AMAnthony P. Morrison

Anthony P. Morrison

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Anthony P. Morrison has a total of 358 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Relative Effectiveness of Cognitive Behavior Therapy, Antipsychotics and the Combination for People with First Episode Psychosis: A Two-Study Pooled Analysis of Individual Participant Data

    by Anthony P Morrison, Melissa Pyle, Heather Law, Richard Emsley, Graeme MacLennan, Jemma Hudson

    Background: There is little head-to-head data comparing cognitive behavior therapy for psychosis (CBTp) and antipsychotic medication (APs). However, several recent trials have been conducted in first episode psychosis. We report a pre-specified individual participant data (IPD) pooled analysis utilizing data from two randomized controlled trials (RCTs) with similar designs to examine relative effectiveness. Study Design: The outcomes were psychiatric symptoms (Positive and Negative Syndrome Scale: PANSS) and recovery (Questionnaire about the Process of Recovery: QPR) at 6 months. One-stage and two-stage IPD meta-analyses were performed based on the intention-to-treat principle. Serious adverse events are also summarized. Study Results: Two RCTs were included in the pooled IPD analysis which provided 136 participants. For PANSS total at 6 months, the one-stage meta-analysis found no evidence of a difference between CBTp alone and APs alone (mean difference 2.58, 95% CI, -2.83 to 7.98; P-value 0.35) and CBTp alone compared with APs plus CBTp (MD -5.91; 95% CI, -12.14 to 0.31; P-value 0.063). For APs alone compared to CBTp plus APs there was evidence of a difference (MD -8.49; 95% CI, -14.65 to -2.33; P-value 0.007) favoring the combined treatment. For user-defined recovery (QPR), there was a difference favoring CBTp plus APs in comparison to both CBTp alone (P-value 0.029) and APs alone (P-value 0.026), but no difference between the monotherapies (P-value 0.91). The most common serious adverse events were psychiatric hospital admissions. Conclusions: Cognitive behavior therapy for psychosis and APs did not differ in their effects on symptoms or recovery, but there were suggestions that the combined treatment may be superior. A definitive RCT is warranted.

  • Open Access

    Treatment effect modifiers of cognitive behaviour therapy in people with psychosis: an individual participant data meta-analysis of RCTs

    by Filippo Varese, Maria Sudell, Anthony P Morrison, Eleanor Longden, Catrin Tudur Smith

    Background: Cognitive–behavioural therapy is a recommended intervention for the treatment of schizophrenia and related psychoses, but there is considerable uncertainty on whether its effectiveness is moderated by patient characteristics and/or intervention characteristics. Objective(s): To identify treatment effect modifiers of cognitive–behavioural therapy in people with schizophrenia spectrum diagnoses. Design: An individual participant data meta-analysis of randomised controlled trials comparing cognitive–behavioural therapy to treatment as usual or control active psychosocial control interventions. Setting: Community and inpatient settings. Participants: Individuals with schizophrenia spectrum diagnoses. Interventions: Cognitive–behavioural therapy, as defined by the criteria outlined in the National Institute for Health and Care Excellence guideline on treatment and management of schizophrenia in adults. Main outcome measures: Overall symptom change as measured by assessments of overall psychotic symptom severity (e.g. the Positive and Negative Syndrome Scales). Data sources: Corresponding authors of 110 trials identified from the database searches conducted as part of a related aggregate data meta-analysis in February 2018 (later updated in January 2019) were invited to share their trials’ individual participant data, and additional trial documentation, when available, pertaining to relevant individual participant data metadata, statistical analyses plans and characteristics of the cognitive–behavioural therapy interventions evaluated in their eligible trials. Review methods: Reports of retrieved and unretrieved trials were assessed using the Cochrane Risk of Bias tool. Data were cleaned and standardised to allow pooling and analysis. We conducted a series of two-stage individual participant data random-effect meta-analyses across four treatment comparisons: cognitive–behavioural therapy versus treatment as usual; cognitive–behavioural therapy versus other psychosocial interventions/active comparisons (active control psychosocial interventions); cognitive–behavioural therapy integrating additional elements from other therapies (‘cognitive–behavioural therapy+’) versus treatment as usual; and cognitive–behavioural therapy+ versus active control psychosocial interventions. Treatment by covariate interaction analyses were carried out to examine potential treatment effect modifiers, including participants’ demographic characteristics (age, gender, ethnicity), clinical characteristics (illness duration, phase of illness, duration of untreated psychosis, initial severity of psychotic symptoms and affective symptoms), and specific intervention characteristics (treatment duration, number of therapy sessions, level of therapists’ training/competence, use of manualised interventions, formulation-based interventions; individual vs. group interventions). Results: A total of 53 trials were retrieved. Of these, 27 trials ( n = 2870) were available for the planned cognitive–behavioural therapy versus treatment as usual primary outcome analyses, 11 trials ( n = 961) for the cognitive–behavioural therapy versus active control psychosocial interventions analyses, 14 trials ( n = 1985) for cognitive–behavioural therapy+ versus treatment as usual analyses, and 3 trials ( n = 28) for the cognitive–behavioural therapy+ versus active control psychosocial interventions analyses. There was no reliable evidence indicating that any of the covariates considered in this evidence synthesis significantly impacted the efficacy of cognitive–behavioural therapy in this client group. Limitations: Only 54% of the individual participant data requested were provided by data owners, and there is considerable heterogeneity in the features of the cognitive–behavioural therapy interventions included in this evidence synthesis. Conclusions: The effectiveness of cognitive–behavioural therapy for overall symptom change in this patient group is not significantly affected by the covariates examined in this individual participant data meta-analysis. Cognitive–behavioural therapy should continue to be offered equally to service users irrespective of their demographic or clinical characteristics. Future work: Other potential avenues to explore moderators of cognitive–behavioural therapy efficacy are suggested, including the fine-grained analysis of specific intervention components and the added value of interventionist-causal paradigms. Study registration: This study is registered as PROSPERO CRD42017060068. Funding: This award was funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme (NIHR award ref: 15/187/05) and is published in full in Health Technology Assessment ; Vol. 29, No. 53. See the NIHR Funding and Awards website for further award information.


  • Open Access

    Online Hearing Voices Peer Support Groups: Assessing Feasibility and Acceptability Within UK National Health Service Settings

    by Alison Branitsky, Samantha Bowe, Anthony P Morrison, Eleanor Longden, Sandra Bucci, Lee D Mulligan, Filippo Varese

    Background: User-led hearing voices groups (HVGs) have existed for the past 40 years, but little research has been conducted into if and how this approach can be implemented in statutory mental health settings, and whether they can be delivered effectively when held online. The aim of this study was to conduct a feasibility and acceptability trial of an online HVG for the UK National Health Service (NHS) users who hear voices, to inform a future larger-scale trial. Methods: A mixed-methods, nonrandomized feasibility study of an online HVG was conducted with 9 participants. Participants attended weekly online meetings for 6 months and completed measures of social connectedness, voice hearing, personal recovery, as well as semi-structured interviews, at baseline and postintervention (26-weeks). Primary outcomes were qualitative and quantitative assessments of feasibility and acceptability. Results: Thirty-eight participants were referred to the trial, 9 of whom were recruited (100% of the target sample). The trial had high retention (100%) and engagement (mean = 13.2 groups attended). Participants reported positive features of attending this digital intervention, with 85% reporting it was helpful to meet other voice hearers, that the group helped them make sense of voice hearing experiences, that they received positive messages about recovery, and that the group represented a form of support they could not get elsewhere. Discussion: The findings indicate it is feasible and acceptable to run an online HVG within an NHS setting. A larger trial is needed to further investigate the utility, efficacy, and cost-effectiveness of running online HVGs in the NHS. Trial Registration: ISRCTN11873550.


  • Open Access

    Study protocol for an adaptive, multi-arm, multi-stage (MAMS) randomised controlled trial of brief remotely delivered psychosocial interventions for people with serious mental health problems who have experienced a recent suicidal crisis: Remote Approaches to Psychosocial Intervention Delivery (RAPID)

    by Melissa Pyle, Lucy Loftus, Richard Emsley, Daniel Freeman, Steven Gillard, Andrew Gumley, Justyna Sierpatowska, Lisa Wood, Rory C. O’cOnnor, Paul Pfeiffer, et al.

    Background: People with serious mental health problems (SMHP) are more likely to be admitted to psychiatric hospital following contact with crisis services. Admissions can have significant personal costs, be traumatic and are the most expensive form of mental health care. There is an urgent need for treatments to reduce suicidal thoughts and behaviours and reduce avoidable psychiatric admissions. Methods: A multi-stage, multi-arm (MAMS) randomised controlled trial (RCT) with four arms conducted over two stages to determine the clinical and cost effectiveness of three psychosocial treatments, compared to treatment as usual (TAU), for people with SMHP who have had recent suicidal crisis. Primary outcome is any psychiatric hospital admissions over a 6-month period. We will assess the impact on suicidal thoughts and behaviour, hope, recovery, anxiety and depression. The remote treatments delivered over 3 months are structured peer support (PREVAIL); a safety planning approach (SAFETEL) delivered by assistant psychologists; and a CBT-based suicide prevention app accessed via a smartphone (BrighterSide). Recruitment is at five UK sites. Stage 1 includes an internal pilot with a priori progression criteria. In stage 1, the randomisation ratio was 1:1:1:2 in favour of TAU. This has been amended to 2:2:3 in favour of TAU following an unplanned change to remove the BrighterSide arm following the release of efficacy data from an independent RCT. Randomisation is via an independent remote web-based randomisation system using randomly permuted blocks, stratified by site. An interim analysis will be performed using data from the first 385 participants from PREVAIL, SAFETEL and TAU with outcome data at 6 months. If one arm is dropped for lack of benefit in stage 2, the allocation ratio of future participants will be 1:1. The expected total sample size is 1064 participants (1118 inclusive of BrighterSide participants). Discussion: There is a need for evidence-based interventions to reduce psychiatric admissions, via reduction of suicidality. Our focus on remote delivery of established brief psychosocial interventions, utilisation of different modalities of delivery that can provide sustainable and scalable solutions, which are also suitable for a pandemic or national crisis context, will significantly advance treatment options. Trial registration: ISRCTN33079589. Registered on June 20, 2022.


  • Open Access

    Group Cohesion and Necessary Adaptations in Online Hearing Voices Peer Support Groups: Qualitative Study With Group Facilitators

    by Alison Branitsky, Eleanor Longden, Sandra Bucci, Anthony P Morrison, Filippo Varese

    Background: Face-to-face hearing voices peer support groups (HVGs), a survivor-led initiative that enables individuals who hear voices to engage with the support of peers, have a long-standing history in community settings. HVGs are premised on the notion that forming authentic, mutual relationships enables the exploration of one’s voice hearing experiences and, in turn, reduces subjective distress. As such, group cohesion is assumed to be a central mechanism of change in HVGs. The rise of digital mental health support, coupled with the COVID-19 pandemic, has resulted in many HVGs adapting to online delivery. However, to date no studies have examined the implementation of these online groups and the adaptations necessary to foster cohesion. Objective: This study aims to understand the experience of group cohesion among HVG facilitators in online groups compared with face-to-face groups. Specifically, we examined the ways in which the medium through which groups run (online or face-to-face) impacts group cohesion and how facilitators adapted HVGs to foster group cohesion online. Methods: Semistructured qualitative interviews were conducted with 11 facilitators with varied experience of facilitating online and face-to-face HVGs. Data were analyzed using reflexive thematic analysis. Results: The findings are organized into 3 themes and associated subthemes: nonverbal challenges to cohesion (lack of differentiation, transitional space, inability to see the whole picture, and expressions of empathy); discursive challenges to cohesion (topic-based conversation and depth of disclosure); and necessary adaptations for online groups (fostering shared experience and using the unique context to demonstrate investment in others). Despite challenges in both the setting and content of online groups, facilitators felt that group cohesion was still possible to achieve online but that it had to be facilitated intentionally. Conclusions: This study is the first to specifically investigate group cohesion in online HVGs. Participants noted numerous challenges to group cohesion when adapting groups to run online, including the unnaturally linear narrative flow of dialogue in online settings; lack of transitional spaces, and associated small talk before and after the session; ease of disengagement online; inhibited sharing; and absence of shared physical presence online. Although these challenges were significant, facilitators nevertheless emphasized that the benefits provided by the accessibility of online groups outweighed these challenges. Necessary adaptations for cultivating group cohesion online are outlined and include capitalizing on moments of humor and spontaneity, using group activities, encouraging information sharing between participants using the chat and screen-sharing features, and using objects from participants’ environments to gain deeper insight into their subjective worlds.


  • Open Access

    Group Cohesion and Necessary Adaptations in Online Hearing Voices Peer Support Groups: Qualitative Study With Group Facilitators (Preprint)

    by Alison Branitsky, Eleanor Longden, Sandra Bucci, Anthony P Morrison, Filippo Varese

    BACKGROUND Face-to-face hearing voices peer support groups (HVGs), a survivor-led initiative that enables individuals who hear voices to engage with the support of peers, have a long-standing history in community settings. HVGs are premised on the notion that forming authentic, mutual relationships enables the exploration of one’s voice hearing experiences and, in turn, reduces subjective distress. As such, group cohesion is assumed to be a central mechanism of change in HVGs. The rise of digital mental health support, coupled with the COVID-19 pandemic, has resulted in many HVGs adapting to online delivery. However, to date no studies have examined the implementation of these online groups and the adaptations necessary to foster cohesion. OBJECTIVE This study aims to understand the experience of group cohesion among HVG facilitators in online groups compared with face-to-face groups. Specifically, we examined the ways in which the medium through which groups run (online or face-to-face) impacts group cohesion and how facilitators adapted HVGs to foster group cohesion online. METHODS Semistructured qualitative interviews were conducted with 11 facilitators with varied experience of facilitating online and face-to-face HVGs. Data were analyzed using reflexive thematic analysis. RESULTS The findings are organized into 3 themes and associated subthemes: nonverbal challenges to cohesion (lack of differentiation, transitional space, inability to see the whole picture, and expressions of empathy); discursive challenges to cohesion (topic-based conversation and depth of disclosure); and necessary adaptations for online groups (fostering shared experience and using the unique context to demonstrate investment in others). Despite challenges in both the setting and content of online groups, facilitators felt that group cohesion was still possible to achieve online but that it had to be facilitated intentionally. CONCLUSIONS This study is the first to specifically investigate group cohesion in online HVGs. Participants noted numerous challenges to group cohesion when adapting groups to run online, including the unnaturally linear narrative flow of dialogue in online settings; lack of transitional spaces, and associated small talk before and after the session; ease of disengagement online; inhibited sharing; and absence of shared physical presence online. Although these challenges were significant, facilitators nevertheless emphasized that the benefits provided by the accessibility of online groups outweighed these challenges. Necessary adaptations for cultivating group cohesion online are outlined and include capitalizing on moments of humor and spontaneity, using group activities, encouraging information sharing between participants using the chat and screen-sharing features, and using objects from participants’ environments to gain deeper insight into their subjective worlds.


  • Open Access

    Estimating the Economic Value of Automated Virtual Reality Cognitive Therapy for Treating Agoraphobic Avoidance in Patients With Psychosis: Findings From the gameChange Randomized Controlled Clinical Trial

    by James Altunkaya, Michael Craven, Sinéad Lambe, Ariane Beckley, Laina Rosebrock, Robert Dudley, Kate Chapman, Anthony Morrison, Eileen O'REgan, Jenna Grabey, et al.

    Background: An automated virtual reality cognitive therapy (gameChange) has demonstrated its effectiveness to treat agoraphobia in patients with psychosis, especially for high or severe anxious avoidance. Its economic value to the health care system is not yet established. Objective: In this study, we aimed to estimate the potential economic value of gameChange for the UK National Health Service (NHS) and establish the maximum cost-effective price per patient. Methods: Using data from a randomized controlled trial with 346 patients with psychosis (ISRCTN17308399), we estimated differences in health-related quality of life, health and social care costs, and wider societal costs for patients receiving virtual reality therapy in addition to treatment as usual compared with treatment as usual alone. The maximum cost-effective prices of gameChange were calculated based on UK cost-effectiveness thresholds. The sensitivity of the results to analytical assumptions was tested. Results: Patients allocated to gameChange reported higher quality-adjusted life years (0.008 QALYs, 95% CI –0.010 to 0.026) and lower NHS and social care costs (–£105, 95% CI –£1135 to £924) compared with treatment as usual (£1=US $1.28); however, these differences were not statistically significant. gameChange was estimated to be worth up to £341 per patient from an NHS and social care (NHS and personal social services) perspective or £1967 per patient from a wider societal perspective. In patients with high or severe anxious avoidance, maximum cost-effective prices rose to £877 and £3073 per patient from an NHS and personal social services perspective and societal perspective, respectively. Conclusions: gameChange is a promising, cost-effective intervention for the UK NHS and is particularly valuable for patients with high or severe anxious avoidance. This presents an opportunity to expand cost-effective psychological treatment coverage for a population with significant health needs. Trial Registration: ISRCTN Registry ISRCTN17308399; https://www.isrctn.com/ISRCTN17308399 International Registered Report Identifier (IRRID): RR2-10.1136/bmjopen-2019-031606


  • Open Access

    Protocol for individual participant data meta-analysis of randomised controlled trials of patients with psychosis to investigate treatment effect modifiers for CBT versus treatment as usual or other psychosocial interventions

    by Maria Sudell, Catrin Tudur-Smith, Xiaomeng Liao, Eleanor Longden, Graham Dunn, Tim Kendall, Richard Emsley, Anthony Morrison, Filippo Varese

    Introduction Aggregate data meta-analyses have shown heterogeneous treatment effects for cognitive–behavioural therapy (CBT) for patients with schizophrenia spectrum diagnoses. This heterogeneity could stem from specific intervention or patient characteristics that could influence the clinical effectiveness of CBT, termed treatment effect modifiers. This individual participant data meta-analysis will investigate a range of potential treatment effect modifiers of the efficacy of CBT. Methods and analysis We will perform a systematic review and meta-analysis of studies investigating CBT versus treatment as usual, or CBT versus other psychosocial interventions, for patients with schizophrenia spectrum diagnoses. The Cochrane Central Register of Controlled Trials (CENTRAL), PubMed, EMBASE and the online clinical trials registers of the US government, European Union, WHO and Current Controlled Trials will be searched. Two researchers will screen titles and abstracts identified by the search. Individual participant data will be requested for any eligible study, for the primary outcome (overall psychotic symptoms), secondary outcomes and treatment effect modifiers. Data will be checked and recoded according to an established statistical analysis plan. One-stage and two-stage random effects meta-analyses investigating potential treatment effect modifiers will be conducted. A list of potential treatment effect modifiers for CBT will be produced, motivating future research into particular modifiers. Ethics and dissemination This study does not require ethical approval as it is based on data from existing studies, although best ethical practice for secondary analysis of clinical data will be followed. The findings will be submitted for publication in peer-reviewed journals, and promoted to relevant stakeholders. PROSPERO registration number CRD42017060068.


  • A treatment protocol to guide the delivery of dialogical engagement with auditory hallucinations: Experience from the Talking With Voices pilot trial

    by Eleanor Longden, Dirk Corstens, Anthony P. Morrison, Amanda Larkin, Elizabeth Murphy, Natasha Holden, Ann Steele, Alison Branitsky, Samantha Bowe

    Purpose: To present a treatment protocol for delivering Talking With Voices, a novel intervention for people with psychosis that involves dialogical engagement with auditory hallucinations. Method: This paper presents a manualized approach to therapy employed in the Talking With Voices trial, a feasibility and acceptability randomized control trial of 50 adult participants. A rationale for following a treatment manual is provided, followed by the theoretical underpinnings of the intervention and its principles and values, including the main tenet that voices can often be understood as dissociated parts of the self which serve a protective function by indicating social–emotional vulnerabilities. The four therapy phases for improving the relationship between the voice‐hearer and their voices are outlined: (1) engagement and psychoeducation, (2) creating a formulation, (3) dialoguing with voices, and (4) consolidating outcomes, including key milestones at each phase. Implementation issues are discussed, as well as recommendations for best practice and future research. Results: The Talking With Voices treatment protocol indicates that it is feasible to manualize a dissociation‐based approach to support service users who are distressed by hearing voices. Conclusion: For some individuals, it is possible to engage in productive dialogue with even extremely hostile or distressing voices. Developing coping strategies, creating a formulation, and ultimately establishing a dialogue with voices has the potential to improve the relationship between voice(s) and voice‐hearer. Further research is now required to evaluate feasibility, acceptability, and efficacy. Practitioner points: It is feasible to integrate a dissociation model of voice‐hearing within a psychological intervention for people with psychosis. Combining psychosocial education, formulation and direct dialogue can be used to facilitate a more peaceful relationship between clients and their voices. Practitioners trained in other therapeutic modalities can draw on existing transferrable skills to dialogue with their clients’ voices. The input of those with lived experience of mental health difficulties has an important role in guiding treatment design and delivery.


  • Open Access

    Psychological intervention, antipsychotic medication or a combined treatment for adolescents with a first episode of psychosis: the MAPS feasibility three-arm RCT

    by Anthony P Morrison, Melissa Pyle, Rory Byrne, Matthew Broome, Daniel Freeman, Louise Johns, Anthony James, Nusrat Husain, Richard Whale, Graeme MacLennan, et al.

    Background: When psychosis emerges in young people there is a risk of poorer outcomes, and access to evidence-based treatments is paramount. The current evidence base is limited. Antipsychotic medications show only a small benefit over placebo, but young people experience more side effects than adults. There is sparse evidence for psychological intervention. Research is needed to determine the clinical effectiveness and cost-effectiveness of psychological intervention versus antipsychotic medication versus a combined treatment for adolescents with psychosis. Objectives: The objective of Managing Adolescent first-episode Psychosis: a feasibility Study (MAPS) was to determine the feasibility of conducting a definitive trial to answer the question of clinical effectiveness and cost-effectiveness of these three treatment options. Design: This was a prospective, randomised, open-blinded, evaluation feasibility trial with a single blind. Participants were allocated 1 : 1 : 1 to receive antipsychotic medication, psychological intervention or a combination of both. A thematic qualitative study explored the acceptability and feasibility of the trial. Setting: Early intervention in psychosis services and child and adolescent mental health services in Manchester, Oxford, Lancashire, Sussex, Birmingham, Norfolk and Suffolk, and Northumberland, Tyne and Wear. Participants: People aged 14–18 years experiencing a first episode of psychosis either with an International Classification of Diseases, Tenth Revision, schizophrenia spectrum diagnosis or meeting the entry criteria for early intervention in psychosis who had not received antipsychotic medication or psychological intervention within the last 3 months. Interventions: Psychological intervention involved up to 26 hours of cognitive–behavioural therapy and six family intervention sessions over 6 months, with up to four booster sessions. Antipsychotic medication was prescribed by the participant’s psychiatrist in line with usual practice. Combined treatment was a combination of psychological intervention and antipsychotic medication. Main outcome measures: The primary outcome was feasibility (recruitment, treatment adherence and retention). We used a three-stage progression criterion to determine feasibility. Secondary outcomes were psychosis symptoms, recovery, anxiety and depression, social and educational/occupational functioning, drug and alcohol use, health economics, adverse/metabolic side effects and adverse/serious adverse events. Results: We recruited 61 out of 90 (67.8%; amber zone) potential participants (psychological intervention, n = 18; antipsychotic medication, n = 22; combined treatment, n = 21). Retention to follow-up was 51 out of 61 participants (83.6%; green zone). In the psychological intervention arm and the combined treatment arm, 32 out of 39 (82.1%) participants received six or more sessions of cognitive–behavioural therapy (green zone). In the combined treatment arm and the antipsychotic medication arm, 28 out of 43 (65.1%) participants received antipsychotic medication for 6 consecutive weeks (amber zone). There were no serious adverse events related to the trial and one related adverse event. Overall, the number of completed secondary outcome measures, including health economics, was small. Limitations: Medication adherence was determined by clinician report, which can be biased. The response to secondary outcomes was low, including health economics. The small sample size obtained means that the study lacked statistical power and there will be considerable uncertainty regarding estimates of treatment effects. Conclusions: It is feasible to conduct a trial comparing psychological intervention with antipsychotic medication and a combination treatment in young people with psychosis with some adaptations to the design, including adaptations to collection of health economic data to determine cost-effectiveness. Future work: An adequately powered definitive trial is required to provide robust evidence. Trial registration: Current Controlled Trials ISRCTN80567433. Funding: This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 25, No. 4. See the NIHR Journals Library website for further project information.