500032X. Res. 19 (6), e232. https://doi.org/10.2196/jmir.7126.
Grant, S., Mayo-Wilson, E., Montgomery, P., Macdonald, G., Michie, S., Hopewell, S., Morrison, A.P., 2017. A manualised treatment protocol to guide delivery of evidence-
Moher, D., on behalf of the CONSORT-SPI Group, 2018. CONSORT-SPI 2018 based cognitive therapy for people with distressing psychosis: learning from clinical
Explanation and Elaboration: guidance for reporting social and psychological trials. Psychosis 9 (3), 271–281. https://doi.org/10.1080/17522439.2017.1295098.
intervention trials. Trials 19 (1), 406. https://doi.org/10.1186/s13063-018-2735-z. Morrison, A.P., Pyle, M., Gumley, A., Schwannauer, M., Turkington, D., MacLennan, G.,
Greenwood, K.E., Sweeney, A., Williams, S., et al., 2010. CHoice of Outcome in Cbt for Norrie, J., Hudson, J., Bowe, S.E., French, P., Byrne, R., Syrett, S., Dudley, R.,
psychosEs (CHOICE): the development of a new service user-led outcome measure of McLeod, H.J., Griffiths, H., Barnes, T.R.E., Davies, L., Kingdon, D., Aydinlar, S.,
CBT for psychosis. Schizophr. Bull. 36 (1), 126–135. et al., 2018. Cognitive behavioural therapy in clozapine-resistant schizophrenia
Griffiths, S.L., Bogatsu, T., Longhi, M., Butler, E., Alexander, B., Bandawar, M., (FOCUS): an assessor-blinded, randomised controlled trial. Lancet Psychiatr. 5 (8)
Everard, L., Jones, P.B., Fowler, D., Hodgekins, J., Amos, T., Freemantle, N., https://doi.org/10.1016/S2215-0366(18)30184-6.
McCrone, P., Singh, S.P., Birchwood, M., Upthegrove, R., 2023. Five-year illness Myers, E., Startup, H., Freeman, D., 2011. Cognitive behavioural treatment of insomnia
trajectories across racial groups in the UK following a first episode psychosis. Soc. in individuals with persistent persecutory delusions: a pilot trial. J. Behav. Ther. Exp.
Psychiatr. Psychiatr. Epidemiol. https://doi.org/10.1007/s00127-023-02428-w. Psychiatr. 42 (3), 330–336. https://doi.org/10.1016/j.jbtep.2011.02.004.
Haddock, G., McCarron, J., Tarrier, N., Faragher, E.B., 1999. Scales to measure National Institute of Clinical Excellence (NICE), 2002. Schizophrenia Core interventions
dimensions of hallucinations and delusions: the psychotic symptom rating scales in the treatment and management of Schizophrenia in Primary and Secondary Care.
(PSYRATS). Psychol. Med. 29 (4), 879–889. https://doi.org/10.1017/ NICE Clinical Guideline 1.
S0033291799008661. Neil, S.T., Kilbride, M., Pitt, L., et al., 2009. The questionnaire about the process of
Hayward, M., 2018. Evidence-based psychological approaches for auditory recovery (QPR): a measurement tool developed in collaboration with service users.
hallucinations. BJPsych Adv. 24 (3), 174–177. https://doi.org/10.1192/ Psychosis 1 (2), 145–155. https://doi.org/10.1080/17522430902913450, 2009.
bja.2017.11. NHS Long term plan, 2019. https://www.longtermplan.nhs.uk/publication/nhs-long-t
Hazell, C.M., Hayward, M., Cavanagh, K., Jones, A.M., Strauss, C., 2018. Guided self-help erm-plan/.
cognitive-behaviour Intervention for VoicEs (GiVE): results from a pilot randomised Oei, T.P.S., Green, A.L., 2008. The Satisfaction with Therapy and Therapist Scale–
controlled trial in a transdiagnostic sample. Schizophr. Res. 195, 441–447. https:// Revised (STTS-R) for group psychotherapy: psychometric properties and
doi.org/10.1016/j.schres.2017.10.004. confirmatory factor analysis. Prof. Psychol. Res. Pract. 39 (4), 435–442.
Hollis, C., Sampson, S., Simons, L., Davies, E.B., Churchill, R., Betton, V., Butler, D., Stevens, L.H., Turkington, D., Drage, L., Morrison, T., Muncer, S., Spencer, H.M.,
Chapman, K., Easton, K., Gronlund, T.A., Kabir, T., Rawsthorne, M., Rye, E., Dudley, R., 2019. Investigation of a traumatic psychosis subgroup: a cluster analysis
Tomlin, A., 2018. Identifying research priorities for digital technology in mental of an antipsychotic free cohort. Psychosis 11 (4). https://doi.org/10.1080/
health care: results of the James Lind alliance priority setting partnership. Lancet 17522439.2019.1628290.
Psychiatr. 5 (10), 845–854. https://doi.org/10.1016/S2215-0366(18)30296-7. Teare, M.D., Dimairo, M., Shephard, N., et al., 2014. Sample size requirements to
Horvath, AO.Greenberg LS., 1986. The development of the working alliance inventory. estimate key design parameters from external pilot randomised controlled trials: a
In: The Psychotherapeutic Process: A Research Handbook. Guilford Press, simulation study. Trials 15 (1), 264.
pp. 529–556. Turkington, D., Kingdon, D., Turner, T., Insight into Schizophrenia Research Group,
Killikelly, C., He, Z., Reeder, C., Wykes, T., 2017. Improving adherence to web-based and 2002. Effectiveness of a brief cognitive-behavioural therapy intervention in the
mobile technologies for people with psychosis: systematic review of new potential treatment of schizophrenia. Br. J. Psychiatr. 180, 523–527. https://doi.org/
predictors of adherence. JMIR M and uhealth 5 (7), e94. https://doi.org/10.2196/ 10.1192/bjp.180.6.523.
mhealth.7088. https://mhealth.jmir.org/2017/7/e94. Turkington, D., Spencer, H., Lebert, L., Dudley, R., 2017. Befriending: active placebo or
Lewis, S., Tarrier, N., Haddock, G., Bentall, R., Kinderman, P., Kingdon, D., Siddle, R., effective psychotherapy. Br. J. Psychiatr. 211, 5–6.
Drake, R., Everitt, J., Leadley, K., Benn, A., Grazebrook, K., Haley, C., Akhtar, S., Van Lieshout, R.J., Goldberg, J.O., 2007. Quantifying self-reports of auditory verbal
Davies, L., Palmer, S., Faragher, B., Dunn, G., 2002. Randomised controlled trial of hallucinations in person with psychosis. Can. J. Behav. Sci. 39, 73–77.
cognitive-behavioural therapy in early schizophrenia: acute-phase outcomes. Br. J. Ware, J.E., Sherbourne, C.D., 1992. The MOS 36-item short-form health survey (SF-36).
Psychiatry 181 (Suppl. 43). https://doi.org/10.1192/bjp.181.43.s91. I. Conceptual framework and item selection. Med. Care 30 (6), 473–483.
Longden, E., Corstens, D., Bowe, S., Pyle, M., Emsley, R., Peters, S., Branitsky, A., Waters, F., Blom, J.D., Jardri, R., Hugdahl, K., Sommer, I.E.C., 2018. Auditory
Chauhan, N., Dehmahdi, N., Jones, W., Holden, N., Larkin, A., Miners, A., hallucinations, not necessarily a hallmark of psychotic disorder, 4. In: Psychological
Murphy, E., Steele, A., Morrison, A.P., 2022. A psychological intervention for Medicine, vol. 48, pp. 529–536. https://doi.org/10.1017/S0033291717002203.
engaging dialogically with auditory hallucinations (Talking with Voices): a single- Cambridge University Press.
site, randomised controlled feasibility trial. Schizophr. Res. 250, 172–179. https:// Yiend, J., Lam, C.L.M., Schmidt, N., Crane, B., Heslin, M., Kabir, T., McGuire, P.,
doi.org/10.1016/j.schres.2022.11.007. Meek, C., Mouchlianitis, E., Peters, E., Stahl, D., Trotta, A., Shergill, S., 2022.
Lovibond, P.F., Lovibond, S.H., 1995. The structure of negative emotional states: Cognitive bias modification for paranoia (CBM-pa): a randomised controlled
comparison of the depression anxiety stress scales (DASS) with the beck depression feasibility study in patients with distressing paranoid beliefs. Psychol. Med. 1–13.
and anxiety inventories. Behav. Res. Ther. 33 (3), 335–343. https://doi.org/10.1017/s0033291722001520.

296